Subaru WRX Performance Design Guide

Introduction
This guide focuses on the hardware selection, maintenance, and operation.
The fueling system
By the suicidal eggroll – original post - https://forums.nasioc.com/forums/showthread.php?t=2339842

Alright, this is going to be a big one. The return-style fueling system in our cars is straight
forward, but there are a lot of different pieces to cover. I'm going to start out with a description
of what AFR is, why it's important, and how the ECU controls the fueling system. Then I'll talk
about the fuel system from a mechanical perspective, and how the different components work
together.

So what is the air/fuel ratio, and how does the car's computer control it? Combustion in an
engine deals heavily with stoichiometry, the study of reactants and products in chemical
reactions (after all, fire is a very common chemical reaction). Contrary to popular belief, while
fuel is what gives the engine its energy/power, AIR is what's really important to making power.
After all, if fuel was the only thing that mattered, why wouldn't people just throw giant fuel
systems on tiny engines and make obscene amounts of power? The answer is stoichiometry.
While fuel contains the energy you need to extract in order for the engine to make power, air is
what allows you to extract that energy. The ratio of air to fuel is critical in this process. Too
much air and the fuel can't burn smoothly and completely. Too little air and the fuel simply can't
burn, because there isn't enough air to burn it all. We can inject as much fuel as we need, so the
focus in making power is trying to cram as much air as we can into the engine, so that we can
add the right amount of fuel and burn it.

The stoichiometric ratio is the amount of air required to properly burn a given amount of fuel.
The stoichiometric ratio for gasoline is around 14.7 parts air to one part fuel. This is where you
get a good balance of a stable, clean burn, and relatively complete combustion of the fuel. Now
while the stoichiometric ratio is the "ideal" AFR for complete combustion, due to the nature of a
realistic combustion chamber in realistic conditions, at stoich you aren't actually burning all of
the fuel or all of the air, due to the simple fact that some of the air can't "find" some of the fuel
when combustion occurs. Some engines are alright with an AFR (air/fuel ratio) slightly above
14.7, allowing you to more completely burn the fuel and improve mileage, but it all depends on
the design of the engine and how the air and fuel mix together. On the other end of the spectrum,
adding more fuel can let you burn all of the air completely, allowing you to make some extra
power, up to a point. The ideal power AFR for gasoline is somewhere around 12.5:1.

So all cars run an AFR of 14.7 at idle and cruise for gas mileage, and an AFR of 12.5 at wide
open throttle for power, right? Unfortunately it's not that simple. Most naturally aspirated cars do
exactly that, but forced induction cars (turbo, supercharged) are a completely different animal.
While 12.5 is still the "ideal" power AFR, most forced induction cars running regular pump gas
can't actually do it. An AFR of 12.5 is so energetic and unstable, that the heat created by
compressing the huge amount of air in the cylinder of a forced induction car running high boost
can actually cause the air/fuel mixture to ignite by itself, while the piston is still compressing,
before the spark plug fires. This is called detonation, pinging, or knocking, and is very harmful to
the engine. As a result, most forced induction vehicles have to add additional fuel beyond 12.5
for no other reason than to cool the air/fuel mixture and reduce its tendency to pre-ignite. On
regular gasoline, most turbo 4-cyl cars (Evo, WRX/STi, and so on) run an AFR around 10-11.
With higher octane race gas or additives (methanol injection, etc), it is possible to lean out the
AFR to closer to 12.5 without running into detonation, allowing more complete combustion and
providing significantly more power.

So how does the fueling system work, exactly? Air entering the engine is measured by the mass
airflow (MAF) sensor. There are many types of MAF sensors, but the one used on our car is also
referred to as a hot wire anemometer. This type of MAF works by electrically heating up a very
thin wire which is placed near the middle of the air stream (stock location is between the air filter
and the turbo). As air passes over the wire, the wire is cooled. The MAF sensor measures the
amount of power required to maintain the wire at a fixed temperature, and sends this information
to the ECU.

The ECU reads the output of the MAF sensor, which is nothing more than a voltage between 0-
5v. The ECU then references a huge lookup table which contains the relationship between MAF
voltage and air flow rate. For example, on a stock 2005 STi, a MAF reading of 3.95 volts is
equivalent to an incoming air flow rate of 166.14 grams/second. Once the ECU has an air flow
measurement, it then looks at the crank position sensor to get the current engine RPM, for our
example say it's 4000 RPM. The combination of the two give the ECU a value for the "load",
which is a measurement of the amount of air per engine revolution (MAF*60/RPM). For our
example, this would be 166.14*60/4000 = 2.49 grams/revolution. Using the load and RPM, the
ECU then references another lookup table which tells it what AFR it should run. Again for the
2005 STi, at 4000 RPM and a load of 2.49, the target AFR would be 11.33. At this point, the
ECU knows the engine is currently drawing in 2.49 grams of air per revolution, and the AFR
should be 11.33, so it knows it needs to inject 0.22 grams of fuel per revolution. It then adds any
necessary fueling adjustments (such as adding more fuel during warmup, etc).

At this point, the ECU references another lookup table, the injector scaling. This tells the ECU
how long the injectors need to be open in order to inject the necessary 0.22 grams of fuel per
engine revolution. The ECU must also look at another table, which tells the ECU how long it
needs to TELL the injectors to open for, in order for them to actually open the right amount of
time. This value is the injector latency or dead time, and is the amount of time between when the
injector is told to open, and when it actually opens, and is usually on the order of 0.5-1.5
milliseconds.
The ECU finally knows how long it needs to open the injectors, so it does, they spray the fuel,
the spark plug fires, and you get combustion. Now there are two modes in which the fueling
system in the ECU operates while you're driving: closed loop and open loop. Open loop fueling
stops here; the ECU injects the fuel, you get combustion, and then the process starts over for the
next engine revolution. Closed loop fueling takes things one step further. After combustion, the
exhaust exits the engine and passes through the exhaust system. On its way, it passes by the front
O2 sensor, which tells the ECU what the ACTUAL AFR ended up being. The ECU compares the
actual AFR against what the AFR was supposed to be, and modifies the fueling on the next
combustion cycle to get closer to target (this is one of those fueling adjustments mentioned
before). Closed loop is only used at idle and cruise when the target is 14.7, open loop is used at
high throttle and when the engine is first started, before the O2 sensor has come up to
temperature.

That’s it, that’s the fueling system from the ECU’s perspective. Now it’s time to look at the
fueling system from a mechanical perspective. Mechanically, the fuel system is very straight
forward. Following the diagram below, you can see the pump pressurizes the fuel, pumps it
through the filter, the first rail, the second rail, and finally to the FPR. The FPR is a valve that
keeps the fuel pressure 43.5 psi above the manifold pressure, and any extra fuel is then bled off
in the return line which dumps back into the fuel tank.
The fuel pump is a positive displacement pump, just like the oil pump. A positive displacement
pump is a constant-flow pump, unlike many other pumps that are constant-pressure, etc. An ideal
positive displacement pump always pumps out the same volumetric flow rate of fluid regardless
of the pressure required to do so. Whether it’s pumping into nothing, into a fuel pressure
regulator, or into a brick wall, it will always push the same amount of fluid. Now the fuel pump
isn’t an ideal pump, but that’s still the principle under which it works. A real fuel pump does
have some pressure dependence (with flow decreasing as pressure increases) but at most normal
pressures this dependence is relatively small.

Since the fuel pump is constant-flow, how is the pressure controlled in the fuel system? The
answer is the fuel pressure regulator. The FPR actually looks a lot like a wastegate internally.
There is a spring holding a valve closed. Manifold pressure pushes with the spring helping to
keep the valve closed, while pressure at the fuel inlet pushes against the spring. When the fuel
pressure exceeds the manifold pressure by 43.5 psi, the spring compresses and the valve opens,
allowing the fuel to exit the FPR and return to the fuel tank. In the real world (with a positive
displacement fuel pump), the FPR is almost always partially open, allowing some fuel to return
to the tank while keeping the rail pressure 43.5 psi above the manifold pressure.

If the fuel pump is too large and the FPR is too small (a common problem when people upgrade
their fuel pump too much without touching the rails or FPR), there is actually too much fuel for
the FPR to bypass it all, and the pressure starts rising uncontrollably, causing the car to run rich
(too much fuel for the amount of air in combustion). This is mostly a problem at low load
(idle/cruise) when the injectors aren’t pulling off very much of the fuel coming from the pump,
so almost all of it has to be bypassed and returned to the tank. This problem is helped in part by
fuel pump controllers, which vary the voltage or duty cycle of the power going to the pump to
reduce their flow rate at low load.

If the fuel pump is too small for the setup, the injectors might pull off all of the fuel coming from
the pump, so there’s nothing left for the fuel pressure regulator to regulate. In this case, the
pressure will start dropping uncontrollably, which is what causes cars to run lean (too little fuel
for the amount of air in combustion) at high load and high RPM when their fuel pump isn’t
adequate for the amount of power they’re making.

The FPR is quite possibly the most critical part in the entire fuel system. The entire principle on
which the ECU operates (opening the injectors a fixed amount of time to add the right amount of
fuel) relies on the difference between the fuel pressure behind the injector and the air pressure in
front of the injector being a constant. Without the fuel pressure regulator holding the fuel
pressure a fixed 43.5 psi above the manifold pressure, as manifold pressure rose, fuel flow would
drop, since the injectors would have to spray into a higher pressure environment. Similarly, as
the manifold pressure dropped below atmospheric, fuel flow would increase, since in addition to
the fuel pressure pushing fuel through the injector, the vacuum in the intake manifold would
actually be sucking fuel out of the injector as well. The reference line running between the
manifold and the FPR is quite possibly the most critical hose in the entire car. What’s more,
Subaru decided it would be a good idea to just slip the hose on, with no clamps. If you ever have
the engine bay disassembled to the point where you have easy access to the FPR and this hose,
do yourself a favor and put some zip ties or hose clamps on each end to hold it on there.

Injectors are very precise valves which are pulsed by the engine’s computer to let a fixed amount
of fuel enter the combustion chamber. They are rated in their flow rate (volume/time), which is
an indication of how much fuel they are able to flow when running at close to 100% duty cycle.
The larger the injector, the more fuel they can flow (supporting higher horsepower setups), but
the less control they have over exactly how much fuel is being injected. Large injectors also tend
to have worse spray patterns, so the fuel doesn’t atomize as well as it should, and combustion
suffers as a result. This is why larger injectors can lower fuel economy, because the worse spray
pattern and worse atomization means you need more fuel for proper combustion, lowering gas
mileage. Usually this isn’t a big concern until you get to very large setups though, 1200cc/min
and larger.

If you’re interested in fuel system calculations (hose diameter, pressure loss, etc.), check out the
excel worksheet I made. It’s a little too advanced for these “how stuff works” articles, so I’m
keeping it separate:
http://forums.nasioc.com/forums/show....php?t=2158062
FAQ
Q: Will larger injectors/fuel pump add power?
A: No. Remember what I said earlier, it’s AIR that’s important when it comes to adding power,
not fuel. The fuel system is a support system; it adds the fuel to support the amount of air you’re
bringing in to maintain the proper AFR. Larger injectors/fuel pump do nothing more than
increase the CAPACITY of the fuel system, allowing you to run larger turbo setups without
running the fuel system dry. The only time that upgrading the injectors or fuel pump will add
power is when you’ve already exceeded their limits, and have had to de-tune the engine to work
correctly on the restricted fuel system capacity. In that case, upgrading the fuel system will let
you tune it up to take full advantage of your turbo system. If you ever run into this situation, then
you need to take two steps back and re-think your researching process, because upgrading to a
turbo setup that your fuel system can’t support is step 1 in the unwritten book of how to blow up
your engine. If you don’t know how much fuel you need for a given setup, use the calculator
linked above. If you don’t know what any of it means, then you should be using a professional to
help you design your setup.

Q: Is the O2 sensor narrow or wideband?

A: It is a wideband. If you disagree, then just stop talking and listen. There are numerous tests,
diagrams, plots, measurements not just suggesting this, but PROVING this. The problem with
the O2 sensor is not its accuracy away from stoich, the problem is its installation location. When
the boost rises, the exhaust pressure rises, and the exhaust pressure throws off the readings of the
O2 sensor. For this reason, even though the O2 sensor reads down to 11.0, in the factory location
its measurements can not be trusted once you get into boost. Many people have relocated the O2
sensor after the turbo where the pressure is more or less normal, and have had great success
when compared to aftermarket widebands installed in the same location. If you haven’t done this
though, don’t pay attention to the factory O2 once the boost starts rising.

Q: What is the closed loop/open loop delay?

A: As I said before, closed loop is when the ECU uses feedback from the O2 sensor to maintain
target AFR, while open loop is just “inject it and forget it”. Since the O2 sensor is only accurate
at idle/cruise, that’s when the ECU is in closed loop with a target of 14.7. As you get on the
throttle, at some point the ECU has to make the switch from closed loop to open loop and drop
the target AFR. Now ideally this switch would occur when the RPM and/or load increased to the
point where the target AFR started dropping below 14.7, and this is exactly when most tuned
cars will make the switch. The problem is the factory decided to implement a delay in this switch
to improve emissions and fuel economy (they didn’t want the car constantly switching to open
loop and dropping the AFR every time you blipped the throttle on the freeway). This is called the
closed loop/open loop delay (CL/OL delay). When you get on the throttle and one of many
switches is triggered (high RPM, high load, high injector pulse width, high throttle), the ECU
starts a counter. When this counter exceeds the CL/OL delay, the ECU finally makes the switch
to open loop and drops the target AFR from 14.7 to whatever it is supposed to be given your
current RPM/load.

Unfortunately, on late model cars, this counter threshold is very, very high, and it’s entirely
possible for the car to be at peak boost and STILL be stuck in closed loop running 14.7. This
causes significant detonation and pulled timing, and will eventually damage the engine. The 02-
03 WRX had either no delay, or a delay that was so small that it didn’t matter. Over time Subaru
has increased the delay due to increasing emissions restrictions. Even through the 06 STi it
wasn’t terrible, but in 07 the counter threshold was jacked way up, and has since been believed
to be the primary cause of engine failures at stock to stage 2 power levels (blaming the failures
on tuned vehicles to damage that had already been done before the tune was implemented). If
you’ve ever noticed a hesitation on your 07+ WRX/STi at high throttle between 3-4k RPM,
that’s probably the ECU pulling timing to combat detonation caused by the CL/OL delay.

Q: Is driving without a front O2 sensor bad?

A: No. When the O2 sensor is disconnected, the car is immediately switched to open loop.
Unless there’s a major problem with the fueling system (untuned intake, untuned injectors,
clogged fuel system, a vacuum leak, etc), the car will continue to run just like normal, maybe
even better than normal (see above about the CL/OL delay). The only problem with this is that
the ECU will not correct for global fueling problems, such as a change in stoich from ethanol-
infused gasoline, etc. If you ever drive without your front O2 sensor operating, it would be very
wise to install an aftermarket wideband sensor so you can monitor the AFR yourself to check for
any fueling problems. If your O2 sensor ever fails and starts reporting incorrect AFR values to
the ECU, causing bad gas mileage or stalling, it is MUCH better to unplug the O2 sensor entirely
and let the car run in open loop than to allow the car to continue running under those conditions.
Now keep in mind that this ONLY applies to the front O2 sensor. The rear O2 sensor is used to
check for cat efficiency, and on the 16-bit ECU (02-05 WRX) it can be unplugged without a
problem (other than the resulting CEL which can be disabled), but on the 32-bit ECU (everything
else), unplugging it will drop the closed loop AFR target from 14.7 to closer to 13.5, hurting gas
mileage.
Q: Ok, so once and for all, what mods will, and what mods won’t affect the AFR on our
cars?
A: Good question me! Here’s a handy dandy list:
Stock sized intake: most likely
Big MAF intake: absolutely
Turbo inlet: it will have a small effect, it has been debated whether or not it’s even worth
worrying about
Turbo: same as turbo inlet
Intercooler: no
Intake Manifold: it will affect the per-cylinder fueling compensation, and maybe tip-in
enrichment
Cams: might affect tip-in enrichment
Porting: same as cams
Valves: same as cams
Header: no
Uppipe: no
Exhaust: no
Fuel pump: no, not unless the previous pump was dropping pressure, AND the car was
previously tuned taking that into account, or the pump is overrunning the fuel pressure regulator
and causing an uncontrollable rise in fuel pressure
Fuel rails: it might affect the per-cylinder fueling compensation
Fuel injectors: definitely, even if the flow rate is the same, chances are the latency is not
Fuel pressure regulator: not unless the fuel pressure is being changed, or the previous regulator
was being overrun AND the car was previously tuned taking that into account

Q: Why does a stock sized intake require tuning?

A: Remember, the MAF does not measure the air flow rate, it is actually measuring the velocity
of the air that crosses the hot wire. The MAF scaling table then extrapolates that velocity to get
the total air flow rate, taking into account the diameter of the tube around the MAF and the air
flow pattern of the air around the MAF. While a stock sized intake is the same diameter, the filter
location, bend locations, and bend radii will be different than stock. These all affect the air flow
pattern across the hot wire, which will affect the MAF scaling.
Q: When it comes to the fuel system, more capacity is always better, right? It gives me
more room to grow in the future.
A: Not necessarily. The fuel system is one of those things where you want to run what you
NEED, and no more. As far as injectors go, as long as you’re under about 1000cc it doesn’t
really matter, you might as well get as big injectors as you think you might need in the future.
Above 1000cc you should take a step back and look at your setup to see if you really need that
much fuel, at that point you’re starting to get into negative side effects.

This oversizing applies to pumps too people! Everybody is so gung-ho about getting the biggest
baddest drop in fuel pump they can find, because more capacity is always better, right? Wrong.
Look at the fuel system diagram above again. What do you see? In addition to many other things,
you see a series rail setup. The fuel comes from the pump, passes through one rail, then through
the other rail, then to the regulator. Our fuel lines are tiny guys, ¼” inside diameter. That causes
a lot of pressure drop, and the more fuel you’re flowing, the worse it is. The pressure drop
between the pump and the first injector doesn’t really matter, more pressure drop just means the
pump has to operate at a slightly higher pressure. As covered before, since the pump is positive
displacement, this pressure increase doesn’t have much of an effect on the flow rate. It hurts it a
little bit, but not much.

The bad part is there is about 3-4 feet of fuel line between the first injector and the last injector.
The more fuel you’re trying to force through this line with your gigantic pump, the larger the
pressure drop in this section of line, and the larger the pressure difference between the first and
last injector will be. This will cause the first and second cylinders to receive fuel to run rich, and
the third and fourth cylinders to receive fuel to run lean. If you want to see how bad it is, use the
4th tab in the excel spreadsheet linked above to find out. All of the equations are there, just plug
in your numbers and see how it looks. In my opinion, a DW 300 LPH is about the largest you
should go on the stock rail setup. Above that, and the flow will start to cause some problems with
the pressure imbalance in the rails. Just to be clear, if you NEED the flow, then by all means go
for it, but if you don’t, a huge oversized pump will just make things worse. If you NEED more
than a DW 300 pump, you should really start to look into either converting your rails to parallel,
or going with aftermarket rails and at a minimum a nice, fat line connecting them to keep the
pressure imbalance low.
Q: Is a series or parallel rail setup better?
A: I’m not touching this one with a 10 foot pole, hehe. Both have their advantages and
disadvantages, and you’ll find numerous arguments on both sides of the fence. A parallel setup
certainly has a smaller pressure difference between the two banks, that can’t be argued, but a
parallel setup can introduce other problems such as vapor lock. Speak to your tuner about this
before you decide on one way or the other. For what it’s worth, when I built my fuel system I
went parallel. Cruise and full throttle were fine, but I ran into problems at idle. I converted to
series, and it literally ran exactly the same. Turns out my problems were being caused by the
injectors, but I still noticed absolutely no difference between the series and parallel setups in
regards to drivability, AFR, etc. After fixing the problem, I just stayed with series, no particular
reason.

Turbo/Engine/Fuel Calculator
I posted this on the RR forums as well, but a lot of people here don't frequent RR.

This is a simple calculator that lets you estimate certain things about your car (IE: air flow rate,
target boost given a HP goal, HP given a target boost, etc). It makes no assumptions, everything
is input. There are a couple of tools to help you figure out certain parameters based on log data if
you don't have a good grasp on them yet.

A similar calculator was posted last year in the Newbies forum, unfortunately it had some bugs
in the calculation, and the developer abandoned the project almost immediately. The spreadsheet
was so locked down that you couldn't even see or modify the equations to fix the bugs either.
This is laid out similar to his, but I've developed it from scratch and left everything open.

Sheet 1 lets you estimate your flywheel and wheel horsepower (also your total fuel flow rate and
required injector size) given several inputs, namely AFR, BSFC, VE at peak power, RPM at
peak power, compressor and intercooler efficiencies, and atmospheric conditions.

Sheet 2 is the reverse, it lets you estimate the required manifold pressure (also your fuel flow and
injector size) given a target flywheel horsepower, AFR, BSFC, VE at peak power, RPM at peak
power, compressor and intercooler efficiencies, and atmospheric conditions.

It also outputs air flow rate in lb/min, CFM, and g/s as well as pressure ratio, so you can
reference the compressor map for your turbo to get an accurate value for the compressor
efficiency.
Sheet 3 has a couple of calibration tools, returning VE at peak power given efficiencies,
atmospheric conditions, RPM, manifold pressure, and air flow rate. The other tool returns BSFC
given calculated horsepower (from virtual dyno software or a real dyno), as well as drivetrain
loss, AFR, and measured air flow rate.

There are multiple intended uses for this spreadsheet. The ones that come to mind are:

1 - determine turbo/boost requirements for a given HP goal

2 - determine fuel flow requirements for a given boost or HP goal
3 - view changes in HP as a result of changes in your setup (changing intercooler
efficiency, changing turbo inlet pressure drop, changing atmospheric pressure or
temperature, etc)

Please offer any suggestions for improvement or critiques. There shouldn't be any bugs, but if
you find one please let me know so I can fix it.

http://www.thesuicidaleggroll.com/ho..._calc_v3.0.xls
Fuel Pump FAQ
What are the OEM fuel pump sizes?

2002-2005 WRX fuel pump 130 lph

2006-2007 WRX fuel pump 145 lph
2008+ WRX fuel pump 155 lph
2004-2007 STI fuel pump 145 lph
2008+ STI fuel pump 175 lph

What is the best fuel pump? The most commonly used fuel pump is the Walbro 342 fuel pump.
It is a drop in unit that supports most aftermarket turbo fueling requirements. Other fuel pumps
that are commonly used are:

a. Walbro 341 fuel pump. Functionally identical to the 342. Requires slight modification
for use.
b. DeatschWerks
d. Aeromotive fuel pump. Generally for high HP applications only though.

Can I use XXX brand fuel pump? Most likely yes, but you should refer to the manufacturer's
website for sizes and compatibility as the above listed brands are the commonly used ones today
in the Subaru community.

What about the 08+ WRX/STI? There have been changes to the fuel pump on these models
over to the Legacy style pump, so they are no longer "plug and play" swaps like previous
models. There is a DIY option to do this as well and it can be found via this link.

Is there anything to look out for with fuel pumps? When using aftermarket fuel pumps, many
re-use the stock WRX fuel pump filter. This filter contains a hard plastic skeleton inside to
prevent the pump from ingesting the soft "bag filter". Aftermarket pumps do not have the hard
plastic part, and many users enjoy the additional safety that the OEM filter provides over the
aftermarket fuel pump's filter.

Do I need to tune for a fuel pump? No.

Can I run an aftermarket fuel pump by itself without new injectors/turbo/management?

During the turbo upgrade path, it may be easier to purchase and install a fuel pump ahead of
time. It is cheap, easy to install, and removes an eventuality from the upgrade chain ahead of
time. You can run an aftermarket fuel pump with no other modifications. Adding a fuel pump
will not give you more fuel (i.e. richen your car), it simply allows for more fuel to be delivered if
you run higher boost or larger injectors. The stock RRFPR keeps the pressure at the stock 43 psi
+ manifold pressure no matter how much fuel may be available from the pump.
How hard is it to install a fuel pump? Allow around one hour for install time. Professional
installation, depending on your area, is around $75. This is one vehicle modification that is very
simple and can be successfully accomplished by even the greenest shade tree mechanic.

How do I install a fuel pump? Refer to the fuel pump manufacturer's instructions. For fuel
pumps without instructions, below is are some links to some of the better known fuel pump
installation instructions:

scoobymods.com instructions
scoobymods.com instructions
scoobymods.com instructions
TurboXS's instructions
Quantum Racing's instructions
Fuel pump install for 08+

Injector FAQ

What are the best fuel injectors? Fuel injectors should be purchased based on the best possible
match between the flow rating and needs of your aftermarket turbo. This should be a
determination by your tuner to provide the best possible match. It is better to opt on the side of
too large if there is any possibility of maxing out the injectors with your upgrade plans. Bigger is
not always better with injectors so it is always best to purchase injectors best suited for your
application and upgrade as need as opposed to starting off too large. This online calculator may
assist you in the decision process. This post by hotrod will also help. As well, ask your tuner for
his recommendation on brands. It will be cheaper to buy the brand he is familiar with rather than
wasting dyno time + tuner fees while he fiddles around with your rogue brand/size injectors!
ALWAYS take your tuner's advice on these matters no matter what the internet tells you!

What type of injectors do I need? Aside from matching the injector flow for your turbo/tuning
needs, you need to get the correct type for your vehicle:

02+ WRX uses high impedance top feed injectors

04-06 STI uses high impedance side feed injectors (also called bottom feed)
07+ STI uses high impedance top feed injectors

What is the difference between low and high impedance? Low impedance injectors need to be
electrically adapted to work properly, and usually come with the required electronic components
for this.
Do I need to consult with my tuner about injectors? Yes. Items in your car that can be tuner
specific: manual boost controllers, electronic boost controllers, upgraded wastegates, restrictor
pills, wastegate helper springs, external wastegates (type/size/brand), and injector
types/brands/sizes. ALWAYS take your tuner's advice on these matters no matter what the
internet tells you!

Can I run a aftermarket injectors without tuning? No. You MUST have your car tuned
during the installation process. Most tuners can send you a "limp map" based on your mods to
tide you over during the break-in period or to get you to their shop for a tune days or weeks later.
This map will be safe, but you should drive like your car is a 1990 Ford Escort until your final
tune.

The map notes from XXX tuning company say pinks, but 06 injectors flow the same so
can't I use them? No. Don't think you can out-think a tuner. If you have a question, ASK
FIRST. As above, don't trust the internet, trust your tuner.

Are there any cautions with injectors? UTEC (older engine management system) users should
be aware that they should not use low impedance injectors.

What are IDCs? The injector duty cycle is the proportion of time that the injector is open,
squirting fuel. When choosing injectors, many try to choose an injector size to place IDCs in the
85-90% range. This assures proper fueling with a little bit of headroom.

Is there a list of top feed injector sizes for the 2002+ WRX and 2007+ STI?

OEM 2002-2005 WRX 420 cc

JDM STi "pinks" 550 cc
OEM 2006+ WRX 560 cc (also known as "blues")
OEM 2007+ STI 560 cc
DeatschWerks 565 cc
DeatschWerks 650 cc
RC Engineering 650 cc
Fuel Injector Clinic 650 cc
Injector Dynamics 725 cc
DeatschWerks 750 cc
RC Engineering 750 cc
Fuel Injector Clinic 775 cc
Modified Stock WRX 850 cc
DeatschWerks 850 cc
Injector Dynamics 850 cc
Fuel Injector Clinic 900 cc
DeatschWerks 1000 cc
Injector Dynamics 1000 cc
Fuel Injector Clinic 1050 cc
Fuel Injector Clinic 1100 cc
DeatschWerks 1300 cc
Injector Dynamics 1300 cc
Injector Dynamics 2000 cc
Fuel Injector Clinic 2150 cc
DeatschWerks 2200 cc

Is there a list of side feed injector sizes for the 2004-2006 STI?

OEM 2004-2006 STi 535 cc

RC Engineering 650 cc
Fuel Injector Clinic 650 cc
Injector Dynamics 725 cc
DeatschWerks 740 cc
RC Engineering 750 cc
Fuel Injector Clinic 775 cc
Modified Stock Injectors 817 cc
DeatschWerks 850 cc
Injector Dynamics 850 cc
Fuel Injector Clinic 900 cc
DeatschWerks 1000 cc
Injector Dynamics 1000 cc
Fuel Injector Clinic 1100 cc
Injector Dynamics 1300 cc
Injector Dynamics 2000 cc
Fuel Injector Clinic 2150 cc

Can I use XXX brand injectors? Most likely yes, but you should refer to the manufacturer's
website for sizes and compatibility as the above listed brands are the commonly used ones today
in the Subaru community.

Who manufactures injectors?

DeatschWerks (also known as DW Injectors)

Fuel Injector Clinic (also known as FIC Injectors)
Injector Dynamics (also known as ID Injectors)
RC Engineering
Can I modify stock 02-05 WRX injectors for increased flow? Yes, though this route requires
additional research, thought, time, and care than purchasing larger injectors.

600 cc injectors
740 cc injectors (Though this is the thread title and commonly known name, these modified
injectors actually flow 850 cc when tested at the normal static fuel pressure of 43psi)

Can I modify stock 06+ WRX injectors or STi pinks for increased flow? Yes, but that would
be a dumb move. These babies go for like $300 used. Sell them, buy 02-05 WRX injectors for
like $150 and have them modded.

Are there any other mods to increase flow? Yes, you can modify your stock FPR when using
an aftermarket fuel pump to increase the cc rate of your existing injectors. View this thread or
this thread for details.

How hard is it to install fuel injectors? Allow around two hours for install time. Professional
installation, depending on your area, is around $150. This vehicle modification more difficult
and time consuming than most "bolt on" installations.

How do I install fuel injectors? Refer to the fuel injector manufacturer's instructions. For fuel
injectors without instructions, below is a link to one of the better known fuel injector installation
instructions:

TurboXS's instructions
Useful NASIOC thread
2011 WRX injector install instructions
Fuel Rail FAQ
What are fuel rails? Aftermarket fuel rails are designed to increase the internal diameter of the
stock fuel lines as well as convert the OEM fuel rail set up from series to parallel. The theory
behind these two changes is to increase fuel flow and equalize fuel distribution.

When are fuel rails needed? This is a very hard question without a clear answer as it varies
from tuner to tuner, but GENERALLY speaking, once you are on the road to over 400WHP, fuel
rails become a viable option, but consult with your tuner for the best answer. Additionally, if you
are a show car type person, you may find that fuel rails improve the look under the hood as a
reason for this modification, which is fine, but do know that they require tuning to operate.

Are there any fuel rail myths? Yes. Many years ago, it was theorized by some that the stock
serial layout fuel lines caused a lean cylinder condition which carried away the #3 or #4 piston
(depending on what story you believe) and that fuel rails with their parallel layout would cure
this. After time though, this theory has been generally disregarded save for a few curmudgeons.

Who manufactures fuel rails?

Aeromotive
Crawford Performance
Fuel Injector Clinic
Perrin Performance
Rocket Rally Rails
TurboXS
Ultimate Racing
Boomba Racing

Can I use XXX brand fuel rails? Most likely yes, but you should refer to the manufacturer's
website for sizes and compatibility as the above listed brands are the commonly used ones today
in the Subaru community.

Is there a "fuel rail mod"? Yes. Many have modified their stock fuel rails into parallel fuel
lines and/or parallel with bigger lines.

scoobymods.com instructions
supporting/supplemental NASIOC thread
supporting/supplemental NASIOC thread
How hard is it to install fuel rails? Allow around six hours for install time. Professional
installation, depending on your area, is around $400. This vehicle modification is more difficult
and time consuming than most "bolt on" installations.

Do I need to tune for a fuel rails? Yes, due to the change in fuel rail diameter they will flow
more to your injectors than the stock rails.

How do I install fuel rails? Refer to the fuel rail manufacturer's instructions. For fuel rails
without instructions, below are some links to installation instructions:

Perrin's instructions
Perrin's instructions
APS's instructions
Vishnu's instructions
Mike Egan's instructions
Jorge's instructions

Editors Note

This post was created because I wasn't able to find a good FAQ on these parts. I came up with
the text based on LOTS of searching here. Upon reading this you should have an idea of what
type of fuel supporting modifications best suit your needs. The types and manufacturers are up to
you.

If you find an error in this FAQ, please PM me with factual details and I will update this post.
Responses such as, "I have XXX's injectors and they are great!" or "XXX's fuel rails cracked
after 1 month" are not appreciated here, that is what the Car Parts Review Forum is for.

Last edited by Unabomber; 05-15-2014 at 10:17 AM.

Exhaust FAQ
Original thread by Unabomber

The primary purpose of an aftermarket exhaust is to remove or replace the stock catalytic
converter(s) with a better flowing unit. It also increases the exhaust diameter for better flow. The
term "aftermarket exhaust" can be broken down into the three main types:

1. Turbo back exhaust (TBE) covers items 1-4 as described below.

2. Cat back exhaust (CBE) covers items 3&4 as described below.
3. Axle back exhaust (ABE) covers item 4 as described below.

The stock WRX Subaru exhaust consists of:

1. Downpipe (with cat) | 2. Midpipe (with cat) | 3. Rear Exhaust Pipe | 4. Axle Back Muffler

Keep in mind that the above information is catered to the 2002-2005 WRX and is likewise
applicable to all years of the STi, though there are slight variations. For example, the STi (all
years) has a catless uppipe and the 2006+ WRX has a one piece downpipe/midpipe.

HP gain is dependent on the type of exhaust chosen. TBE gains are 20-30HP. CBE gains are 5-
15HP. ABE gains are 5-10HP. These figures are highly debated as different manufacturers use
different dynos with different cars with different levels of mods. It also varies because exhausts
use different amounts of catalytic converters or are catless.
What about overall fitment between the WRX and STi or different model years? All OEM
or aftermarket exhausts designed for 02-07 WRX/STi will fit either the 02+ WRX sedan/wagon
or 04+ STi. Occasionally, you will have a slight fitment issue with some models (moreso if your
car has the optional rear differential cover) that usually get be fixed via hanger adjustment,
longer exhaust hangers etc. 06+ models have fitment issues with some exhausts (mainly the
muffler portion) due to the new rear diffuser.

I have an 06+ WRX, is the downpipe the same? No. The 2006+ WRX has a unique exhaust in
that it's downpipe is both the downpipe and midpipe sections of the "older" exhaust. This means
you must use a "long" downpipe to bolt up to the rest of your exhaust system. A full TBE will fit
fine, but when replacing just the downpipe, you must use a long downpipe. Check with your
Vendor for other possible fitment issues prior to ordering.

I have an 08/09 WRX, is the downpipe the same? No. The 08/09 WRX has a unique exhaust
in that it's downpipe is the same, fitment wise, as the Legacy GT. So if you have one of these
models, you must ensure you state your model/year to your vendor or specifically request the
"Legacy GT downpipe" for your car in order to assure proper fitment.

I have an 08/09 WRX/STI, is the exhaust the same as the 02-07 models? No. The 08/09
WRX has a unique exhaust in that it's axle back portion is mounted differently. So if you have
one of these models, you must ensure you state your model/year to your vendor or specifically
request the "08/09 exhaust" for your car in order to assure proper fitment. Yes, the picture above
isn't a perfect representation of your car but it gives you an idea of where to pieces are located.

Can I use an OEM STi exhaust on my WRX? Yes, but do not use it thinking you now have an
upgraded exhaust. While the piping is slightly larger in diameter, the catalytic converter is larger
and more restrictive so any gains are negated. About the only reason to swap out components is
to go to the STi axleback muffler which is slightly louder than the OEM WRX unit.

How do I stop my car from sounding like a "Honda Killer Bee Exhaust"? Subaru boxer
configured engines, even equipped with a "Honda Killer Bee Exhaust" will NOT sound like a
buzzy, annoying exhaust due to the motor's natural sound. The only way to become buzzy is to
switch to an equal length header. The OEM exhaust manifold and aftermarket unequal length
manifolds (headers) retain the exhaust pulses at different intervals which create the unique boxer
sound. This is why your stock exhaust sounds similar to a VW Bug or Porsche engine.

Which manufacturer is best? This topic is highly debated. There have been no reported
consistent "bad" exhausts on the market. Obviously, there may have been bad exhausts sold, but
not enough to report as "bad" overall.
What exhaust metal material is best? Exhausts are made from mild steel, stainless steel (304 &
321), and titanium. There is no irrefutable evidence that one material is better than the other.
Obviously, corrosion levels are higher with mild steel (coated or otherwise). Article on exhaust
materials.

Which exhaust diameter is best? Exhausts have 3 diameters:

1. 2.5": Good for 300-350 HP

2. 3.0": Good for 500-600 HP
3. 3.5": Good for 600HP +

There is no irrefutable evidence that one diameter is better than the other. Purchasing a larger
exhaust than needed for your level of modification has no disadvantages other than cost and
possible less ground clearance issues.

What other differences are there with exhausts? The other differences are as follows:

1. High flow cat(s) or catless.

2. Exhaust pipe length. Some manufacturers have proprietary exhaust component lengths. This
means you have to use all of their exhaust piping or some of it with a custom piece. As well,
JDM specification exhausts are different length and require all JDM specification piping or
some of it with a custom piece. Additionally, some manufacturer's components may take the
place of two OEM components. This generally occurs with the downpipe section of a TBE
where the "downpipe" is actually a downpipe+midpipe and these are commonly referred to
as "long downpipes".
3. Flanges/mating surfaces. Exhaust components are either equipped with a flat flange or have a
donut gasket designed to flex during movement. Flat goes to flat and donut goes to donut,
ensure you double check prior to purchase which you have and what your new exhaust has.
Some companies do make adapters of various kinds to mate say diameter A to diameter B
piping and/or flat flange to donut gasket.

Which exhaust provides the least back pressure? No one is really sure, but theory suggests
that larger piping is better than smaller, baffleless mufflers are better than baffled mufflers, and
no cats/freer flowing cats are better than OEM/poorer flowing cats. This thread provides limited
testing that bears this theory out. This link shows a graph from that thread.

Which exhaust has the best sound? This subject is up to the individual. It is highly subjective
and can lead to trouble with local noise abatement ordinances. That great sounding exhaust may
be the cause of some expensive and/or time consuming noise abatement tickets. Arguably, the
three most famous/popular "best" sounding exhausts are the ScoobySport, Stromung, and
Prodrive. A smart move is to find someone in your local area with the exhaust component(s) you
are interested in to hear it in person. Seek out locals via the NASIOC Chapters or NASIOC
Regional Areas.
Which exhaust is the quietest? This question can be answered by listing hard data. By using
hard data, you remove human judgement and don't get stuck with a recommend "quiet exhaust"
from someone that in your opinion is now a loud and expensive nuisance. Keep in mind that
these quiet exhausts only protect you from the chance encounter in areas with strict noise
abatement laws. At wide open throttle (WOT) any aftermarket exhaust is considerably louder
than OEM and even the quiet ones will not protect you from folly.

Full dB testing data can be found on this link.

Which exhaust is the loudest?

Full dB testing data can be found on this link.

How do I quiet down an already loud exhaust?

1. Install an aftermarket silencer. This is an insert that is fitted into your tailpipe.
2. Install a resonated exhaust tip.
3. Have an inline resonator welded into your exhaust.
4. Install an auger style muffler or vortex insert cone.
5. View this thread for sound deading tips and a new found source of noise.

One special note about silencers. While they do quiet down your exhaust, they generally work by
restricting your exhaust flow down to a 1 1/2" hole. Strongly consider what this does to your
backpressure and the effects of backpressure on a turbocharged engine.

Which exhaust has the best gains? There has not been a 100% conclusive test of every
aftermarket exhaust to prove or disprove best gain level. Suffice it to say though that
theoritically, the exhaust with the largest pipe diameter along with a 100% baffless muffler
would be better than a smaller diameter and/or baffled muffler.

Are there any downsides to exhausts? There have not been significant amounts of problems
with exhausts. For STi and some 2.5L WRX owners, there have been reports of overboosting
issues with some TBEs. For more information read this post and the follow on posts it contains.

Where do I buy an exhaust? Every Subaru/Import performance store sells exhausts. For
purchasing, support your local economy or the NASIOC Vendors.

What should I decide upon before purchasing to narrow down my decision?

a. Sound level.
b. Exhaust tip(s) diameter and shape.
c. Catalytic converter issues with regard to your particular state's emissions laws.
d. Outside appearance, straight back exhaust or angled exhaust?
What questions should I ask of the retailer or manufacturer before purchasing?

1. Will this unit mate up to the rest of my exhaust? Some exhausts will fit the stock
exhaust, or must be fitted to a larger exhaust.
2. Will this unit throw a CEL? Some do, some don't, so find out beforehand.
3. Does it come with new gaskets? It is always a good idea to replace exhaust gaskets.
4. Does it come with a CEL fix if needed and is it the right one? There are two unique
electrical fixes, one for the 02/03 and one for the 04. The 05+ does not currently have an
electrical fix.
5. Will this exhaust clear the rear differential protector, if applicable?
6. Does this exhaust drone? Drone is a term to describe an annoying monotonous buzzing
sound generally heard during long highway trips. This sound can be more apparent with
wagons, due to the lack of a trunk to deaden the sound.
7. For 06 users: Will this exhaust clear my rear diffuser?

If after installation, my exhaust drones, how do I fix it? Many users report great success by
using audio sound deadening material in the trunk, back seat area, and/or underneath your
vehicle. View this thread for sound deading tips and a new found source of noise in the rear.

What is some good background information about exhaust pipe fitment?

Generally speaking, there are two types of catbacks:
Type 1: It has a sliding flange designed to bolt up to the stock downpipe.
Type 2: It has a flat flange designed to bolt up to 3" downpipes.

Generally speaking, there are two types of downpipes:

Type 1: It is designed to bolt up to the stock catback and other type 1 catbacks.
Type 2: It is designed to bolt up to only type 2 downpipes.

Type 1 catbacks do not bolt up well to type 2 downpipes, and type 2 catbacks do not bolt up well
to type 1 downpipes.
Type 1 catbacks and downpipes taper to 2.5" where they meet.
Type 2 catbacks and downpipes are full 3".
Type 2 downpipes are usually less expensive than type 1 downpipes.
Type 1 and 2 catbacks are relatively similar in price.

All 2.5" catbacks are type 1.

Some 3" catbacks are type 1, and some are type 2.
How hard is it to install an exhaust? Allow around one hour for install time. Professional
installation, depending on your area, is around $75. This is one vehicle modification that is very
simple and can be successfully accomplished by even the greenest shade tree mechanic.

How do I install an exhaust? Refer to the exhaust manufacturer's instructions. For exhausts
without instructions, below is a link to one of the better known exhaust installation instructions:
TurboXS's Instructions
Perrin's Instructions

Do I need aftermarket exhaust hangers? This can depend on many factors and the only clear
indication is during the post installation driving test. Some exhausts work perfectly fine with the
OEM hangers and some can benefit from the added stiffness or extended length of the
aftermarket hangers provide. They are fairly inexpensive and can be good insurance for a perfect,
rattle free exhaust install.

What about ceramic or other heat wrap coating? Though there are pros and cons to heat
retention treatments with exhaust components, the general consensus with headers is that heat
retention benefits outweigh the liabilities.
Pros:
Increased heat retention, which is beneficial for exhaust flow.
Reduces underhood temps, which decreases IC heatsoak.
Cons:
Additional expense.
Additional lead time, in the case of coatings.
Increased heat retention, which in some cases, can accelerate metal fatigue, especially at weld
joints.
No proven HP increase.
If you plan on using a coating, ensure you are aware of what impact this will have on your
header warranty if applicable. Coating or wrapping your header can void exhaust component
warranties. If using a ceramic-type coating, ensure it's working heat range is above the known
EGT. EGTs can be as high as 1400-1600 degrees in the extremes. Some normal coatings are not
rated to this temperature level.
What types of ceramic or other heat wrap coating are available? The two most popular
manufacturers of heat wrap are Thermo Tec and DEi. The two most popular coating
manufacturers are Jet Hot and Swain Tech Coatings. Wrapping is a DIY project with a shorter
life span than the professional only coating with a longer life span.

Excellent coating thread

.pdf file on coatings
Car Craft magazine's testing of Jet Hot

What is the CEL fix? It stops the rear CEL code P0420 Catalyst System Efficiency Below
Threshold from appearing, this code being caused by the lack of a catalytic converter in the
exhaust. This is usually fixed with a Pony Express MIL Eliminator (no longer in business),
Casper Electronics fix, or Perrin Performance fix. The 05+ does not currently have an electrical
fix. Another fix is the mechanical fix, which works on all models. ***NOTE*** There are two
CEL fixes for the WRX, the other one is the 2.2 KOhm 1/2 watt resistor. It is used to combat the
CEL code P0546 = Exhaust Gas Temperature Sensor Circuit Malfunction (High Input) in the
uppipe. This notice, though not important to this thread, acknowledges there are two "CEL fixes"
and defines them individually to avoid possible confusion.

Many aftermarket engine management systems can also be used to prevent the CEL light from
showing up.

Which is the best CEL fix, mechanical or electrical? Both work fine, though personnel living
in states with OBDII testing prefer the mechanical fix. The reason for this is during state
mandated OBDII testing, the electrical fix can show up as a malfunctioning sensor. For some
states, this one malfunctioning sensor is fine, for others, you will have to fix it and return. In the
end, the mechanical fix is cheaper and easier for most users.

Do I need engine management with an exhaust? For 02/03 WRX and STI users, no. For 04+
WRX users, it's pretty much a requirement as the open loop/closed loop fueling issue mandated
by the EPA causes long term reliability problems with using a downpipe and no tuning. For STI
users, the STI was the one exception manufacturers are allowed to have, but it's a wise move
regardless as the STI and 06+ WRX users have a good chance of boost creep occurring with
exhaust use. Tuning can ease this issue, but the cause/effect/cure of boost creep can be
researched via this link. This advice is for FULL exhausts, from the turbo back. If you are just
going the catback or axleback, there is no need for engine management.

Should I reset my ECU after this mod? It is never a bad idea to perform an ECU reset after a
mod. The traditional route is to disconnect the negative battery terminal, press the brake pedal
for a few seconds to bleed the system of charge, and reattach. Some use the more advanced
Vishnu Reset.
What about emissions problems like emissions tests and legality? Using a catless exhaust is a
violation of section 203 of the Clean Air Act. Each state has statutes that cover this subject as
well. As far as passing state emissions tests, as long as you still have one cat in your system, you
should be fine unless you run into a sharp-eyed and knowledgeable inspector who knows there
should be two in the main exhaust. The exhaust should be properly warmed prior to any testing
to ensure it is operating efficiently. Another smart move is to find someone in your local area
with the same set up to compare notes. Seek out locals via the NASIOC Chapters or NASIOC
Regional Areas. (A nice article on cat testing can be found here.)

What are some other good articles on exhausts?

Vishnu Design 101
Exhaust Design, Turbo and Naturally Aspirated

Editors Note

This post was created because I wasn't able to find a good exhaust FAQ. I came up with the text
based on LOTS of searching here. It was also created to be as brand neutral as possible so that it
serves as a stepping stone for further research. Certain brand names were mentioned in this post
though ONLY because they are the specific answers to specific questions. Upon reading this you
should have an idea of what type of exhaust best suits your needs. The manufacturer is up to you.

If you find an error in this FAQ, please PM me with factual details and I will update this post.
Responses such as, "I have XXX's exhaust and it's great!" or "XXX's exhaust leaked after 1
month" are not appreciated here, that is what the Car Parts Review Forum is for.
Wastegates
By the suicidal eggroll , original post

Also see Unabomber's Manifesto on the topic: http://forums.nasioc.com/forums/show....php?t=1190530

Wastegates are valves that are placed in the exhaust stream pre-turbo, and are used to bypass a
fixed amount of exhaust around the turbine of the turbo. By allowing some of the exhaust gasses
to bypass the turbine, the wastegate can be used to regulate the speed of the turbo, and ultimately
the boost pressure.

Both internal wastegates (IWG - the valve is built into the turbo) and external wastegates (EWG
- the valve is a separate unit installed in the uppipe before the turbo) operate the same way. You
have a port that allows exhaust to bypass the turbine, and you have a valve blocking that port.
The valve is attached to a diaphragm in an air chamber with an external port, and on the other
side of the diaphragm is a spring that keeps the valve closed when at rest. When pressure is
applied to the port, it pushes the diaphragm, compressing the spring, and opening the valve. See
the following two figures for a graphical description.

Low pressure, WG closed

High pressure, WG open

Typically, the port is plumbed into the pressurized section of the engine's intake, somewhere
between the outlet of the turbo and the intake manifold. The boost pressure is what is used to
push on the diaphragm and open the wastegate. It's a closed loop control system, where the boost
pressure is actually used to regulate itself by means of the wastegate.

The spring rate of the wastegate spring is adjustable in an EWG, but typically not in an IWG.
Stiffer springs require more pressure on the diaphragm to compress, which means the car will
run more boost before the wastegate opens and starts to bypass exhaust around the turbo.
Wastegate springs are actually rated at the pressure required to compress the spring, this is called
the wastegate spring pressure.
For example, take a wastegate with a 10 psi spring. When the boost pressure is well below 10
psi, the spring is able to keep the valve closed and force all of the exhaust through the turbine.
When the boost pressure approaches 10 psi, the spring starts to compress, allowing just a little bit
of exhaust to bypass the turbine. When the boost pressure reaches and exceeds 10 psi, the spring
compresses significantly, allowing a lot of exhaust to bypass the turbine. The happy middle
ground, where the spring is compressed just enough, to let just enough exhaust bypass the turbo,
to keep the turbo spinning at just the right speed, to keep the boost at a steady pressure, will be
right around 10 psi, the wastegate spring pressure (give or take a psi depending on external
conditions, such as temperature, atmospheric pressure, etc).

See the section on boost controllers to learn how they manipulate this feedback system to
increase boost without requiring a stiffer spring.

The following are images of actual wastegates. First a Tial 44mm EWG, then the IWG on a
VF39 turbo (the stock turbo on the 2004-2006 USDM STi).

Tial 44mm EWG:

VF39:
FAQ

Q: Why would somebody choose one size wastegate spring over another size? What is the
difference?
A: The answer is that each spring has an effective boost range that it can be used over. On the
low side, you can't have a boost target lower than the wastegate spring pressure. So if you want
to have a partial throttle boost target of 10 psi, then a 10 psi spring is the largest you can use. On
the high side, the further you deviate from the spring pressure, the more unstable your boost will
be. If you use a 5 psi spring and are trying to run 40 psi, your boost will be all over the place and
swing wildly with temperature, etc. Your compensations would have to be tuned PERFECTLY
in order to get something resembling reliable boost. Most people consider the maximum
reliable/stable boost level to be double the wastegate spring pressure. The combination of these
two restrictions gives you a valid range for your target boost. A 10 psi spring can be used with a
target boost anywhere between 10-20 psi, a 15 psi spring can be used with a target anywhere
between 15-30 psi, and so on.

Q: What is boost creep?

A: Boost creep occurs when the wastegate cannot physically bypass enough exhaust to keep the
turbo from spinning faster and the boost pressure rising uncontrollably. It can be helped by
porting the wastegate to allow it to bypass more exhaust, or by switch to an external gate with a
much higher flow capacity. Boost creep (true boost creep) cannot be fixed with tuning. If your
overboost problem was fixed with an adjustment to the tune, then the problem was not boost
creep, it was a bad tune. True boost creep is a mechanical problem, and must be addressed with a
mechanical solution.

Q: What is the difference between different sized wastegates?

A: Well the size obviously (just kidding). The larger the valve, the more exhaust it can bypass
around the turbine, so the lower you can set your boost target and the less chance you have of
encountering boost creep. That's about it, and the price. Though the smaller gates do tend to be a
bit more "screamy" and less "roary" when they open, if that matters to you.
Boost Controllers
Before reading this, please familiarize yourself with the operation and function of the wastegate here

The simplest boost control system available is just a hose running from the manifold or
intercooler piping straight to the wastegate. As the turbo spools up and pressure starts to build,
this pressure is fed directly to the wastegate. When the pressure reaches the wastegate spring
pressure, the wastegate will open, and the boost level will stop increasing, as any additional
pressure will just open the wastegate further and slow the turbo back down. The wastegate spring
pressure is about 7 psi on the stock turbo. Most people run around 12-15 psi wastegate springs on
EWG setups. See the diagram below for how this works:

The job of any boost controller on the planet is single-fold: to reduce the pressure seen by the
wastegate (versus the above example), so you can run more boost before it opens.
Factory boost control setup

The factory boost control setup is a bleed style. What this means is you have a hose coming off
of the intercooler piping (or turbo outlet, same thing), that runs through a restrictor pill to reduce
the flow rate, and then it runs into a T. One branch of the T goes to the wastegate (just like the
simple example above), the other branch of the T goes to the boost control solenoid, and then out
to the atmosphere or the intake. When the solenoid is closed, pressurized air from the turbo outlet
goes through the restrictor pill, through the T, and straight into the wastegate. The wastegate sees
the full pressure from the turbo, and will open once you hit wastegate spring pressure. When the
solenoid is open, pressurized air from the turbo outlet goes through the restrictor pill, through the
T, and is then bled off through the solenoid. The wastegate sees some pressure since the solenoid
can't bleed off everything, but the pressure it sees is lower than spring pressure, so the wastegate
doesn't open. End result is when the solenoid is at rest, it's closed and you run wastegate spring
pressure. When the solenoid is energized, it's open and you run infinite pressure.
See the diagrams below for how a bleed-style solenoid works:
Most aftermarket boost control systems are an interrupt style. In this setup, you have a hose
coming off of the turbo outlet (no restrictor pill), straight into the solenoid, and then the other
side of the solenoid goes straight to the wastegate. When the solenoid is open, air passes straight
from the turbo outlet, through the solenoid, to the wastegate which opens at spring pressure.
When the solenoid is closed, air passes from the turbo outlet to the solenoid, and is then blocked,
so the wastegate sees no pressure. The solenoid is also plumbed so it follows "reverse"
energized/rest logic compared to the bleed setup. End result is when the solenoid is at rest, it's
open and you run wastegate spring pressure. When the solenoid is energized, it's closed and you
run infinite pressure. See the diagrams below for how an interrupt-style solenoid works:
With either setup, the ECU rapidly pulses the solenoid to raise or lower boost as necessary. The
wastegate duty cycle (WGDC) is the percent of the time that the solenoid is energized. A WGDC
of 0% means the solenoid is always at rest and your boost will be as low as it can be. A WGDC
of 100% means the solenoid is always energized and your boost will be as high as it can be. A
WGDC of 50% means the solenoid is energized half the time, and at rest half the time. The ECU
constantly monitors the boost pressure and compares it to the target. It then increases or
decreases the WGDC as necessary to push the boost closer to target.
A manual boost controller (MBC) works similarly, and is also available in bleed or interrupt
styles. Rather than being open and closed electronically, an MBC has a fixed setting, which
always bleeds (bleed style) or blocks (interrupt style) a certain amount of air, to reduce the
pressure seen by the wastegate.

MBCs are less flexible than BCSs because they only have one setting (BCSs have an infinite
number of "settings", controlled by the WGDC). This means that MBCs are unable to run less
boost at partial throttle or high RPM, and they're unable to drop boost to wastegate spring
pressure in the event of a malfunction, for example. The reduction in flexibility is counteracted,
in part, by MBCs' good reliability and consistency.

BCSs need a significant amount of compensation in order to stay controlled. Without this
compensation, small changes in atmospheric temperature or pressure would cause drastic
changes in boost. This is because BCSs rely on the amount of time they're open or closed in
order to do their job. 5 milliseconds of energized time won't always bleed of the same amount of
pressure, for example, especially if the atmospheric pressure has changed.

MBCs, on the other hand, always drop off a fixed amount of pressure, so the wastegate always
sees the same thing and opens at the right time. Wide swings in atmospheric temperature or
pressure rarely cause more than ~1 psi of drift with an MBC.
For this reason, a popular boost control strategy is to actually use a combination of the two. An
interrupt style BCS for partial throttle and high RPM boost tapering, and an MBC for rock solid
wide open throttle boost. Whenever the user wants the MBC to take over, they set the BCS
WGDC to 100%. This closes the BCS, and allows the MBC to control the pressure drop to the
wastegate. Whenever the user wants the BCS to take over, they drop the WGDC to normal
values. This partially opens the BCS, and allows it to increase the pressure seen by the wastegate
over what the MBC would normally do, allowing the wastegate to open sooner than with the
MBC alone. This truly is a "best of both worlds" setup. It gives the user a minimum boost of
wastegate spring pressure, and a maximum boost controlled by the MBC. They are then free to
do anything they want in between those two extremes with the BCS, without worrying about
boost spikes or unstable boost due to improperly tuned atmospheric compensations. See the
diagram below for how this is set up:
FAQ
Q: Will I gain power with an aftermarket boost controller?
A: The answer is: "it depends". The boost controller itself will do nothing for power, however it
is possible that an interrupt-style boost controller will allow you to run more boost than a bleed-
style, and that will ultimately allow you to make more power. Same goes for an MBC, if you run
the same boost, you'll have the same power.

Q: Will I have to get re-tuned for any of these modifications?

A: Absolutely yes. Anything you do, at all, to the boost control system (when using an electronic
solenoid) needs to be tuned for. It's VERY easy to make the boost control system unstable,
causing under boosting, over boosting, oscillating boost, fuel cuts, etc. MBCs don't need the tune
to be altered when the boost level is changed per-se, but at a minimum the tune needs to be
checked to see if everything is still kosher at your new boost level.

Q: What happens if I run a stock (bleed) setup without a restrictor pill?

A: Without the restrictor pill, the solenoid won't be able to bleed off enough air to keep the
pressure down, so the wastegate will open early and you'll have very little boost (just a little
above wastegate spring pressure).

Q: Will turning up the boost make me run lean?

A: NO, no no no no no. I don't know who started this misconception, but it is completely, 100%
wrong. This idea is based on thought that our fueling system uses no sensors at all, that it just
reads your current throttle position and estimates how much boost you must be running, and how
much air flow that is. This is the most ridiculously oversimplified view of a modern fueling
system that I have ever heard. The fueling system in our cars uses a MAF sensor to measure the
ACTUAL incoming air flow rate. It then adds the necessary amount of fuel to hit the target AFR.
It doesn't care what the boost is, it injects fuel based on how much air there actually is. There are
only TWO ways that increasing boost will make you run lean, and they are "ifs", not certainties.
1) There is so much more air flow at this higher boost level that your fueling system simply can't
keep up. The ECU tries to inject the fuel for it, but there isn't enough capacity to actually do it,
due to a weak fuel pump or small fuel injectors. 2) There is so much more air flow at this higher
boost that you actually saturate the MAF sensor, so it can't even read how much air there actually
is. Other than these two extremes, you can do anything you want with the boost level without
affecting your AFR. Now that doesn't mean it's advisable, or even safe. There are many things
that can go wrong with arbitrarily turning up the boost level, but a lean AFR is not one of them
(usually).
Q: What is partial throttle full boost (PTFB) in regards to MBCs and is it dangerous
A: Short answer: with tuning, no it's not dangerous; without tuning: it's possibly dangerous. With
an electronic boost control solenoid (EBCS), the ECU is able to taper boost at partial throttle. So
if your wide-open boost target is 20 psi, you might want it to taper boost to 10 psi at only half
throttle. With an MBC, this is not possible; your target is full boost all the time. With a small
turbo, it's very easy to hit full boost with only a little bit of throttle input. With a TD04 at 3000
RPM, you might be hitting full boost at only 30% throttle. Now normally this would not cause a
problem, because the fueling system in our cars doesn't just guess what the air flow rate should
be given your current throttle position, it KNOWS what the air flow rate is using the MAF sensor
(see the fuel system article for more on this). If you're running 10 psi more boost at partial
throttle, it will measure the increase in air flow and inject the fuel for it, that's not the problem.

The problem is closed loop fueling (see the fuel system article for more on this). The closed loop
system operates on idle/cruise and uses a feedback system with the O2 sensor to maintain an
AFR of 14.7. When you get hard on the throttle, the ECU switches to open loop and the target
AFR drops to whatever it should be given your current RPM/load. The problem is the factory has
implemented a delay in this switch, called the closed loop/open loop delay (CL/OL delay). There
are many triggers that will cause the switch from CL to OL, one of which is throttle position, the
others are RPM, injector duty cycle, and more. If you're at low RPM and hammer the throttle, the
throttle position will trigger the switch to open loop sooner than the others, causing the car to
drop the AFR as necessary sooner than it would if it had to rely on the other switches. If you're at
low RPM and go to 50% throttle, this won't be enough to trigger the throttle switch, so it'll stay
in closed loop. With a low partial throttle boost target, this is fine, because the boost won't be
high enough to really NEED a low AFR.

With an MBC and PTFB though, your boost can spike very high before the other switches have a
chance to move you into OL and drop the AFR. The reason I said that with tuning it's not
dangerous, is because one of the first things that any tuner does on our cars is eliminate this
CL/OL delay. With no delay in place, the ECU will switch to OL as soon as the target AFR
requests it, regardless of what the current throttle position is. So while your gas pedal might only
have two settings (slow and holy****), at least the AFR will be safe regardless.
Blow off valves
By the suicidal eggroll - https://forums.nasioc.com/forums/showthread.php?p=36880743
Also see Unabomber's Manifesto on the topic: http://forums.nasioc.com/forums/showthread.php?t=468038

For some reason a lot of people get confused between about wastegates and blow off valves. Let
me start by saying that wastegates and blow off valves are not the same thing. They are not
installed in the same place, they do not have the same function, they do not make the same
sound, they do not make sound at the same time, and the operation of one is not affected in any
way, shape, or form by the operation of the other. They are as different as your eyeball and your
elbow. Now that that's out of the way, let's continue.

To understand what a blow off valve does, imagine you're driving. You're on the throttle hard,
the turbo is spinning at 100,000 RPM and is sucking in air through the filter like a cracked out
vacuum cleaner and pumping it out faster than a leaf blower on steroids. This air is flying into
the intercooler at 100 mph, into the engine, and forcing itself into the cylinders. The engine is
barely able to suck down all of this air, and the wastegate is bypassing exhaust around the turbo
to keep it spinning at the right speed. Got the picture? Now you let off the gas, the throttle plate
closes, the air flow into the engine drops to almost zero, what happens? Without a blow off
valve, as far as the turbo is concerned it just ran full throttle into a brick wall. It takes a long time
to spin down from 100,000 RPM, and the turbo has approximately 0.00001 seconds to do it
before ramming into pressure spike so high it would make your head cave in. Well, it can't, it
rams into the pressure spike, air starts flowing backwards through the compressor wheel, the
wheel all but stops in its tracks, and you get a loud pulsing/chugging sound as it does so. There
are all kinds of arguments on both sides of the debate about whether this is actually bad for the
turbo, I'm not going to get into that, but I will say that while it might not be bad for the turbo, it
certainly isn't good for the turbo.

Now introduce the blow off valve. This is a valve not unlike the wastegate or FPR, which is
placed somewhere in the intercooler piping between the turbo and the throttle body. Its sole
purpose is to stay closed until the pressure difference between the intercooler piping and the
manifold exceeds some threshold, at which point it opens. Throttle plate open, the pressure in the
intercooler piping before the throttle plate, and the pressure in the manifold after the throttle plate
is more or less equal, so the blow off valve is closed. Throttle plate closed and boost is high, the
pressure in the intercooler piping before the throttle plate is high, the pressure in the manifold
after the throttle plate is very low, so the blow off valve opens. The blow off valve gives the
turbo an outlet through which it can pump air while it slows down, since you closed the throttle
plate and the engine can't suck in any air anymore.
Let's repeat the above scenario now that we have a blow off valve. You're on the throttle hard,
turbo is spinning at 100,000 RPM, pumping air into the engine. You let off the gas, throttle plate
closes, manifold pressure drops, blow off valve opens. The turbo continues to pump air into the
intercooler piping, through the blow off valve, until it has slowed down to the point where the
intercooler piping pressure drops enough for the blow off valve to close (or until you hit the gas
and open the throttle plate back up, giving the turbo an outlet again and closing the blow off
valve). This gives the turbo time to slow down gently. While it does vent off all of the pressure
in the intercooler piping between shifts, it also keeps the turbo spinning faster through the shift,
so boost will come on quicker when you get back on the gas in the next gear.

Your car, in fact all turbo Subarus since god knows when, already has a blow off valve from the
factory, and it does its job well. There are two styles when it comes to blow off valves:
atmospheric, and recirculating (aka: bypass valves). Their names are pretty self-explanatory; at
atmospheric blow off valve vents the air to atmosphere, while a recirculating blow off valve
(bypass valve) plumbs this vented air back into the turbo inlet before the turbo. The OEM valve
is the latter.

What is the difference between a BOV and a BPV? Other than where the vented air goes,
nothing. They do the same thing in the same way. Where the vented air goes does matter though,
a lot. If you understand how a MAF-based fueling system like ours works (if not, read the article
on the fuel system, primarily the section on the ECU logic and how it uses the sensors to control
fueling), then this should make sense. Our cars use a draw through MAF system. This means that
the order of operations goes: filter -> MAF -> turbo -> intercooler -> engine. Notice that the
MAF is located before the turbo, the turbo has to draw the air through the MAF sensor, hence the
name "draw through MAF". Similarly, a "blow through MAF" system goes: filter -> turbo ->
intercooler -> MAF -> engine. I'm sure you can figure out why it's named blow through.

Now let's revisit our earlier example with the draw through MAF setup in mind. The turbo is
spinning at 100,000 RPM, sucking air through the filter, through the MAF (where the ECU is
measuring it to figure out how much fuel to add), and then blowing it through the intercooler and
into the engine. You let off the gas, throttle plate closes, blow off valve opens, charge air from
the turbo is diverted through the blow off valve rather than into the engine (since the throttle
plate is closed and the engine is drawing in very little air). With a recirculating bypass valve, this
air is routed back to the intake between the MAF sensor and the turbo. This creates a loop, the
turbo sucks in say 100 grams per second of air, it compresses and blows out 100g/s of air, 2g/s
enters the engine, the other 98g/s passes through the bypass valve back into the intake. There it's
joined by 2g/s being drawn in through the MAF sensor, and re-enters the turbo as 100g/s again.
The turbo is pumping 100g/s, but only 2g/s is being metered by the MAF, and only 2g/s is
entering the engine, everything is happy. Now what happens with an atmospheric valve? The
turbo sucks in 100g/s, it compresses and blows out 100g/s, 2g/s enters the engine, the other 98g/s
vents out into the atmosphere. The turbo is still drawing in 100g/s though, and since nothing is
coming in from the valve, the full 100g/s is passing through the MAF sensor. This means the
ECU is measuring 100g/s of air flow, and injecting fuel for 100g/s of air flow, but only 2g/s of
that is actually entering the engine. The AFR swings stupid rich, so rich it can't even combust,
the engine misfires, you get a huge fireball out of the exhaust, and the fueling system takes a
while to recover, delaying spool up in the next gear.

Not only that, but atmospheric valves can cause problems at idle as well. Remember that blow
off valves operate on the pressure difference between the intercooler piping and the manifold?
Well when you're sitting at idle, the intercooler piping is close to 0 psig, while the manifold is
somewhere around -10 psig. Depending on the valve, this could be enough to open it, just sitting
there at idle not doing anything. With a recirculating valve, that's not a problem. If it's open, that
just creates another path to the intake where it can grab metered air. With an atmospheric valve,
if the intercooler piping isn't EXACTLY at 0 psig, some air will either be entering or leaving
through the open valve, making you run lean or rich accordingly. This can cause stumbling,
hesitation, or even stalling at idle.

Some people claim that their atmospheric valve didn't harm their driving experience much at all.
I do not think that these people are lying, at all, but I DO think that they are more the exception
than the rule. These claims should be taken with a grain of salt, because for every one person
who doesn't have a problem with an atmospheric valve, there is at least one person who does. It
all depends on the specific valve being used, the spring in the valve, the specific car, and the
temperature and pressure where they live. If you want to avoid problems, stick with a
recirculating bypass valve.

There are some exceptions to the rule. Earlier I mentioned blow through MAF setups where the
MAF sensor is located after the turbo before the throttle body. In these setups, as long as the
valve is located before the MAF sensor, it can be atmospheric or recirculating, it makes no
difference, because the air being vented has not been metered by the MAF sensor yet. Another
exception is speed density, in which there is no MAF, the ECU uses the temperature and pressure
of the air in the intake manifold to estimate the amount of air entering the engine (using a
volumetric efficiency table instead of a MAF scaling table). As long as the air is being vented
before it reaches the intake manifold (which it has to be, otherwise it would defeat the purpose of
the blow off valve in the first place), then it doesn't matter if it's atmospheric or recirculating.
FAQ
Q: You spoke about the disadvantages, but are there any advantages to an atmospheric
blow off valve?
A: No, none. At BEST an atmospheric valve will make the vehicle run the same as with a
recirculating valve. At worst, the vehicle won't even run. In the real world, it's going to be
somewhere in between. The only thing, and I mean the ONLY thing that you gain with an
atmospheric valve, is that it's louder. If that matters to you, I would suggest installing an
aftermarket intake first, as the sound from the recirculating valve will resonate in the intake and
make it significantly louder. Remember though, that any aftermarket intake MUST be tuned for
(see the fuel system article for why).

Q: Will I gain power with an aftermarket valve?

A: Possibly. In an ideal world no, but we don't live in an ideal world. The OEM valve does leak
at high boost (old ones start leaking around 20 psi, new ones start around 17 psi). If your valve is
leaking at high boost, replacing it with one that doesn't leak will obviously help.

Q: Can the rich swing on shifts with an atmospheric valve be tuned out?
A: No. Contrary to popular belief, you can not "tune" for an atmospheric blow off valve. How
would you? There is absolutely no way to know how much air is being vented by the valve
versus how much is actually entering the engine, so there is no way to correct for it. There is also
no table in the ECU that says "remove X g/s from the measured MAF for Y seconds after I let
off the gas when the boost is above Z psi". None that are even close.
Blow Off Valve FAQ
By Unabomber - https://forums.nasioc.com/forums/showthread.php?t=468038

Term usage: "Blow off valves" go by several names, among them are compressor bypass valve
(CBV), air by-pass valve, bypass valve (BPV), blow off valve (BOV), Diverter valve, and
possibly a few others. BOV is the common and incorrect term that lumps true blow off valves
and bypass valves under the same term. For the sake of correctness, this post will refer to either
aftermarket BOV, aftermarket BPV or OEM BPV as these are the most correct terms.

What is the function of a blow off valve (BOV)? To release pressure from the intake tract of a
turbo car when the throttle closes. It is a vacuum-actuated valve designed to releases the air to
the atmosphere.

What is the function of a bypass valve (BPV)? To release pressure from the intake tract of a
turbo car when the throttle closes. It is a vacuum-actuated valve designed to recirculate the air
back into the intake before the turbo inlet, but after the airflow sensor.

What is the purpose of a BOV/BPV? When the throttle closes and the intake system is under
pressure, the high-pressure air entering the motor will bump into the closed throttle plate, and in
the absence of a BOV/BPV, a pressure wave will travel back to the turbocharger. The result is
that the compressor wheel will stall (a phenomenon known as “compressor surge”) and slow
down very quickly. This is hard on the bearings and decreases the turbo’s lifespan, but it also
means the turbo will take longer to spin up the next time the throttle is opened.

Are aftermarket BOVs necessary with Subaru turbos? No. The OEM BPV is perfectly fine
up to 20psi of boost. For applications using higher boost levels, an aftermarket BOV/BPV should
be considered.

Can I mod my stock BPV to hold higher boost? Yes. This link gives detailed instructions on
how to do so.

Is the STi BPV better than the WRX BPV? No. They are the same. However, the JDM STi
BPV will hold more boost as it is physically different than both the USDM STi BPV and the
WRX BPV. The specific PSI rating of the JDM STi BPV is unknown, but users have reported it
is good up to 25 PSI.

Have the OEM BPVs changed over the years? Yes. 08+ Subarus have a BPV that is plastic.
So if you for some reason you have a plastic BPV, it will not hold as much boost as the old metal
ones. The plastic ones will only hold ~17PSI give or take.
Is an aftermarket BPV better than the stock BPV? No. Unless you are considering an
aftermarket BPV solely for the purposes of holding higher boost levels. An aftermarket unit
should sound just like the OEM unit.

Which manufacturer is best? This topic is highly debated. There have been no reported
consistent "bad" aftermarket BOVs. Obviously, there may have been bad ones sold, but not
enough to report as "bad" overall.

What are the different types of aftermarket BOVs/BPVs? Different manufacturers use
different methods. There are three basic types:
1. Aftermarket BPV: Similar in function to the OEM BPV where 100% of the air is recirculated.
2. Atmospheric BOV: 100% of the air is vented to the atmosphere.
3. Hybrid BOV: These depend on the manufacturer and end user settings. These can either be
adjustable or manufacturer set for different percentages of atmospheric/recirculation dumping.
They can also be set to work as recirculation during lower boost conditions and 100%
atmospheric during higher boost conditions.

Are there any downsides to aftermarket BOVs? There have not been significant amounts of
problems with BOVs. Aftermarket BOVs can and do require some light end user maintenance to
keep them performing perfectly. For aftermarket hybrid BOVs that have end user defined
settings, there will be an initial period of adjustment to obtain the desired
recirculation/atmospheric ratio. As well, most aftermarket BOVs will require "tuning" (usually
via supplied washers, a screw, or other mechanism on the BOV itself) to allow them to idle
correctly and blow off at the right time.

Are there any negative effects with aftermarket BOVs? Yes. The downside of releasing the
air to atmosphere is that it has already been metered by the mass air sensor, and when it blows
off, the ECU will be injecting the wrong amount of fuel into the cylinders. The engine
temporarily runs rich, meaning too much fuel is injected into the cylinders. On most tunes the
target A/F under boost is @11.1:1 or so. Say you are at 11.1:1, then you shift and it vents. It will
swing rich, typically to around 9.5:1. That is not that rich and this period lasts for under one
second...again, nothing to write home about.

This temporary rich condition isn’t usually that harmful. Technically, it can eventually foul spark
plugs and even clog the catalytic converter as unburned fuel on the catalytic converter burns very
hot, and too much of it can melt the cat. The odds of either of these two conditions actually
happening is very, very low though, but that's the theory.

As to blanket "you'll run rich" statements, a BOV will only run "rich" during hard acceleration
and shifting as 99% of the time it stays closed.
Can my tuner or engine management tune out this rich period? Yes. There are some forms
of engine management that can tune this out. Buying your engine managment soley for this
purpose is a poor method of choosing an engine managment system though.

What about a 50/50 or BOVs that you can portion the VTA portion? This is a bad analogy,
but if a BOV is a person in a wheelchair, a 50/50 BOV is a person in leg braces. It's not as bad,
but not good enough to say bolt it up. If you find a deal on one or happen to like the sound of a
particular model, go for it, but don't think you are doing your car better vs. a 100% VTA model.

Won't I be labeled as a ricer? It's your car, do with it what you like, but be aware that many
people are prejudiced against BOVs just like neon underbody lights and other accessories. Be
aware that a BOV can be seen as a dinner bell for street racers though. It attracts attention both
good and bad.

So a BOV is not bad then right? Not so fast...most BOVs leak. Even ones that are religiously
maintained, installed, and are from quality manufacturers. A leak in essence means less boost
and less power or your car will be working harder to produce boost than it means, which can
decrease reliability.

How do I set up my BOV to not leak/stand less chance of leaking? Read the BOV
manual/instructions first off. Learn how to adjust it. If it doesnt have a manual, just look at the
valve. You can almost aways figure out how to adjust it. Some have screws, some use washers
under the spring, others use replacement springs. Now that you know how to adjust it, here is
what you do:

First off, start the car, get out and open the hood. If you can see the piston of the valve (look in
the hole it vents out of if its a VTA valve), check to see if it is moving at all at idle, or worse,
partially open. You want that piston rock solid, not moving at idle.
Now give it a little gas. It shouldnt move when you step on it, and when you lift, it should pull up
a bit. Depending how hard you hit the gas, it will either JUST move, or fully open.
If its doing anything but what it should be, adjust it a little tighter. Renew your makeup or beauty
inventory with Priceline Catalogue.

Now go drive the car

When you shift at light throttle, is it venting? If so, is it a pure straight vent, or do you get a little
bit of a "chufchchchc" noise right before it vents? That noise is compressor surge. Which isnt as
bad as some people like to think and say. Ideally, you are looking to hear a little surge then a
vent. Or no sound at all. Either is ok. Now drive it hard, does it surge a little or just straight vent?
Get out of the car and tighten the valve and drive it again. Repeat this until you get a LOT of
surge when you lift under boost. Now start loosening it until you JUST start to hear no almost
surge.
What the surge means is that it fought opening just a little bit. If you hear a LITTLE surge just as
it begins to vent, that means that the valve is holding itself shut as tightly as it is capable of,
without being so tight that it doesnt vent properly.

I have had XXXX brand valve for years, it has never leaked. How do you know? By looking
at your boost gauge? Looking at the boost levels in your datalog? Neither of those prove that the
valve isnt leaking. The advice above tells you how to set it and KNOW for 100% sure.

But compressor surge will kill my turbo! Unless its really bad, it really isnt going to kill it.

Where do I buy aftermarket BOVs/BPVs? Every Subaru/Import performance store sells

BOVs. For purchasing, support your local economy or the NASIOC Vendors.

How hard is it to install aftermarket BOVs/BPVs? Allow around 1 hour for install time.
Professional installation, depending on your area, is around $75. This is one vehicle modification
that is very simple and can be successfully accomplished by even the greenest shade tree
mechanic.

How do I install BOVs/BPVs? Refer to the BOV/BPV manufacturer's instructions. For

BOVs/BPVs without instructions, below is a link to one of the better known installation
instructions:
scoobymods.com instructions (with photos)

If you are wanting to be 100% sure you arent losing any power from a BOV, and you are
running under about 22PSI, you are best off sticking with a stock unit.

Editors Note

My thanks to Dan of Mach V Motorsports for writing an excellent BOV article that provided a lot of the
information contained within this FAQ. He also provided additional information that was helpful in the
formulation of this FAQ. Also thanks to my buddy Davenow for his excellent BOV post found HERE
that contains additional information.

This post was created because I wasn't able to find a good blow off valve FAQ. I came up with the text
based on LOTS of searching here. Upon reading this you should have an idea of whether a blow off valve
best suits your needs or not. The manufacturer is up to you.

If you find an error in this FAQ, please PM me with factual details and I will update this post. Responses
such as, "I have XXX's blow off valve and it's great!" or "XXX's blow off valve broke after 1 month" are
not appreciated here, that is what the Car Parts Review Forum is for.
Boost Control FAQ
https://forums.nasioc.com/forums/showthread.php?t=1190530 by Unabomber

The genesis behind this FAQ is mainly due to the rampant "monkey see-monkey do" approach
due to the fairly recent introduction of aftermarket boost control solenoids. While they do play a
role in the overall boost control world, the press they have received has gotten way out of hand,
so this FAQ will discuss your boost control options.

What is the function of a boost control device? To control boost. While this seems to be an
easy definition the reason n behind boost control is lost to many people. A boost controller of
any type is designed to limit boost. Yes, you read that right***8230;.limit boost. Without a boost
controller of any kind, your turbo would spin up until it explodes and produces upwards of
probably 50PSI. This seems to be the most unknown fact of boost controllers, that they limit
boost. Now that you have your mind wrapped around the proper function of a boost control
device, you can explore the rest of the story.

Do I need to consult with my tuner about boost control options? Yes. Items in your car that
can be tuner specific: manual boost controllers, electronic boost controllers, upgraded
wastegates, restrictor pills, wastegate helper springs, external wastegates (type/size/brand), spark
plug types/brands/gaps, and injector types/brands/sizes. ALWAYS take your tuner's advice on
these matters no matter what the internet tells you!

What controls boost on my stock turbocharged car? The two main components are the boost
control solenoid located on the passenger side strut tower and the internal wastegate on the turbo
itself.

What is an internal wastegate? A spring/diaphragm based mechanism which controls the

movement of the wastegate valve. A turbo wastegate is normally closed, forced shut by a
compressed spring inside the actuator canister. As pressure is applied to the nipple of the
canister, the wastegate shaft moves away from the actuator, swinging open the wastegate valve.
When the wastegate opens, this allows air to bypass the turbine inside the turbine housing. Less
flow = slower spinning of the turbo, so you can grasp the concept of how this controls boost.
Can I adjust my internal wastegate? Pre-loading the wastegate actuator arm, also known as
adjustment of the wastegate actuator rod (if the rod length is not fixed and adjustments can be
made) will allow proper calibration and some additional tuning. All IHI turbochargers have a
fixed wastegate actuator rod that cannot be adjusted, while all MHI turbochargers have an
adjustable wastegate actuator rod. If the rod coming out of the wastegate actuator is shortened, it
will pre-load the spring inside the wastegate actuator increasing the pressure level at which the
actuator will allow the wastegate valve to open and the total boost pressure that the turbo can
generate will increase (as long as the turbo is still within its efficiency range). This pre-load will
also limit how far the wastegate valve can open. Pre-loading (shortening) the wastegate actuator
rod too much can potentially create a mechanical boost creep issue that cannot be tuned out. If
the wastegate actuator rod is lengthened, the actuator will decrease the load on the spring and
decrease the pressure level at which the actuator will open and the total boost pressure the turbo
can generate will decrease. If the wastegate actuator rod does not put enough pre-load on the
wastegate valve, then you could see boost fluctuations of + or - 2psi even when the wastegate
solenoid duty cycles are constant. Help to adjust it can be found HERE.

What about a helper spring? A helper spring is a clever, cheap, easy, and removable way of
wastegate adjustment as well. Basically its a small spring that stretches from the wastegate door
arm to the wastegate itself. This puts more tension on the wastegate flapper door so the actuator
has to work harder to open the door. Installation and more helper spring advice can be found
HERE.

What is an external wastegate? External wastegates are used for two purposes:
a. To increase the flow when using high PSI
b. Required for larger turbos that have no internal wastegate

Using an external wastegate on an internally gated model will show benefits as it effectively
reduces choking of the turbo. Choking is where the flow gets backed up due to the more
restrictive internal wastegate. By replacing the internal with an external, this generally allows
one to run more boost safely. While every turbo can benefit from an external unit, it really shines
on larger turbos.

How can I run an external wastegate? You will need some custom fabrication and parts.
Generally speaking you will need an uppipe fitted with a flange to run an external wastegate.
Then you need a wastegate suited to your application. You also need to consider plumbing of the
wastegate gasses.

External wastegate tuning tips can be found HERE

Where do I plumb my wastegate? Plumbing is the routing of wastegate gases. You can "vent to
atmosphere" which is where the wastegate opens and air is directed straight to the outside air.
Downsides to this are the need to route the venting so it won't heat up anything in the engine bay
and the increased noise. If noise is a factor, you might consider having a small motorcycle or 2
stroke muffler welded onto the end of your vent pipe, room permitting. The other, quieter option,
is to plumb the external unit back into your downpipe. The key to successfully doing so is to
ensure it is plumbed far away from the turbo (at least 18" past the downpipe to turbo flange, and
hopefully beyond any catalytic converters in your system) and ensure the external piping is
welded to the downpipe at a narrow angle to encourage aft flow vs. say a 90 degree entry that
will encourage more back pressure into the exhaust. The downside to a "plumbed back"
wastegate is even with a perfectly designed system, it is not as efficient as a vent to atmosphere
system.

Which manufacturer of external wastegates is best? This topic is highly debated, but the
general consensus is TiAL is the king of the hill.

Are there different sizes of external wastegates? Yes. 38 mm and 44 mm are the two most
talked about and utilized sizes.

Which size is right for me? That is a question for your turbo vendor and tuner as sometimes
their advice counters the external wastegate sizing theory. The theory is that a larger wastegate
(44 mm) should be utilized on turbos using low boost vs. turbos that are running high boost
which should run a smaller wastegate (38 mm). Where did this theory come from?
Remember***8230;a wastegate limits boost. If you want to run more boost, you don't have to
limit it as much. Running less boost means you have to limit it more. Think this premise through
to get a good mental picture of it as this concept is terribly misunderstood.

Some also erroneously put credibility on the TiAL's 44 mm model due to its superior V-Band
clamp vs. the 38 mm models flat flange design as the decision maker. While the V-Band is an
overall better design, the problem with TiAL's 38 mm model is that it apparently ships with what
many consider being a sub-standard gasket. Replacing this gasket to start with or after leaks
occur generally solves the issue. Funny how a $3 gasket will convince some to spend $200 more
on a 44 mm unit as well as encourage others to do so.

Useful thread that discusses TiAL pros/cons

What about external or internal wastegate springs? Think back to the wastegate sizing
question above. Once again, you have a theory that needs to be buttressed by the opinions of
your turbo vendor and tuner. The theory is that if you want to run say 15PSI on your turbo, you
want to run a spring that is 50% of that value or as close as you can get it for proper boost
control.
How do I swap out wastegate springs External wastegates are serviceable by the end user, so
refer to your external wastegate documentation for details as well as ordering information for
new size springs. Generally speaking, most internal wastegates need to have the whole assembly
swapped out as removing spring, finding the new correct spring, and replacing the spring is not
able to be performed. There are a few sources of upgraded internal wastegate solenoids as
Deadbolt, AVO, and others sell them.

What is a restrictor pill? This component limits the amount of pressurized air flowing from the
turbo compressor housing. The restrictor pill restricts the air flow so the wastegate solenoid
valve and wastegate actuator are not overdriven, which would force the wastegate valve to open
prematurely.

Can I tune with this restrictor pill? A smaller diameter hole in the center of the brass restrictor
pill will have a higher tendency to create boost spike in the system and require less wastegate
duty cycle to run higher boost. The larger the diameter hole in the center of the restrictor pill, the
less chance the boost control system will boost spike and greater wastegate duty cycle will need
to run in order to produce higher boost. If you have installed a new turbocharger and you are
using the stock boost control system to tune boost, please verify that the vacuum line coming off
the turbo compressor housing (prior to the T-fitting) contains a restrictor pill with a hole
machined in the center of the pill.

The stock boost control system most commonly uses a restrictor pill with a center hole size of
0.040"-0.048" +/- 0.001"

For larger-than-stock turbochargers or turbochargers with a stronger mechanical spring in the

wastegate actuator you will need to use a restrictor with a larger center hole, something along
0.050"-0.055" +/- 0.001"

For similar-to-stock-sized turbochargers with a weaker mechanical spring in the wastegate

actuator you will need to use a restrictor with a smaller center hole, something along 0.030"-
0.040" +/- 0.001". Be very careful when using a restrictor with a center hole of this size, there is
a higher tendency for the system to boost spike and you will need less wastegate duty cycle to
run higher boost.

Where do I get new restrictor pills? You can modify your OEM one via obtaining special
small drill bits and a special small hand drill to drill it out. One can also obtain suitable sized
nitrous jets and then file down the outside housing to fit inside the vacuum line. As well, many
tuners have a collection of various sized restrictor pills.
What are aftermarket boost control solenoids? Perrin and Prodrive make purpose built Subaru
unit, though the GM unit may be retrofitted as well. These units mimic the purpose and function
of the OEM boost control solenoid.

What makes them so special? Nothing really. They serve the exact same purpose as the factory
unit. Since they are slightly different in actuation values, they do require a custom tune. So what
has appeared to happen is people with an off the shelf tune or no tune have purchased these.
Then they get tuned with them. Then they attribute much of their gains, not on tuning, but on the
boost control solenoid. While they can help, this is something to discuss with your tuner as one
option for better boost control as 95% of the good/bad of this modification lies in the tuner's skill
vs. the boost control solenoid. The biggest difference in aftermarket units is that they are
intercept style vs the OEM bleed style boost control. Intercept style allows for better boost
response when compared to the OEM bleed style.

So how does a boost control solenoid work? It is opened via a pulsed duty cycle based on
inputs from the ECU. While you are building boost, the BCS is not operating. This allows the
wastegate to remain closed to built boost quickly. As target boost approaches, it ramps up
quickly to control/limit the boost. At this point is where the BCS is working the hardest.

More BCS interaction help : Article 1, Article 2

Why is tuning needed on aftermarket ones? Port sizing and duty cycles can change with them
which can lead to wild results. Even tiny changes in the turbo/wastegate associated hose length
and diameter can cause drastic changes. Do not use a GM, Perrin, or Prodrive boost control
solenoid until you are ready to be tuned.

What about a manual boost controller or electronic boost controller? With proper use and
tuning they are other possibilities for controlling boost. These can be utilized as standalones for
02/03 WRX or STI models with judicious use and the proper knowledge. They may also be
effective in conjunction with most engine management systems with proper tuning. These need
to definately be discussed with your tuner prior to installation though.
Is there a simple, general guide to all these items? There is now:

Small boost change that require no tuning:

• Pressure check your system for leaks. How to do so can be found HERE.

Small boost change that may require tuning:

• Adjustment of internal wastegate. There are no hard and fast rules, but if you are only
adjusting the wastegate for 1 PSI or so, you can probably safely get away with this.
Anything over that and you want to strongly consider a tune.
• Helper spring.

Big boost changes that require tuning:

• Aftermarket boost control solenoids

• Manual/electronic boost controllers
• Upgraded internal wastegate actuator
• Going to external wastegate
• Changing springs inside the external wastegate
• Restrictor pill swaps

So what do I need to assist controlling my boost? More than likely, nothing. Tuning is what
ultimately controls your boost using the wastegate and BCS. If tuning doesn't cure your ills,
listen to your tuner's advice as to what steps to proceed with. Do not surprise your tuner on the
day of the tune with an aftermarket boost control solenoid or other fix because Tim in Kansas
had "great results" with his. Tuning the boost control system is the hardest and most time
consuming thing a tuner can do. You might have the latest "fad" boost control solution on the
market, but will it be worth it when you put on a $175 wastegate actuator and it takes you tuner 1
extra hour of his and dyno time to the tune of $300? Consider he might be able to do the same
thing in no extra time with a 10 minute swap out of a $2 restrictor pill.

And remember...if you are currently have boost control issues, you should fix them first rather
than trying one of the methods described above to band aid fix the issue. For example, if you are
only seeing 7 PSI, adding a spring to your wastegate to bring it back up to the normal 14.7 PSI is
not the correct way to fix the problem. Troubeshoot and fix the main issue vs. performing a work
around.
Where do I buy aftermarket boost control devices? Many NASIOC Vendors offer them.

How hard is it to install aftermarket boost control devices? Most are fairly simple to install
and many come with directions that should be referred to and followed.

GM install instructions are HERE, though they would help with Perrin and Prodrive as well.
Another useful BCS install thread can be found via this link.

Editors Note

My thanks to Christian of www.cobbtuning.com for his assistance that provided a lot of the
information contained within this FAQ. Another wonderfully informative boost control read that
is authored by him can be found HERE.

This post was created because I wasn't able to find a good boost control FAQ. I came up with the
text based on LOTS of searching here. Upon reading this you should have an idea of whether a
boost control solution best suits your needs. The type/manufacturer is up to you.

If you find an error in this FAQ, please PM me with factual details and I will update this post.
Responses such as, "I have XXX's BCS and it's great!" or "XXX's wastegate broke after 1
month" are not appreciated here, that is what the Car Parts Review Forum is for.
Turbos
By the suicidal eggroll

This is not going to be a turbo selection guide. This is not "what works best on what car", "what
will get you to your target power", etc. There are plenty of those on the forum, just go to Proven
Power Bragging. This is about how a turbo actually works. Once you know how a turbo works,
understanding how the wastegate does its job, for example, becomes a bit easier. This is actually
going to be a pretty short article.

A turbo is essentially two fans connected by a shaft. The exhaust side is called the turbine (or the
"hot side", because of the hot exhaust); the intake side is called the compressor (or the "cold
side", because of the cool intake air). Exhaust gasses at very high temperature and pressure enter
the inlet of the turbine. They use that energy to spin the turbine to very high speeds, losing a lot
of pressure in the process (and as a result, dropping the temperature). After the exhaust gasses
exit the turbine, they enter the downpipe and continue out through the rest of the exhaust. The
turbine is connected by a shaft to the compressor. The compressor's blades are similar to those of
the turbine, but reversed, so that when it is spun by the turbine, it draws fresh air in through the
intake, compresses it, and blows it out of the turbo outlet (essentially the exact same thing the
turbine is doing, but in reverse).

This high pressure before the turbo is why you can't trust the O2 sensor's readings when installed
in the uppipe. The temperature drop across the turbo is why you have to put EGT sensors before
the turbo in order to get good measurements from them.

Now why do we want to compress the air entering the engine? The compressed air naturally is
denser, which allows you to squeeze more air molecules into a smaller space. As discussed in the
fuel system article, air is what matters when it comes to making power. You can inject as much
fuel as you want, but unless you have the air to burn that fuel, it does no good. This extra air
allows a smaller engine to produce the power of a larger engine.

All "normal" car engines have 4 cycles: intake, compression, combustion, and exhaust. In the
intake cycle, the exhaust valves are closed, intake valves are open, and the piston is pulled down
by the crankshaft. This draws air and fuel into the cylinder through the intake valves like a
syringe. In the compression cycle, the intake valves close, and the piston is pushed back up by
the crankshaft. Since there is no outlet, this compresses the air and fuel. Near the top of the
piston's travel, the spark plug fires, igniting the air/fuel mixture and pushing the piston back
down (rotating the crankshaft), this is the combustion cycle. Finally, the exhaust valves open and
the crank pushes the piston back up, forcing the exhaust out of the engine. Then the cycle
repeats. Regardless of the number of cylinders in an engine, all of them will complete one full
combustion cycle every two revolutions of the crank (down up down up). This means the engine
pumps one full displacement of air through the system every two revolutions. For a 4L engine,
that's 4L of air every other revolution, 2L per rev.

Now ignoring inefficiencies, pressure losses through the intake, etc., that means our 4L engine
will pump through 2L of air at atmospheric pressure (14.7 psi) and temp (call it 70F) every
revolution. According to the ideal gas law, PV=nRT, or pressure * volume is equal to the amount
of air * a constant * temperature. When you compress the air, the pressure (P in our equation)
rises. Assuming the temperature stays more or less the same, this allows you to lower V, the
displacement of the engine, without affecting the amount of air that's available to burn (n), and
ultimately the amount of power you can make. Ignoring changes in compression, temperature,
etc., a turbo or supercharged car running 14.7 psi of boost (29.4 psi absolute) can make the same
amount of power as a naturally aspirated car (14.7 psi absolute) with half of the displacement.

Now in the ideal world you have inefficiencies, etc. that don't make it quite this cut-and-dry, but
the result is still pretty much the same. Add boost, and you can reduce displacement without
affecting the amount of air you pump through the system and the power you can make. Or on the
flip side, add boost, and you can drastically increase the amount of power you can make in an
engine without having to increase displacement.

That's about it, turbos in a nutshell. Most people on this site probably already know this, but
that's alright.
Turbo Compressor Maps
By the suicidal eggroll

I apologize ahead of time, this is going to be a math-heavy article. The math is not difficult, but
there will be a lot of it. You can't talk about turbo compressor maps without getting into the
math, after all the ultimate goal of reading the compressor map is getting the compressor
efficiency, but what good is that if you don't have the equation with which to use it?

Alright, so what do turbos do? They take advantage of the high temperature of the exhaust
leaving an engine, and use it to compress the intake air entering the engine. See the turbos article
for why this is advantageous if you don't know already.

Unfortunately, every time you compress a gas, you increase its temperature. A perfect
compressor will do this adiabatically with no input or release of heat. A turbocharger with 100%
efficiency would follow this equation:

𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 = 𝑇𝑇𝑖𝑖𝑖𝑖 + 𝑇𝑇𝑖𝑖𝑖𝑖 ∗ ((𝑃𝑃𝑜𝑜𝑜𝑜𝑜𝑜 /𝑃𝑃𝑖𝑖𝑖𝑖 )0.263 − 1)

(Remember to use absolute units for your temperature, either Rankin or Kelvin)

As an example, imagine a perfect turbo, 100% efficiency. 70F intake temp, 14.7 psi atmospheric
pressure, 14.7 psi of boost (29.4 psia). The outlet temperature would be:

𝑇𝑇𝑜𝑜𝑜𝑜𝑜𝑜 = 70 + ((70 + 459.67) ∗ ((29.4/14.7)^0.263 − 1)) = 175.9 𝐹𝐹

Unfortunately, turbos aren't perfect, they have some inefficiencies that cause them to increase the
temperature even further than this. The real equation is:

𝑇𝑇𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑇𝑇𝑖𝑖𝑖𝑖 + 𝑇𝑇𝑖𝑖𝑖𝑖 ∗ ((𝑃𝑃𝑜𝑜𝑜𝑜𝑜𝑜 /𝑃𝑃𝑖𝑖𝑖𝑖 )0.263 − 1)/𝐸𝐸𝐸𝐸𝐸𝐸

Where Eff is the compressor efficiency as a fraction (0.7 for 70%, etc)

Repeating the above example gives us an outlet temperature of 221.3 F with a compressor
efficiency of 70%.
That's where turbocharger compressor maps come in. These are charts which show the
compressor efficiency as a function of flow rate and pressure ratio (Pout/Pin). An example for
the Garrett GT3076R is shown below:

First, a description of how to read the map:

The x-axis "Corrected Air Flow (lb/min)" is the air flow rate in pounds per minute. The MAF
sensor gives you air flow rate in grams/second, it's just a simple conversion to go between the
two.

The y-axis "Pressure Ratio" is simply Pout/Pin. Make sure you use absolute pressure. So if the
atmospheric pressure is 14.7 and your boost pressure is 12, your pressure ratio (Pr) would be:

(12 + 14.7)/14.7 = 1.82

Now the map itself; you'll see 7 more or less horizontal lines, and on the right side of each are
very large numbers, starting at 55864 and continuing to 144956. This is the RPM of the
compressor wheel. It's not terribly useful for this analysis, but useful fun to know when it comes
to selecting a turbine housing to go with your compressor. You'll also see roughly 8 concentric
rings that look sort of like a target. Each one is labeled with a percentage from 60% to about 77%
(it's hard to read the inside one). This is the compressor efficiency we discussed earlier.

Now it's time for the extremes in the compressor map. On the left side we have the stall region.
This is when the turbo is spun up and blowing out a lot of air, but the engine can't take it all in
due to low RPM or poor efficiency. The pressure builds, but still the engine can't take it all in.
What happens is the pressure builds too much, and it starts pulsing backwards through the
compressor wheel, very similar to what happens when you release the throttle without a blowoff
valve to vent the air. This will cause a bucking or surging feeling/sound, and is usually the result
of a mismatched compressor size and turbine size for the engine.

On the top, we have one last RPM line at 144956 RPM and nothing beyond. This is the choke
point, where the tips of the compressor blades start to break the sound barrier. When that
happens, efficiency drops to almost nill, which is why you see nothing above that line.

On the right side, the turbo will continue to operate, but the efficiency drops off quickly. When
somebody says "if you run that turbo at that boost at high RPM it's just going to blow hot air",
that's what they're referring to. Efficiency starts to drop so rapidly that you really aren't gaining
any flow, you're just increasing the temperature.

On the bottom, I'm not 100% sure on what happens. I believe the turbo simply isn't spinning fast
enough to really compress the air, the air just kind of flows through it. Don't quote me on that
though.

So how do you read this map? It's pretty simple in theory, but a bit more difficult in practice.
We'll start with the simple theory, and then go into the complications.

Pressure ratio is the easy one, manifold pressure divided by atmospheric pressure and you're
done. Air flow rate is a bit more difficult, because the air flow rate into the engine depends on
several things: engine displacement, engine speed, engine volumetric efficiency (rarely known,
but it can be estimated around 85-95%), air pressure, and air temperature. But wait, doesn't the
temperature of the air depend on the efficiency of the turbo? Yep, it's an iterative process. Guess
an efficiency, calculate a flow rate, update the efficiency using the compressor map, update the
flow rate, and so on. Luckily the efficiency doesn't swing too wildly with changing air flow rate
(only maybe 10-15%), so it only takes a few iterations to nail this down.
Before we can start calculating the air flow rate into the engine, we have to talk about another
piece of the puzzle, most turbocharged setups (at least the ones that run relatively high boost) use
an intercooler. The job of the intercooler is simple, to cool this hot air back down before entering
the engine, to help prevent detonation and to increase the density of the air even more.
Intercoolers have an efficiency as well, which describes how efficient the intercooler is at
cooling the charge air from the turbo down to the temperature of the outside air passing over the
intercooler. The intercooler outlet temperature follows the equation:

𝑇𝑇𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑇𝑇𝑖𝑖𝑖𝑖 − 𝐸𝐸𝑓𝑓𝑓𝑓 ∗ (𝑇𝑇𝑖𝑖𝑖𝑖 − 𝑇𝑇𝑎𝑎𝑎𝑎𝑎𝑎 )

Where Eff is the intercooler efficiency as a fraction (0.7 for 70%), Tin is the inlet temperature
calculated above (same as the turbo outlet temperature), and Tamb is the temperature of the
outside air passing over the intercooler. As you can see, a 100% efficient intercooler will cool
Tin to exactly Tamb, while a 0% efficient intercooler will not cool Tin at all.

Actual intercooler efficiencies vary wildly, anywhere from maybe 40% for a heat soaked ebay
top mount at low speed to well over 90% for a high quality front mount on the freeway. This part
of the calculation is easily the least known. Very little people, hell very little COMPANIES
actually go out and measure intercooler efficiency, it's sad really. They all publish how much air
they can flow with minimal pressure drop, but very few of them actually tell you what the
cooling efficiency is while flowing that much air, which is easily the most important part. After
all, are you buying an intercooler so it can flow a lot of air, or are you buying it to cool your air?
A straight pipe will flow a ton of air, but it won't cool anything.

So where are we now? You start with air at atmospheric temperature and pressure. Your turbo
compresses this with some associated efficiency, heating it up. The air passes through the
intercooler which cools it back down, and finally it enters the engine. According to the ideal gas
law, we have everything we need now. Air flow rate should be equal to:

𝑛𝑛 = 𝑃𝑃𝑃𝑃/𝑇𝑇 ∗ 𝐶𝐶

Where P is the pressure, V is the volumetric flow rate (engine displacement * RPM/2), T is the
temperature, and C is a constant.
There's one last piece though, the engine's volumetric efficiency. The value calculated above is
the theoretical maximum air flow rate given the engine displacement, RPM, and the temperature
and pressure of the air. In order to actually make it into the cylinders though, this air has to flow
very quickly through a lot of pieces such as the TGVs and valves. As the air blows through the
engine on its way to the cylinders, there is some pressure loss associated with it. While the
pressure might be 30 psi in the manifold, it might only be 29 psi in the cylinder by the time the
intake valves close and compression starts. This is caused by the engine's volumetric efficiency,
which is expressed as a percentage of the theoretical maximum amount of air (calculated above)
that actually makes it into the cylinders after losses. When you port the heads, remove the TGVs,
install high lift/duration cams and large diameter valves, the volumetric efficiency is what you're
increasing (or at least trying to increase). The VE of most engines is somewhere around 85-95%.
Some very well engineered engines have the ability to approach, or even slightly exceed 100%
due to very clever designs, ours are not those engines.

The VE has a very straight forward effect, rather than the earlier equation of:

𝑛𝑛 = 𝑃𝑃𝑃𝑃/𝑇𝑇 ∗ 𝐶𝐶

You get:

𝑛𝑛 = 𝑃𝑃𝑃𝑃/𝑇𝑇 ∗ 𝐶𝐶 ∗ 𝑉𝑉𝑉𝑉

Where the VE is expressed as a fraction (0.9 for 90%).

Alright, we're just about there. Just a couple more complications and we're done. First is the
pressure loss before the turbo. The act of drawing the air through the filter, MAF, and several
feet of tubing will drop the pressure from atmospheric before it enters the turbo. This doesn't
change your boost target, after all 12 psi above atmospheric is 12 psi above atmospheric, but
what it does do is increase the pressure ratio (Pr) for the turbo.

With 1 psi of loss in the intake, rather than having a Pr of 1.82 at 12 psi at sea level, instead you
have a 𝑃𝑃𝑃𝑃 = (12 + 14.7)/13.7 = 1.95.

The last complication you have to take into account is charge loss. The boost level you look on
your gauge or in your logs at is the pressure at the manifold, but between the turbo outlet and the
manifold there are a lot of pieces the air has to pass through (piping, bends, intercooler, and
throttle body), each one dropping off some pressure. With 1 psi of charge loss, rather than having
a Pr of 1.95 at 12 psi at sea level, instead you have a 𝑃𝑃𝑃𝑃 = (13 + 14.7)/13.7 = 2.02.
Alrighty, that's it. The turbo draws air through the intake (which leaches some pressure),
compresses it and heats it according to its efficiency. The air then passes through the intercooler,
which drops the pressure a bit due to restrictions, and drops the temperature according to its
cooling efficiency. Finally the air enters the manifold, where it then passes through the TGVs,
heads, and valves and into the cylinders, dropping off some more pressure along the way
according to the engine's volumetric efficiency. Finally, the air is in the cylinders, and you get
compression, combustion, and power.

The process for actually doing these calculations involves a lot of iteration, as you can probably
guess. For that reason, I write an Excel spreadsheet to do the calculations for you. You'll still
have to iterate using the compressor map (enter a compressor efficiency, look at the flow rate
and Pr, update the efficiency, and so on), but it definitely simplifies the process. Unfortunately
there's no way to get around the assumptions that have to be made about intercooler efficiency
and engine volumetric efficiency, so take any results with a grain of salt.

Here is the spreadsheet:

http://forums.nasioc.com/forums/show....php?t=2158062

So how do you use all of this information? Simple actually. Estimate your pressure drops,
estimate your intercooler efficiency (try to get this close, but it doesn't need to be exact), estimate
your engine's volumetric efficiency (same), pick a boost target and an RPM, and plug them in to
the calculator (or the raw equations if you want). There will be a value entered for the
compressor efficiency already, so you'll get outputs for the pressure ratio and air flow. Then look
at the compressor map for the turbo you're interested in, use the Pr and air flow from the
calculator to get the actual efficiency from the map, plug that in and iterate a couple of times
until you converge. If the Pr and air flow you calculate are outside the bounds of the compressor
map, the simple answer is that that turbo cannot run at that pressure on that engine at that RPM
for whatever reason (below the surge point, above the choke point, etc).

You'll want to repeat the analysis for a few different RPMs to give yourself a range (before
spool, after spool, peak power, and redline), and then print out your compressor map and draw
lines between each of these points on the map. Make sure your desired spool isn't in the surge
region of the turbo, make sure your redline isn't beyond the choke point of the turbo, and see
how close to the efficiency peak of the turbo that you'll pass through. Then pick a turbo that
meets your requirements, or try to run your existing turbo within its efficiency range to maximize
power.
How to Enjoy Big Turbos
Lots of people on the forums believe that anything bigger than a 20G won’t be fun to drive. They
are wrong. Half of them have no idea what they’re talking about, and the other half don’t know how to
drive. I suspect they got the “OMG lagz WTF” idea from looking at dyno charts that show big turbos
hitting full boost at 4000 RPM or more.

Those charts are very misleading. It’s true that if your turbo hits full boost at 4000 RPM and you
floor it at 3000 RPM, you’re basically going to wait for a naturally-aspirated engine to take you to 3500
RPM before things get interesting. That’s a waste of time. It’s not fun at all. There’s a very easy way to
fix that problem though: stop doing that.

If you have a big turbo, you just have to get into the habit of keeping RPM high when you want
power. You get much less lag that way. If I am trying to accelerate, I’m in the 4000–7000 range. When I
start a pull at 4500 RPM, the chart looks like this:
The other nice thing about keeping RPM high is that it means driving in a lower gear that you
would otherwise. So, not only does boost come on quickly, you get more acceleration from
whatever boost you have, simply because you’re in a lower gear.

Around the time when I was trying to decide on a turbo, I went for a ride in a friend’s Corvette,
with a big naturally aspirated V8. That was when it hit me that when you want acceleration, you
keep RPM high, even if you don’t have a turbo. High RPM is another way of saying “lower
gear” and “lower gear” means more torque at the wheels, and more torque at the wheels means
more acceleration. That was the day that I realized that I would be perfectly happy with a turbo
that didn’t wake up until about 4000 RPM.

At 2500 RPM my car is boring to drive - it’s almost like a naturally aspirated 2.5L. That’s
perfectly fine for getting groceries, so I’m OK with that. Also, that fact has no impact at all on
my car’s performance characteristics. When I want performance, I am in the 4000–7000 range.
Bam. Problem solved.

With a smaller turbo I’d have less torque from the motor, and since it would run out of breath
before redline, I’d have to shift sooner, so I’d be handicapped by a taller ratio as well. Some
people will pay those penalties, just to avoid downshifting. I don’t think that’s a good trade.

Want numbers?

My turbo hits 20psi at about 3800 RPM in 3rd. If I am cruising in 5th or 6th at 2500 RPM, and I
just floor it, nothing happens. It 6th gear I hit 20 psi at about 3200 RPM, which sounds great,
until you realize that in 6th gear it takes ten seconds to get from 2500 to 3200. That hardly even
counts as acceleration, so I have only done that once, just to see how long it would take.

On the other hand, if I am cruising on the freeway at 65mph in 6th, and I downshift to 3rd, I’m at
5000 RPM when I release the clutch. From there I get full boost after about a half-second. And
the car accelerates a lot faster in 3rd than in 6th because not only do I have more boost sooner, I
have much better gearing. So, two seconds after releasing the clutch, I would reach the redline, at
about 85mph. (But that would be illegal, so of course I never do that either.)

And just for completeness…. When I shift while drag racing, I push the clutch pedal, floor it
while shifting gears, and then release the clutch pedal. Boost builds quickly with full throttle at
high RPM. So even though I shift as fast as I can, I’ve got 10psi when I release the clutch, and it
only takes another quarter-second to get 20psi. But, I know exactly where to have my car towed
if I break something while racing, and for DDing I don’t plan ahead that much.

(I have spec B gears, so the numbers above are slightly different than a regular 6MT, but the
principle is the same.)
List of Turbo Spool + Whp characteristics (averages)
By WoohooRX - original post

I've been researching turbos for the past few days, and realized there's no cohesive turbo spool +
whp thread. I understand that it all depends on the TUNE and not all dynos are created equal,
however, there seems to be an AVERAGE/COMMON spool/whp for most turbos.

I was hoping there could be a single thread, where newbies, like myself, could have a starting
point, just so we don't see a million "____ turbo spool characteristics" or similiar type threads
(which there have been a plethora of in the past few days)

This is what I've gathered going through the Factory 2.0L and PPB sections. Please keep in
mind, these are all AVERAGES, there is a wide spectrum, absolutely dependent upon the tune.

I'll also need your help, please add anything I've missed, or correct me if I'm wrong in
something.

The following is some of the more popular turbos which people ask about. A range of average
spool times, and lastly, average Whp with accompanying basic mods (TBE, Injectors, Fuel
Pump, IC if required), all data below is on 91-93 oct. pump gas.

Spool = The rpms listed below reflect the time where the turbo is fully engaged, ie the point
where your head gets pushed back against the seat.

Turbo_____________Full Spool_____________Avg Whp_______Max Whp

Stock Td 04.........................2900 - 3100......................150 - 185

Td 04..................................2500 - 3000.......................210 - 230............. .245
VF 23..................................3300 - 3500........................unsure
VF 34..................................3300 - 3500.......................250 - 270........... ...300
VF 30..................................3500 - 3700.......................240 - 270..............
VF 39..................................3750 - 4000.......................250 - 270..............
VF 22..................................3500 - 3800.......................265 - 290..............
Small 16g.............................3500 - 3700.......................260 - 280..............
Big 16g................................3500 - 3700.......................260 - 290..............
EVOIII 16g..........................3500 - 3700.......................260 - 300..............
18g..................................... 3750 - 4000........................280 - 310+............
20g.................................... .3800 - 4200........................310++................. .

(The use of P&P, alky/meth, AVCS, etc. will usually decrease the spool times, and increase whp)
For the Max Whp line, keep in mind the following:

1. Dyno reading uncertainty

2. Highly modded, usually includes P&P, and may include alky/meth
3. Validated by atleast 1 person

Please don't bite my head off, I know spool times and whp threads usually end up in people
arguing, but hopefully this can be a ROUGH guide.

Let me know if you want to add any other turbos to the list.

VF series turbos: Information taken from amukaoen

(http://forums.nasioc.com/forums/show...highlight=vf23)

VF22:
This turbo has the highest output potential of all of the IHI VF series turbos and is the best
choice for those who are looking for loads of top end power. The top end power however, does
not come without a cost. The VF22 spools significantly slower than the rest of the IHI models
due to the larger P20 exhaust housing and is much less suited for daily driving than some of the
other models. Although the largest VF series turbo, the VF22 is not quite optimal for stroked
engines or those who wish to run more than 20PSI of boost. The VF22 is best mated with the 2.0
block and in applications of 20lbs of boost or less.

VF30:
The VF30 is commonly considered the best bang for the buck turbo in the IHI VF series line. A
relatively new model the VF30 features the the same exhaust housing as the VF24 but a larger
compressor side similar to the VF22. The combination of these two parts results in increased
output potential without the lag associated with the VF22. Although it doesn't offer the top end
supremecy of the VF22, the VF30 is a great compromise between these unit and the quicker
spooling models.

VF34:
The VF34 is nearly identical to the VF30, with the same exhaust housing and compressor.
However the VF34 goes back to the ball bearing design, and in doing so achieves full boost
approximately 500RPM sooner than the comparable VF30. The VF34 is the most recent IHI
design and as such costs slightly more than its counterpart. Top end performance and maximum
output are identical to the 30.
VF23
This turbo is considered a great all-around turbo. Like the VF22 it utilizes the largest P20
exhaust housing. This housing is mated with a smaller compressor housing of the of the VF24
for fast response and excellent low and mid-range performance. It does not have the same top
end power of the VF22, but spools up significantly quicker. Standard ball bearing center section.
Excellent bolt-on replacement for the standard WRX turbo on cars without any other major
modifications. Expect flows around 430 CFM at 18 PSI.

Provided by Dark Rex (*Information was derived from many sources including NASIOC, MRT
website, PPMotorsports website, clubwrx.net, scoobytech.net, WorldOne and Rallispec website.)

Last edited by WoohooRX; 09-21-2007 at 05:55 AM.

Turbo Break-in Procedure
1. Thoroughly clean out any oil supply lines to the turbocharger and oil coolers.

2. Change the engine oil and use a high-quality engine oil for turbocharger break-in. 40wt oil, or
thicker is recommended for our performance turbochargers.

3. Install the turbocharger and attach all bolts, lines, oil supply, etc. but DO NOT START THE
CAR!!! AND DO NOT UNDER ANY CIRCUMSTANCES USE ANY TYPE OF SILICONE
SEALANT ON THE OIL SUPPLY LINE!!!

4. Prime the oil system by cranking the engine over (WITHOUT STARTING IT) for 10-15
seconds, and repeat this step 3-5 times minimum. There are many ways to make sure that your
car will not start, you can unplug the coil wire on vehicles with distributor-style ignition, or on
vehicles with direct fire ignition, you can disconnect the coil packs, or the ignition relay. It is
VERY important that your car DOES NOT START while pre-oiling the turbo.

5. After you have completed the above step you can start the car and allow it to IDLE ONLY.

6. Allow the vehicle to idle for 10-15 minutes without applying any throttle. Even at idle, the
turbo can be spinning at as much as 28,000 RPM’s on certain models. Use this time to check for
any loose fittings, hoses, exhaust leaks, and to bleed the cooling system of any trapped air.

7. After idling the car as prescribed above, you may then begin to drive the car. Make every
attempt to limit boost for the first 100-200 miles.

*For our performance line of journal bearing turbochargers, a upgraded oil feed line and Banjo
bolt is needed to prevent oil starvation.

*For Subaru vehicles you must remove any AVCS banjo bolt filters to avoid oil starvation.
Subaru TSB, “Turbocharger Oil Supply Mesh Screen #02-103-07”
Oil Consumption Causes –
https://www.e46fanatics.com/threads/what-does-oil-consumption-have-todo-with-forged-internals.453625/

What is considered normal or acceptable, however, will vary from one engine application to the
next. For example, Ford Motor Company considers consumption as high as one quart of oil per
1000 miles to be acceptable in a gasoline engine. For large diesel engines used in over the road
trucking applications, many manufacturers are not concerned until consumption reaches one
gallon oil oil per 10,000 miles of operation.

Leaking Turbocharger Seal

A leaking turbocharger seal will draw oil into the combustion chamber where it will burn and
form carbon deposits which contribute to further oil consumption as they interfere with proper
engine function.

Restricted Air Intake

Excessive restriction in the air intake system will increase engine vacuum and can increase oil
consumption as noted in "High Engine Vacuum". A heavily plugged air filter would be one
example of this situation.

Oil Viscosity
The use of an oil with a viscosity that is too light may result in high oil consumption. Refer to the
vehicle owner's manual for the proper oil viscosity to be used under specific driving conditions
or ambient temperatures.

Aftermarket Performance Chips and Modifications

Increasing performance through the use of performance/power enhancement products to a stock
or factory engine can increase the chance of excessive oil consumption.

High Engine Vacuum

Engine vacuum has increased in modern engines due to the fact that engine rpm, valve overlap
and compression habits have also increased with these models. Some of the late model engines
will draw as high as twenty five inches of vacuum on deceleration, as compared to twenty inches
in older engines. This high vacuum characteristic has made it necessary for the development of
an oil ring to seal both (top & bottom) sides of the ring grooves and eliminate oil from passing
around the back and sides under high vacuum or deceleration. Such vacuum could be the main
cause of smoking and oil consumption so it is important that you use a side sealing piston ring
when called for.

Late Valve Timing

Late valve timing will keep the intake valve closed too long after the intake stroke has started,
and will increase the vacuum in the cylinder. The high vacuum will have a tendency to suck oil
up past the piston and rings into the upper part of the cylinder where it will be burned.
Oil Pressure Too High
An incorrect oil pressure setting or a faulty relief valve may cause the oil pressure to be too high.
The result will be that the engine will be flooded with an abnormally large amount of oil in a
manner similar to that which occurs with worn bearings.

External Oil Leaks - Some of the many points where external leaks occur may include: oil
lines, crankcase drain plug, oil pan gasket, valve cover gaskets, oil pump gasket, fuel pump
gasket, timing cover and camshaft bearing seal. No possible source of leakage should be
neglected because even a very small leak will cause extremely high oil consumption. For
example, it has been estimated that a leak of one drop of oil every 20 feet is approximately equal
to a loss of one quart every 100 miles. The best way to check for external leaks is to road test the
vehicle with a large piece of light colored cloth tied under the engine. Oil on the cloth will
indicate a leak which should be traced back to its source.

Front or Read Main Bearing Seals - Worn front or rear main bearing seals almost always result
in oil leakage. This can only be determined when the engine is operated under load conditions.
Bearing seals should be renewed when worn because a slight leak will result in extremely high
oil consumption just as it would with an external oil leak.

Worn or Damaged Main Bearings - Worn or damaged main bearings throw off an excessive
amount of oil which flows along the crankshaft and is thrown up into the cylinders. The amount
of oil throw off increases rapidly when bearing wear increases. For instance, if the bearing is
designed to have 0.0015 inch clearance for proper lubrication and cooling, the throw off of oil
will be normal as long as this clearance is maintained and the bearing is not damaged in any way.
However, when the bearing clearance increases to 0.003 inch, the throw off will be five times
normal. If the clearance is increased to 0.006, the throw off will be 25 times normal. When the
main bearings throw off too much oil, the cylinders are usually flooded with more than can be
controlled by the piston and rings. This causes burning of the oil in the combustion chamber and
carboning of the piston and rings.

In a conventional, full-pressure lubricated engine a large loss of oil at the main bearings may
starve the downstream connecting rod bearings of lubrication to such an extent that sometimes,
especially at low speeds, insufficient oil may be thrown on the cylinder walls. This will cause the
pistons and rings to wear to such an extent that they will not be able to control the oil at high
speeds. The effect of main bearing wear will be high oil consumption.

Worn or Damaged Connecting Rod Bearings

Clearances on connecting rod bearings affect the throw off of oil in the same proportions as
mentioned for main bearings. In addition to this, the oil is thrown more directly into the
cylinders. Worn or damaged connecting rod bearings flood the cylinders with such a large
volume of oil that the pistons and rings, which re designed to control a normal amount of oil or a
reasonable increase in the normal amount, are overloaded to such an extent that some oil escapes
past them to the combustion chamber and causes high oil consumption. CAUTION - Insufficient
bearing clearance can also produce piston, ring and cylinder damage as well as damage to the
bearing itself.
Worn or Damaged Camshaft Bearings
Camshaft bearings are generally lubricated under pressure and, if the clearances are too large,
excess oil will be thrown off. Large quantities of this oil may flood valve guide and stem areas
resulting in increased oil consumption.
Worn Crankshaft Journals
Worn crankshaft journals will have the same effect on oil consumption as worn bearings. When
they are worn out-of-round, they cannot be set up with round bearings to give uniform oil
clearance. A bearing fit to the larger dimension of a worn journal will be loose at the smaller
dimension and throw off many times the proper amount of oil. Journals which are out-of-round,
rough or scuffed should be reground and fitted with undersize bearings of the correct size.

Tapered and Out-of-Round Cylinders

In slightly tapered and out-of-round cylinders, the oil can be controlled by the pistons and
rings. However, with increased taper and out-of-roundness, satisfactory oil control becomes
more difficult to maintain. This is due to a combination of many factors. The increased piston
clearances permit the pistons to rock in the worn cylinders. While tilted momentarily, an
abnormally large volume of oil is permitted to enter on one side of the piston. The rings, also
tilted in the cylinder, permit oil to enter on one side. Upon reversal of the piston on each stroke,
some of this oil is passed into the combustion chamber.
For each revolution of the crankshaft, the pistons make two strokes - one up and one
down. When an engine is running at 3000 R.P.M. (approximately 60 miles per hour) the rings in
tapered and out-of-round cylinders are changing their size and shape 6000 times per minute.
Consequently, at high speeds, the rings may not have time to conform perfectly to all worn parts
of the cylinders on every stroke. Whenever this occurs, the engine consumes higher amounts of
oil due to what is commonly referred to as oil pumping.

Distorted Cylinders
Cylinders which are distorted so that they are out of shape - not from wear, as described under
"Tapered and Out-of-Round Cylinders", but from other causes, such as unequal heat distribution
or unequal tightening of cylinder head bolts - present a surface which the rings may not be able
to follow completely. In this case, there may be areas where the rings will not remove all of the
excess oil. When combustion takes place, this oil will be burned and cause high oil consumption.

Clogged "PCV" Valve

The main purpose of the PCV (positive crankcase ventilation) valve is to recirculate blow-by
gases back from the crankcase area through the engine to consume unburned hydrocarbons.
Blowby is a mixture of air, gasoline and combustion gases forced past the rings on the
combustion stroke. The PCV system usually has a tube leading from the crankcase to the
carburetor or intake manifold. Vacuum within the engine intake manifold pulls blowby gases out
of the crankcase into the combustion chamber along with the regular intake of air and fuel.
A valve can become clogged with sludge and varnish deposits and trap blowby gases in
the crankcase. This degrades the oil, promoting additional formation of deposit material. If left
uncorrected, the result is plugged oil rings, oil consumption, rapid ring wear due to sludge
buildup, ruptured gaskets and seals due to crankcase pressurization, oil thrown out around the
filler cap and consequent rough engine operation.
Honing Abrasive
If cylinder honing or glaze breaking is performed on an engine, cleaning instructions should be
carefully followed to prevent metal fragmentation or abrasive damage to the rings' seating
surfaces.

Cleaning instructions for reconditioned cylinders: After honing thoroughly wash cylinder walls
with soapy water and a scrub brush and oil immediately thereafter, or swab cylinders with No. 10
oil and carefully wipe clean. Repeat until all evidence of foreign matter is removed. In either
method that is used, a white cloth wiped on the surface should remain clean.
Note: Do not use gasoline or kerosene to clean the cylinder walls after honing. Solvents of this
nature will not remove the grit from the cylinder wall and often carry particles of abrasives into
the pores of the metal. Failure to properly clean the cylinder walls will leave abrasives that will
cause rapid wear and ring failure and will result in elevated oil consumption.

Worn Ring Grooves

For piston rings to form a good seal, the sides of the ring grooves must be true and flat - not
flared or shouldered - and the rings must have the correct side clearance in the grooves.
Normally, automotive ring groove side clearance should not exceed .002-.004. As the pistons
move up and down, the rings must seat on the sides of the grooves in very much the same way
that valves must seat to prevent leakage. New rings in tapered or irregular grooves will not seat
properly and, consequently, oil will pass around behind the rings into the combustion chamber.
Worn grooves are usually flared or tapered causing increased side clearances which permit more
than the normal amount of oil to pass the rings into the combustion chamber. Excessive side
clearances also create a pounding effect by the rings on the sides of the piston grooves. This
promotes piston groove wear and, if the condition is not corrected, breakage of rings lands may
occur.
Cracked or Broken Ring Lands
Cracked or broken ring lands prevent the rings from seating completely on their sides and cause
oil pumping by a process similar to that described above in "Tapered and Out-of-Round
Cylinders". In addition to this, they also lead to serious damage of the cylinders as well as
complete destruction of the pistons and rings. Cracked or broken ring lands cannot be corrected
by any means other than piston replacement and this should be done as soon as there is the
slightest indication of a crack.

Worn Valve Stems and Guides

When wear has taken place on valve stems and valve guides, the vacuum in the intake manifold
will draw oil and oil vapor between the intake valve stems and guides, into the intake manifold
and then into the cylinder where it will be burned. If this condition is not corrected when new
piston rings are installed, an engine is likely to use more oil than it did before because the new
piston rings will increase the vacuum in the intake manifold. When gum or deposits on the valve
stems are removed - a procedure recommended when overhauling an engine - the seal previously
formed will be removed and leakage will be more pronounced. This is particularly true on
overhead valve engines where loss of oil may occur on the exhaust valves as well as on the
intake valves. High oil consumption caused by too much valve guide clearance can frequently be
cured by reaming or nerraling the valve stem. In some cases new valves may also be required.
Use of a permanently bonded valve stem seal will give added insurance against oil leakage on
complete engine overhauls or on valve jobs.
Bent or Misaligned Connecting Rods
Bent or misaligned connecting rods will not allow the pistons to ride straight in the cylinders.
This will prevent the pistons and rings from forming a proper seal with the cylinder walls and
promote oil consumption. In addition to this, it is possible that a bearing in a bent rod will not
have uniform clearance on the crankpin. Under these conditions, the bearing will wear rapidly
and throw off an excessive amount of oil into the cylinder.

Worn or Improperly Fit Wrist Pins or the Wrong Pins

The use of worn or improperly fitted wrist pins or the installation of the wrong pins, as in the
case of rifle drilled rods where oil is forced to the wrist pins under pressure, can cause such an
excessive throw off of oil onto the cylinder walls that the piston rings may not be able to control
it. This will not only result in the direct loss of the excess oil but also in the formulation of
carbon which will clog the oil passages and cause the rings to become stuck in the grooves.

Wrist Pins Fit Too Tightly

Wrist pins that are fitted too tightly at both ends prevent the pistons from expanding and
contacting freely under the repeated heating and cooling encountered in engine operation. The
piston distortion results in scuffing or scoring, which inevitably leads to blow-by and high oil
consumption.

Clogged Oil Passages

After an engine has had long, hard service the oil passages in piston rings and pistons will likely
become clogged from carbon or an accumulation of foreign matter in the oil. The passages are
designed for carrying oil - in excess of the amount needed for lubricating the cylinders - back to
the crankcase. When the passages become clogged, oil may be trapped in areas reducing the
indicated level of oil within the engine. It may also pool in areas such as above the valve guides,
which can further promote consumption.

Clogged passages in the rifle drilled rods or any clogged oil line will starve the engine of
lubrication, promote wear and lead to high oil consumption. To avoid clogging of oil passages,
the same precaution should be taken as recommended in "Pistons Rings Stuck in Grooves".
Initial side clearance is not applicable in this case.

Unequal Tightening of Main Bearing Bolts or Connecting Rod Bolts

Unequal tightening of main bearing bolts or connecting rod bolts will throw the bearing bores
out-of-round enough to shorten bearing life and to cause an abnormally large throw off of oil
from the bearings. The effect on oil consumption is described in "Worn or Damaged Main
Bearings" and "Worn or Damaged Connecting Rod Bearings". When bearing bores are originally
machined, at the time of engine manufacture, the bolts are tightened to the manufacturer's torque.
A torque wrench must be used to insure roundness of the bearing bores whenever the bolts are
tightened after having been removed and reinstalled. Unequal tightening of connecting rod bolts
may also cause connecting rod distortion, with results similar to those described in "Bent or
Misaligned Connecting Rods".

Unequal Tightening of Cylinder Head Bolts

The strains developed by unequal tightening of cylinder head bolts may cause serious cylinder
distortion and result in oil pumping as mentioned in "Tapered and Out-of-Round Cylinders" and
"Distorted Cylinders". When re-installing a cylinder head, a torque wrench should always be
used on the head bolts. The engine manufacturer's instructions should be followed for the torque
readings and the sequence in which the bolts are tightened.

Dirty Cooling Systems

Rust, scale, sediment or other formations in the water jacket and radiator, or corrosion of the
water distributing tube, will prevent a cooling system from performing its duties efficiently. This
is likely to cause cylinder distortion with a direct loss of oil as mentioned in "Tapered and Out-
of-Round Cylinders" and "Distorted Cylinders".

A defective cooling system causes overheating of the engine with the possibility of developing
localized hot spots in some of the cylinders. This may also lead to scuffing and scoring of
cylinders, pistons and rings which results in high oil consumption.

Dirty Oil
Failure to change the oil at proper intervals or to take proper care of the oil filter may cause the
oil to be so dirty that it will promote clogging of the oil passages in the piston rings and pistons.
This will increase the oil consumption as described in "Clogged Oil Passages". Dirty oil will also
increase the rate of wear on bearings, cylinders, pistons and piston rings. All of these worn parts,
as explained in individual items on each part, will contribute to a further waste of oil. Note: as a
rule, dirty oil by nature is also consumed at a higher rate than cleaner oil.

Too Much Oil in Crankcase

Due to an error in inserting the oil dip stick so that it does not come to a seat on its shoulders, a
low reading may be obtained. Additional may be added to make the reading appear normal with
the stick in this incorrect position which will actually make the oil level too high. If it gets so
high that the lower ends of the connecting rods touch the oil in a pressure lubricated engine or
the dippers go too deep into the oil in a splash lubricated engine, excessive quantities of oil will
be thrown on the cylinder walls and some of it will work its way up into the combustion
chamber.

Incorrect Piston Rings for Type of Engine or Type of Service

If rings of an incorrect size are installed (for instance, .020" oversize rings in .040" oversize
cylinders) they can readily cause oil pumping because they will not fit the cylinders and will be
unable to keep the oil down from the upper cylinder walls. In this example, ring end gap will also
be greater, resulting in additional oil loss, as described in "Piston Rings Fit with Too Little End
Clearance". Different types of engines and their use in different types of service require
individually engineering ring sets which vary in many ways. Each set has been designed for a
particular purpose, but if one is used in an engine for which it is not intended, it may be
incapable of controlling the oil in that engine. It is extremely important to always make sure that
the correct set is used.

Worn Timing Gears or Chain

Worn timing gears or chain can cause the valves (and sometimes the distributor) to be out of
time with the crankshaft. The large amount of backlash, which is caused by this wear, will
prevent proper engine adjustment because timing may vary from one revolution of the crankshaft
to another. When the valve and piston motions are not synchronized, extremely high oil
consumption may result. This will be caused by excessive vacuum which draws large quantities
of oil into the combustion chamber where it will be burned.

Piston Rings Fit with Too Little End Clearance

When fitting new rings, care must be taken to see that, with the rings in the smallest part of the
cylinder, sufficient end clearance is allowed for expansion due to heat. Normal gap clearance in
automotive engines with cast iron rings usually runs .003-.005 per inch of bore diameter. The
rings will heat more rapidly and will operate at a higher temperature than the cylinder because
they are exposed to the direct heat of the burning gases from the combustion chamber. The
cylinder walls are kept at a lower temperature by the water in the water jacket. This means that
the rings expand more than the cylinder and this expansion must be allowed for by use of a gap -
known as end clearance - between the two ends of each ring. If sufficient end clearance is not
provided, the ends of the rings will butt while the engine is in operation.

Butting will cause scuffing and scoring of rings and cylinders which leads to oil consumption. If
the engine is allowed to be used for continued operation, especially under heavy load, scoring
will become more severe. The ends of the rings will be forced inward - away from the cylinder
wall - so that a space opens up between the rings and the cylinder. This provides a direct path for
hot gases from the combustion chamber to burn the oil on the cylinder and greatly increases the
oil consumption of the engine. Severe cases of butting may also cause ring breakage, with the
same results as described in "Worn or Broken Piston Rings". Excessive ring end clearance leads
to increase oil consumption as well.

Worn or Broken Piston Rings

When piston rings are broken or are worn to such an extent that the correct tension and
clearances are not maintained, they will allow oil to be drawn into the combustion chamber on
the intake stroke and hot gases of combustion to be blown down the cylinder past the piston on
the power stroke. Both of these actions will result in burning and carboning of the oil on the
cylinders, pistons and rings.

Broken rings are especially damaging because their loose pieces with jagged ends are likely to
cut into the sides of the piston grooves. This causes land breakage which results in the complete
destruction of the piston assembly. Instead of reinstalling worn rings during engine overhaul, it is
always advisable to replace them. New rings have quick-seating surfaces which enable the rings
to control oil instantly, unlike rings which have been used in the past. Used rings, even those that
have only slightly worn will still have polished surfaces that will not seat-in properly and will
lead to excessive oil consumption.

Pistons Rings Stuck in Grooves

Obviously, oil cannot be controlled by piston rings which are stuck in their grooves, so every
effort should be made to prevent rings from becoming stuck. First, they should be installed with
sufficient side clearance to enable them to remain free while the engine is working under load at
normal operating temperatures. Second, every precaution should be taken at the time of assembly
to see that all parts of the engine are clean of any dirt particles which might cause the rings to
stick. Third, a good grade of oil should be used to lessen the possibility of carbon or varnish.
Fourth, the oil should be kept clean by regularly scheduled oil changes and proper care of the oil
filter. Fifth, every precaution should be taken to keep the engine from becoming overheated from
any cause.

Piston Slap
Some late model engines meeting the latest emission requirements have changed their piston
design. This can sometimes lead to a light "knock" at startup. In some cases this can increase oil
consumption levels.

Internal Gasket/Intake Breach

Newer engine designs sometimes implement a combination of composite materials and metals.
Gaskets and seals can sometimes breach of become stressed over time to differences in heat
expansion and contraction differences causing oil consumption levels to increase.

Spark Knock
Most new automobiles have knock sensors to adjust timing to reduce emissions as well as
increase engine power and performance. Spark knock is due to premature ignition of the fuel
during the combustion process. Preignition results in surges of pressure being forced upon the
piston. This disrupts the movement of the piston ring, resulting in a loss of ring seal on both the
top and bottom on the ring, and ultimately allowing for increased blowby and oil consumption
past the rings. This may also occur due to a faulty mass air flow sensor or throttle positioning
switch.

Lugging Engine
Lugging is running the engine at a lower RPM in a condition where a high RPM (more
power/torque) should be implemented. This causes more stress loading on the piston and can
lead to increases in engine oil consumption.

Inappropriate Operation of Overdrive

Operating the overdrive mode in conditions where it is not recommended will cause the engine
to consume oil for a variety of reasons. Such conditions include towing or stop-and-go driving in
city traffic. See also "Lugging Engine".

Fuel Dilution
If unburnt fuel is allowed to enter the lubrication system, the oil will become thinner and more
volatile. Both will result in higher oil consumption. Excess fuel can enter and mix with the oil
via a leaking fuel injector, fuel pump problem, restricted air intake or through excessive idling
Engine Venting: PCV, A/O Separators, Catch Cans
Original post by JZeezSTi

Ok... So I searched and searched, and researched and researched, and after many sleepless nights
and reading endless threads regarding the topic at hand, I’ve decided to try and start a thread that
would help everyone, including myself, who was interested in researching such topics, make the
right decisions.

So I’ll start out by adding what I've learned in my studies and what I'm still a little confused
about. If anyone more knowledgeable has anything helpful to add please do so and I’ll keep track
of all the information and Ill summarize it here in the first post so it will be easy to find for
anyone who is just passing through to get information.

What happens in the STi engine crankcase while driving?

In regards to the Subaru STi engine, as I understand it (and I may be wrong, but I would like to
share my thoughts and hopefully confirm them), the stout 2.5 liter boxer motor generates quite a
bit of pressure in the crankcase during high revs. This pressure can put a drag on the engine if not
ventilated efficiently. It can also force oil pass the piston rings, into the cylinders, burning off in
the combustion chamber and causing lean conditions, which can be bad for our motors and all
turbo charged motors as a whole. Under boost the high pressure of the Forced Induction can also
get by the piston rings in the opposite direction entering the Crankcase and adding to the high
pressure all ready in there from revving. This also can introduce contaminants into the oil as well
as un-burned fuel vapor.

What is Crankcase Ventilation and why is it necessary?

All of these ill effects can be avoided by sufficient venting of the engine by utilizing the constant
vacuum in the inlet tube of the intake track, to pull all the added pressure out of the crankcase
with a series of venting tubes, and check valves. Most stock motors have a sufficient venting
system that is designed for their specific performance levels, (i.e. the positive crankcase
ventilation system, aka PCV). There is an upside and a downside to this mechanism. The upside
to this is... It relieves the pressure in the crankcase, as well as re-circulates the un-burned fuel
vapor into the intake track and insures proper burning of the fuel. This helps keep harmful fumes
from polluting our environment and helps improve gas mileage. The downside is... Along with
pressure and fuel vapor coming out of the motor through the venting tubes, Oil Vapor also makes
its way into the intake track (aka Blow By). ESPECIALLY under high revs and high boost. This
is why most modded STi’s experience excessive blow by. Oil vapor can be harmful to the Turbo
and make a mess of things in the entire Intake system and intercooler, as well as causing lean
conditions by being burned in the engine. It can also cause an excess of carbon build up in the
engine (from burning the oil). These problems usually don’t occur in a stock STi because the
boost and performance levels are not high enough to cause any blow by.
How can you avoid “Blow By”?
The way to avoid "Blow By" is with proper venting and by using a device that will take the Oil
vapor out of suspension and maintain the flow of un-burned air/fuel vapor out of the crankcase
and into the Intake. An “Oil Catch Can” accomplishes this by passing the flow through a baffled
can that has inlet and outlet ports for the venting tubes and helps keep the oil in the can and out
of the intake. An Air/Oil separator accomplishes this the same way; however it has an additional
port on the bottom of the canister to allow oil to drain back into the engine.

Now... After covering the necessity of venting the engine, what blow by is, the conditions that
cause blow by and the way to help avoid blow by, the only other thing that needs to be covered is
what I am still a little confused about... The Plumbing!

Plumbing
The way the STi Engine’s stock venting system is plumbed, is obviously done for a reason. Who
are we to argue with years of engineering that the Subaru developers have put into designing the
engine? However, as I said before the Stock engine doesn’t really see much blow by due to its
lack of High Performance and Boost. The STi engine is obviously a high performance motor;
however the stock venting system is designed to be sufficient for its performance level. So when
boost and performance is taken up a notch and the need for a Catch Can or an A/O separator is
required to eliminate blow by, how should it be plumbed? Can simply intercepting the inlet tubes
with a catch can or an a/o separator be enough to get the job done? If so then why do aftermarket
companies have installation guides that indicate certain changes to the plumbing of the venting
system when installing their separators, and why do others simply instruct a simple interception
in the lines? What are the benefits and draw backs of each and are they performance level
dependent or a result of research and development, to prove one method better than another?
So which way is the best and what’s the reasoning behind why each company suggests a certain
way to install their venting tubes?
After I discovered the importance of an air/oil separator I began my research and after all was
said and done, I have a pretty decent understanding of everything except the plumbing. I have yet
to understand why each method is done and what the best way to plumb the venting tubes is for
each specific application. This is what drove me to start this thread. I didn’t want to just post a
simple thread with a question. I wanted to provide a source of information for everyone else who
is haunted by these questions and just the idea of A/O separators and Catch Cans in general.
I recently purchased and installed an Air/Oil separator made by Ixiz Concepts. I'm sure many of
you have heard of a few different ones on the market. I was on the fence between Crawford’s
and this one because of some reviews I’ve read and the overall quality they seemed to have. In
the end it really came down to cost. I found one that was brand new for significantly lower than
retail. So I picked it up and installed it. I really felt that I needed one since I just added a few
more power mods that put my power numbers significantly higher than stock, and I wanted
peace of mind and to be able to run said power safely for as long as I can. When it came time to
installing it I decided to use the Crawford Installation guide to plumb my tubes, despite the fact
that Ixiz’s installation guide is actually different from Crawford’s, I felt that Crawford’s seemed
a little more logical to me after my studies of the STi’s ventilation system. I've made a few
diagrams of the plumbing in detail to show you the difference and hopefully get some help to
figure out which way is the best and why. I used a diagram that I got off NASIOC as a starting
point and added to them for further clarification.

This is the complete unit you will receive from Subaru. The hose with the white connection goes
to the intake and the smaller longer tube runs beneath throttle body. The larger short hose on the
bottom goes to the engine block.
Figure 1: The Stock STi Engine PCV plumbing.
Figure 2,3: The Stock system under Vacuum and Boost.
Tuning
https://forums.nasioc.com/forums/showthread.php?p=36880803 by the suicidal eggroll

Also see Unabomber's Manifesto on the topic: http://forums.nasioc.com/forums/showthread.php?t=795608

Let me start off by saying that this is not a tuning guide, in any way, shape, or form. This is
simply an overview of what tuning means, what it does, the options available, and how they
work. If you're looking for a tuning guide, Bad Noodle has written a pretty comprehensive one
on NASIOC, check it out.

So how does a modern ECU-controlled vehicle work? The simple answer is it uses a lot of
sensors to measure what's going on, and it uses a pre-programmed set of lookup tables (generally
referred to as a "map") to see how to respond to the input from those sensors. Everything from
fueling, to timing, to boost is controlled by the ECU using these lookup tables. I don't want to
clutter this section with examples, so if you're interested look at the fuel system section for a
specific example of how these tables are actually used in practice.

"Tuning" on these cars USUALLY means using a piece of software to overwrite these lookup
tables with new values. If you want the ECU to inject more fuel at a certain RPM, you modify
the fueling table and lower the target AFR at that RPM, easy peasy.

There are two main choices for tuning, the Cobb AccessPort, and Open Source (RomRaider and
ECUFlash). I will not go into which choice is better or why, since it is all a matter of opinion and
I want to keep this guide as objective as possible. In general, both of these options do the exact
same thing. They both use the same factory ECU, the same lookup tables, and they modify and
flash them in the same way. A professional tune using AP will be exactly the same as a
professional tune using OS. They are just two different paths to the same destination.

That's not to say there aren't differences though. While the end result is the same regardless of
which path you choose, the path itself does differ. The AccessPort works by using your computer
to download pre-existing (aka off-the-shelf, or OTS) maps from any number of sources (Cobb
themselves, or other tuners). You also have the option of downloading the Access Tuner Race
software from Cobb and modifying the map yourself (DON'T DO THIS UNLESS YOU KNOW
WHAT YOU'RE DOING, YOU CAN BLOW UP YOUR ENGINE). Once you have the map
you want to use loaded onto your computer, you load it onto the AccessPort, then plug the
AccessPort into your car and flash the map onto the ECU. Once the map has been loaded, you
can then use the AccessPort to monitor sensors to see how the engine is running. You can make
"logs" of this sensor data, and then load the log onto your computer to look at a time series of
data during your drive, to see how the engine was running and responding.
Open source is very similar. Just like with the AccessPort, you download pre-existing OTS maps
from any number of sources (other users, other tuners, etc). You can also use the RomRaider or
ECUFlash programs (both freely available) to modify the map yourself (AGAIN, DON'T DO
THIS UNLESS YOU KNOW WHAT YOU'RE DOING). Once you have the map you want to
use loaded onto your computer, you plug the laptop itself into the car using a Tactrix cable, and
use ECUFlash to flash the map onto the ECU. Once the map has been loaded, you can use your
laptop to monitor the various sensors and make logs, just like with the AccessPort.

BOTH AP and OS allow you to run off-the-shelf maps. BOTH AP and OS allow you to get
professionally tuned. BOTH AP and OS allow you to tune the car yourself. A lot of people get
confused here...they think that AP only allows you to run off-the-shelf maps, or OS only allows
you to tune the car yourself. Neither could be farther from the truth.

The main difference between the two options is the device you use to flash and log. In the case of
the AccessPort, this device is a little handheld box with a screen and a few buttons. In the case of
Open Source, this device is your laptop with an OBDII to USB cable plugged into it. Beyond
that, there aren't many differences. The AccessPort has a little more functionality in regards to
base maps vs real time maps, real time tuning (making modifications while the car is running, so
you don't have to shut the car off, flash the ECU, and start it back up), etc. The last time I
checked, OpenSource had more lookup tables in the ECU defined and available for modification
than AP. It's a tradeoff, and a decision you'll have to make yourself. Both choices are incredibly
popular, and have almost a cult-like following. You will find no shortage of reviews, opinions,
and heated arguments on the topic of which is better.

Beyond the two big dogs (AP and OS), there are several other choices as well. Ecutek works
similarly (reflashes the OEM ECU). UTEC is a piggyback device that always stays connected to
the ECU. It allows the ECU to control things like idle and cruise, and then it "takes over" at high
RPM and high load. There is also a myriad of stand-alone ECUs, such as the Hydra, AEM, and
others, which completely replace the ECU and control everything themselves. I'm not going to
cover the different stand-alone options, because if your build is to the point where you need a
stand-alone, you won't be needing these guides to assist you.

Oh, and PLEASE don't turn this thread into an open source vs AP debate. There are hundreds of
those elsewhere on the forum, so take those kind of comments elsewhere. That's not what this
thread is about, this is about what the options are and how they work, not which one is better
than the other.
FAQ
Q: Does the AccessPort need to be plugged into the ECU permanently to function?
A: No. It of course needs to be plugged in in order to log data from the ECU as you drive, but
once you flash the map onto the ECU, you can unplug the AccessPort and toss it in your glove
box or closet without the ECU reverting back to stock calibration. Same goes for Open Source,
once the ECU has been flashed, you can unplug the laptop from the ECU and go on your merry.

Q: Will resetting the ECU cause it to revert back to stock calibration?

A: No. With either AP or OS, the ECU is completely reprogrammed with the new settings. There
is no copy of the stock calibration on the ECU for it to revert to, the only thing it knows is the
new map. Now with the AP, there is one "gotcha" here. When the AP is first installed and a map
is selected, the above is correct. It is completely reprogrammed, a reset will not bring you back to
stock. However, when you go to change maps on the AP, there are two ways to do it. You can
flash the new map as a base map, or as a real time map. Base maps overwrite the programming
in the ECU, so it is persistent through battery disconnects or resets. Real time maps are only
temporary, they will stay in effect until the battery is disconnected or the ECU is reset, at which
point the ECU will revert back to whatever base map was last used.

Q: Can an AccessPort be used on multiple vehicles?

A: No. You must "marry" the AP to your ECU when first setting it up. This pairs the ECU with
the AP, and the AP with the ECU. In order to use the AP on another ECU, it must first be
"divorced" from the previous ECU, which will flash that ECU back to stock in the process. If
you ever buy a used AP, make 100% sure that it has been divorced from the previous owner's
ECU, otherwise you won't be able to use it on yours.

Q: Can tuning be used to disable CELs?

A: Yes, almost any code can be disabled with AP or OS, including the resulting limp mode
caused by several of the more "serious" codes. Just because they can be disabled, doesn't mean
they shouldn't be though. Only codes that you are 100% certain do not matter should be disabled
(such as the TGV codes if you've deleted your TGVs). Other codes, even the "annoying" ones,
can be critical in debugging problems you might run into down the road.
Q: Can you go a little deeper into the benefits of me tuning my own car with AP/OS vs.
having it professionally tuned?

Also, I understand that tuning the ECU can give you power gains, but what about
aftermarket ECUs? Are there any ECUs that are 'better' than the stock ECU, or does it
100% depend on the tune?

A: The professionals have access to the same tuning software you have (they might have a little
bit more power than you do in the AP domain, but in OS it's the same), so it really just comes
down to experience. While each car is different, the professionals know what generally does and
does not work, and can get you to maximum power much faster than you can on your own.

That said, it's not their car, most of the time they're trying to get you out of the door and will
often take shortcuts with the tune (not every tuner, but a lot of them). If the engine blows, 99% of
the time they won't pay for it. Also, it's impossible to get a complete tune in the few hours your
car will spend with them. You need to accept the fact that either you'll be going back to them for
touch ups probably every few months for the first year, or you'll have intermittent problems like
cold weather overboosting, bad AFR in the summer, etc. The closer to stock you are, the less of a
concern this is, but when you start getting into big MAF intakes, aftermarket turbos, EWG, E85,
etc., you're going to be going back to them quite a bit (unless you live in California where the
weather never changes). This isn't their fault, they just aren't able to tune the car for anything
other than the weather conditions at which they do the tune. They can guess how things will
change as the weather changes, but if you're far from stock they won't necessarily be right.

It's just a tradeoff between having to learn how to tune and taking days/weeks/months to dial
your car in yourself, versus getting in and out of the door in a few hours, having to go back a
couple of times for touch ups as the weather changes, and putting 100% faith in the ability of
your tuner.
Q: Now my only question is does accessport allow for several interchancheable maps
during driving such as better mpg, racing, winter, summer etc.

My understanding was you could have several maps programmed that you can change
while driving with the push of a button on the screen. So instead of running to get retuned
for summer, winter, etc. you can just pick the maps that fit the conditions and revert back
when conditions change as long as your programmed without visiting the tuner shop.

Or... Does visiting the shop for several retunes eventually allow for the tuner to get to
where one map will be smooth regardless of conditions?

A: Yes, both AP and OS allow you to hold multiple tunes on the flashing device, and swap them
out at will. You can't do it while driving though, you would need to pull over, shut off the car,
connect the green test connectors, flash the ECU, disconnect the test connectors, start the car
back up, then start driving again.

As you pointed out at the end of your post, there is no need to have separate maps for winter,
summer, etc. though. If your map has issues in winter that it doesn't have in summer, you need to
go back to the tuner to have the low temperature compensation tables adjusted. Unless you're
actively tuning the car or switching between fuels, you should be reflashing the ECU very, very
rarely.
Engine Management FAQ
Why do I need engine management? Consider your stock engine management for just a
moment. Your stock engine control unit (ECU) is a very complex piece of circuitry that
calculates hundreds of variables every second. All of these variables rely on inputs within a + or
- range. When you modify your vehicle, these values change. As long as the changes are within
the values the ECU expects to receive, your engine runs fine. Once the values are exceeded, the
ECU is programmed to compensate to return the values to normal levels.

This is a layman's explanation of how your stock ECU can actually work against you when
modifying your vehicle. This also explains why modifications can feel great once they are bolted
on but the butt dyno results seem to fade over time. This is due to ECU compensation.

What is the first step in finding what engine management I need? Finding a tuner. The Tuner
FAQ will help with the general rules of finding a good tuner. Remember, it's always better to
have a custom tune vs. a plug and play or "staged" map. Always defer to the tuner's advice as to
what to choose as ultimately he will be the one to provide custom support. Discuss your goals
and budget and your tuner should set you on the right path. If you are a "plug and play" kind of
person, review the options below and decide for yourself along with input from locals in your
regional forum and the car parts review forum.

What will engine management do for me? Generally speaking, engine management optimizes
several engine functions to create more horsepower and efficiency. The stock ECU is designed to
ensure your car runs fine and monitors the engine's output parameters. Utilizing an aftermarket
engine management solution takes this to the next level.

Often times, car manufacturers will program the stock ECU with a known amount of "play".
This amount of play allows the manufacturer an extra level of safety and/or the ability to utilize
this at a later date so they can have an increase in HP in later years. An example of this is where
a 2001 car has 200HP and a 2002 car has 215HP. Some manufacturers couple this with
additional parts to get increased HP levels, but this should give you some idea of the concept.
From a marketing and sales aspect, this ensures that the public will continually be interested in
the new year models, even if the body style does not change. While this FAQ is not about the
science and art of auto sales, this may give you an insight as to why the stock ECU is not 100%
perfectly tuned from the factory.

What are the least talked about benefits of engine management? Engine management
solutions to one degree or another can reduce or eliminate black tailpipes, improve driveability
throughout the powerband, eliminate the open/closed loop delay in 04+ Subarus, and increase
MPG.
What Subarus NEED engine management? While every Subaru will benefit from engine
management, the 04+ turbo models (except STI) "require" it. The reason for this is the EPA
mandated greater restrictions on ECUs for 2004+ and allow manufacturers to exclude one model.
Subaru chose the STI as the exclusion, so the restrictions are not on the STI. The restrictions
have to do with the open and closed loop fueling. To put the restrictions in a nutshell without
being technical, there is a delay between open and closed loop fueling that can allow your 04+
Subaru to run lean during this crossover point in how your car gets fuel. Lean is dangerous as it
produces detonation which is the #1 factor is blown pistons. In stock configuration, 04+ turbo
models are fine, but modifications such as exhaust work or higher really necessitate the use of
engine management as those mods cause the fueling issue to rear it's head. Yes, you can run
certain mods for a short time until you get engine management, but you should never kid
yourself on 04+ turbo models that bolt on modifications are fine without engine management.

What about manual/electronic boost controllers or air fuel controllers?

A manual/electronic boost controller or air fuel controller really isn't engine management. MBCs
are fine when used correctly on 02/03 WRX and 04+ STI by a judicious user. The issue gets
further clouded on 04+ turbo models less STI, due to the open/closed loop delay. One can lump
in the mix air/fuel controllers as well for the same reason. Usually those two devices are used by
either cheap skates, n00bs, or advanced users who combine them with other forms of real engine
management.

What types of engine management solutions are available? They fall into one of these general
categories:

a. Reflashed ECU
b. Custom Tuned ECU
c. Piggyback Engine Management
d. Stand Alone Engine Management
e. Open Source Engine Management

Reflashed off the shelf ECU:

This is your stock ECU that has had the programming modified. This form of engine
management is best suited for people who:

a. Have a "set it and forget it" attitude towards engine management

b. Live in remote areas and do not have easy access to a tuner
c. Do not want to learn or are uneasy doing their own tuning
d. Want to modify their vehicle to a certain level and quit or add parts very infrequently
It is important to note that while reflashed ECUs are considered a static engine management
option, they can be custom tuned by the end user by utilizing additional add-ons from the
manufacturer or via a custom tuned reflash by an authorized tuner.

In addition, also consider that EcuTek tuners might have reflashes for your exact equipment set-
up based on their prior custom tunes. This means that someone who has a full TBE might find
the Cobb AccessPort a better plug and play solution, while someone with a full TBE, uppipe,
headers, & a lightweight pulley may find an EcuTek reflash a better plug and play solution if
they can find an EcuTek tuner with that exact map.

Recently for AccessPORT users, "custom maps" have become vogue. These come in the form of
more specific maps for specific mods. Let say you have a stage 2 car with headers, TGV deletes,
and a pulley. You email this information to a tuner and they will create a custom map for those
mods to use. Some will send you a map and logging software, you load the map and data log,
then the tuner sends you a final customized tune. While not as perfect as an in-person/on-dyno
custom tune, it's a great resource for those in areas without tuners. Companies that do this are
PDX Tuning, Perrin, and Clark Turner.

Examples of reflashed ECUs are the Cobb AccessPort and EcuTek (including Prodrive's PPP &
Vishnu's reflash).

Reflashed custom tuned ECU:

This is the next evolution to a reflashed ECU. This allows either the end user or a professional
tuner to custom tune your vehicle to your specific modifications, wants and desires, type of
gasoline used, and geographic area. This form of engine management is best suited for people
who:
a. Will probably modify their vehicle frequently and require additional tuning
b. People with the ability to do their own tuning
c. People that live close to a tuner
d. People that want to get the maximum power and safety out of their car

Examples of Custom Tuned ECUs are Cobb Tuning's StreetTUNER for end user tuning, Cobb
Tuning's ProTuner for professional custom tunes, EcuTek custom tune via an EcuTek tuner, and
EcuTek's DeltaDash Live User Tuning, an end user tuning solution.
Piggyback Engine Management:

This is an engine management option that works in conjunction with your stock ECU.
Depending on the manufacturer, this solution works by the piggyback unit controlling some
engine management functions and the stock ECU controlling others. This form of engine
management is best suited for people who:
a. Will probably modify their vehicle frequently and require additional tuning
b. People with the ability to do their own tuning
c. People that live close to a tuner
d. People that want to get the maximum power and safety out of their car

It is important to mention that most piggyback units come with base maps. These base maps
work very similar in function to a reflashed ECU whereas you can run the base map and be
100% fine, or when the day comes for someone to tune their own car or have it professionally
tuned, they may do so.

Examples of Piggyback Engine Management are Unichip, Xede, UTEC, and others.

Stand Alone Engine Management:

This is an engine management solution that totally replaces the stock ECU and controls 100% of
the engine's functions. This form of engine management is best suited for people who:
a. Will probably modify their vehicle frequently and require additional tuning
b. People with the ability to do their own tuning
c. People that live close to a tuner
d. People that want to get the maximum power and safety out of their car

This form of engine management is generally reserved for more advanced users and people
going for really high levels of performance.

It is important to mention that most stand alone systems do come with base maps. Unlike the
base maps that come on reflashed ECUs, these base maps are meant for your vehicle to run for a
short period of time and are not meant to be used as a permanent solution as is the case with the
other base maps as described above. Consider these base maps as merely as short term option
until end user or a professional tuning.

Examples of stand alone engine management are MoTeC, Hydra, AEM, and others.
Open Source Engine Management:
This can be the cheapest source of engine management available. In essence, you use a laptop,
software, and a cable to reflash your stock ECU. Can be used to flash "staged maps" as a set it
and forget it option or as a dynamic tuning tool either through tuners or by the end user. This
form of engine management is best suited for people who:

Group 1:

A. Already have a good degree of tuning knowledge and understand the logic of the
factory ECU (or have a strong desire to learn both).
B. Want to constantly tweak and experiment with their own tune. They actually enjoy the
process.
C. May be changing/upgrading mods frequently.

Group 2:

A. Looking to get a custom tune from a professional but cannot afford or do not want to
spend the money on license fees and/or hardware costs required of Cobb, Ecutek, etc.
That is, they want their car to have a custom tune from a pro at the least cost. More and
more professional shops are offering open source tunes and they can be just as capable as
tunes from the commercial software.
B. Want an inexpensive (ex. XPT) or free (created by another user) OTS map. Might not
go with EM otherwise because they feel it is too expensive. They are willing to learn the
basics of logging with RomRaider to make sure the tune doesn't have any issues with
their car.

Those people where Open Source Engine Management would be a BAD choice (assuming they
are doing it themselves and not a professional tuner):

1. Want the easiest to use, troubleshoot, and closest thing to "set and forget" EM solution.
2. Are computer illiterate.

Examples of Open Source Engine Management are RomRaider, formerly known as Enginuity,
EcuFlash, and others.
Can the dealer detect my reflashed ECU (AccessPort/EcuTek/Open Source)?

YES the checksum of the ECU changes.

YES the dealer can easily read the checksum.

NO the dealer has nothing to compare the checksum against there are many revisions of the
WRX ecu they all have different checksums. If the dealer had someway of putting this checksum
into a database he COULD verify that the code had been modified but at this time he doesn't.

The 05+ ECU and some of the 04s have the VIN in the ECU code. The current versions of
reflashed ECUs only change the tables so the VIN will report when queried.

Bottom line: If you don't want modifications to be detected, don't modify the car.

What are some specific types of engine management While this FAQ does not go into
specifics for every type of engine management, this thread covers many types. This link offers
several comparisons of different system features as well. There are some not covered in these
threads though and may be considered as well.

Generally speaking, what engine management option will give me the most power? Custom
tuned engine management solutions will always give you more power. Every off the shelf engine
management solution has a built in safety factor. This depends on the manufacturer. This is
because their "Stage 1" or "VF-30" map has to safely make power for someone living in
Phoenix's heat and 91 octane, to Denver's high altitude, to Boston's cold and 93 octane. The
gasoline and geographic variances can leave horsepower on the table.

More specifically, what engine management option will give me the most power? This is one
question without a correct answer. Let's say that you research a very comprehensive stand alone
engine management system such as a MoTeC unit and decide that it's the "best" for your car. At
the end of the day, it's about what the tuner is most comfortable with. Some tuners may be able
to extract better results from a "lesser" system simply because they understand the interface
better.

For someone interested in tuning their own vehicle, they should match their tuning skills with an
engine management solution that they are able to understand and use correctly. For someone
interested in professional tuning, they should consult with their tuner for their recommendations.
Both of these actions will ensure a good tune with a minimum amount of rework, guessing, and
trial and error. While most tuners are capable of learning new or more advanced engine
management systems, consider the benefits of an "older/worse system" your tuner is familiar
with vs. your tuner learning a "newer/better system" at $200/hour plus possible dyno time.
I have a reflashed ECU and am not seeing their advertised HP, why? First off, have you met
their criteria EXACTLY? The #1 cause of low HP with reflashed ECU owners is their lack of
meeting the manufacturer's requirements. If they require a full turbo back exhaust and you only
have a downpipe and a cat back exhaust, 100% of the blame is on the end user. As well, if they
require a full exhaust and you have a full exhaust, uppipe, headers, and a bigger top mount
intercooler, this can cause problems as well.

Also realize that HP figures vary. You cannot compare (for example) a manufacturer's Mustang
Dyno HP figures to your local DynoJet HP figures. Even comparing identical dynos to each
other is futile as dyno software set-up, altitude, temperature, humidity, and other factors do not
ensure an equal result.

In addition, realize that a reflashed ECU still utilizes many of the stock ECU's learning functions.
This means that in a perfect world, you will see the advertised HP from your reflashed ECU.
This may also mean that on the day of dyno testing, your octane, the temperature, humidity, and
many other factors are considered by the reflashed ECU when determining total power output. If
the advertised numbers aren't there on dyno day, it doesn't necessarily mean there is a problem,
but rather, your ECU is protecting your engine from low octane, high temperature, high
humidity, or other factors.

How much is a custom tune by a professional? Expect to pay $100-$150 per hour for the dyno
time. Your tuner's fee depends on their level of experience and pricing. Tuners generally charge
$100-200 per hour for their time. The amount of total tuning time depends on the tuner and the
amount of time you wish for them to tune your vehicle. Most tuners can get your vehicle within
90-95% of its maximum power within 1-2 hours.

How do I go about learning tuning? The quickest source of tuning information lies with local
DIY tuning personnel. Get to know locals via your local forum. Ask for advice, tips, and see
some real world tuning. Also realize that tuning can become a very personal subject and there is
more than one way to get to XXX power level. An excellent example of this is the U.S. vs
Japanese tuning philosophy. U.S. tuners tend to run more boost and less timing and Japanese
tuners tend to run less boost and more timing. Both can create safe power using these methods;
it's just using two different methods. While this is an over simplification and a generalization,
this should give you the wisdom that there are different methodologies that people have
successfully used.
Source of tuning information:

a. This thread and this thread though UTEC specific, contain tuning information that may
serve useful for other engine management solutions.
b. For personnel interested in advanced tuning theory, or for professionals, consider the
tuning articles on www.sae.org as a source of information.
c. How to Tune and Modify Engine Managements Systems by Jeff Hartman is a book
dedicated to tuning that may prove useful.
d. EFI University offers classes nationwide on tuning fundamentals and principles.
e. Innovate Motorsports' website not only sells tuning products, but contains a wealth of
tuning information.
f. UTEC Quick Tuning Guide While UTEC specific, this contains some good tuning
information. (Requires registration on wrxhackers.com to view)
g. RomRaider FAQ

What are some helpful tools to use if I want to monitor my engine's functions? Some engine
management solutions come bundled with engine monitoring software, which is an excellent
way to see hard data on your tuning results. If this software does not come bundled with your
engine management, consider one of the following solutions:
Simple data logger
TARI DL1 (Diagnostics Level 1) Data Logger
DeltaDash
LogMapper
MPS UTECLogger
Auto-Logger
Scanmaster 3

What gauges/tools are important if I wish to tune my own vehicle? These are considered the
"must have" tuning items:

a. Wideband O2 sensor/gauge
b. EGT gauge
c. Boost gauge
What are some other useful tuning tools?
a. Knock detection hardware
b. SMC 4 channel EGT
c. Useful tuning tools from Cobb Tuning

Editors Note

My special thanks to Trey Cobb and Daniel Ponze for their assistance in the formulation of this
FAQ.

This post was created because I wasn't able to find a good engine management FAQ. I came up
with the text based on LOTS of searching here. It was also created to be intentionally brand
neutral so that it serves as a stepping stone for further research. Upon reading this you should
have an idea of what type of engine management solution best suits your needs. The
manufacturer is up to you.

If you find an error in this FAQ, please PM me with factual details and I will update this post.
Responses such as, "I have XXX's engine management and it's great!" or "XXX's engine
management runs too rich" are not appreciated here, that is what the Car Parts Review Forum is
for.
Motor Mounts, Bushings etc.
Original post by Scooby921

Any increase in stiffness of an engine mount is going to give you better

performance from the car, to a certain degree. With less energy lost due to motion
of the motor you are able to transfer more of it to the wheels. Being that a boxer
motor doesn't have much of a roll moment, they don't provide nearly the same
benefit as they do on a V motor. Additionally, the technology of subaru engine
mounts is about 100 years old. It doesn't get much simpler than a piece of rubber
mounted in a compression/shear orientation. The group N mounts are stiffer
because they add a steel plate in the middle of the rubber which reduces deflection.
Not rocket science. Its not anything close to the electronically controlled,
switchable rate, active mounts I design for Honda. (yup, I design engine mounts
and suspension bushings for a living)

The real benefit you are going to get from them on a Subaru is less overall
movement of the powertrain. It'll move around less when you hit a bump or rip
around a corner. Things you want in a race car and don't necessarily need in a street
car. It really just makes the vehicle dynamics and weight transfer calculations that
much easier for a group N rally car. The trade-off is an increase in noise/vibration
transmission into the vehicle cabin. While you may not notice a change through
most of the day-to-day driving that you do, that one vibration that makes it through
a stiffer mount could cause your seat to vibrate, you glove box to rattle, or some
other piece of trim to annoy you to no end.

For what you are planning for your car I don't think they are really that necessary.
They really won't hurt much, but I think you'd be better off doing a group N trans
mount and an STi or group N pitch stop to reduce engine/drive shaft wind up under
load. That'll do much more for performance than the engine mounts will. Just
realize that a stiffer trans mount is going to transmit more gear whine into the cabin
and a stiffer pitch stop is going to transmit more engine and trans noises into your
firewall. Again, the level at which this happens is different for everyone. Some
people don't care about noise, others simply don't notice it. What works and makes
no change for one could just be the end of the world for the next.

Personally, I have a group N trans mount on my WRX. I also have a lot of goodies
from KB and TiC to stiffen up the whole chassis for autocross. I love the
performance of my car, but the amount of noise that I hear while driving down the
road can get annoying. Figure out what you want. Ultimate performance and
ultimate ride quality are mutually exclusive.
E85 fuel FAQ
By hotrod - originalPost

Part 1
Introduction to my experiment with E85 circa 2004

This is a summary of past posts from various forum discussions

I have been running a high ethanol fuel blend in my WRX for nearly 2 years now, with no
problems traceable to the fuel. The car really likes the high alcohol blend, with an obvious
improvement in performance comparable to high octane so called "race fuels".

The Stock WRX ECU can handle up to approximately 33% E-85, by volume before you get a
check engine light for "too lean" due to the fuel trims being pushed to the max rich settings. Add
just a bit of pump premium and the CEL will clear.

I recently shifted to STi 550 injectors without any special changes in the engine management.
Due to the cars larger injectors the car now prefers a higher blend. I am now running about 95%
E-85, but have run the car on blends as low as about 30% E-85 without any serious problems. At
lower blends the car has some surging due to the over size injectors at higher throttle settings, but
can be easily driven as a daily driver if you stay out of the >60% throttle position transition to
open loop.

The one problem is that currently E-85 sites are a little difficult to find, but the more of it we use
the more available it will become.

The folks backing E-85 production are pushing as hard as they can to get stations to make it
available, but its a slow process. They need customers to start asking station owners if they plan
on carrying it to motivate stations to add a pump/tank.
There are currently dozens of FFV's out there that are designed to run on the stuff if folks can
find a place to buy it. Simple way to drastically reduce oil demand as well, and put money in our
economy instead of some other countries coffers.

I currently drive 20 miles each way to get to the nearest E-85 stations, but it is still a good deal as
I am paying $1.59.9 / gal for the E-85, vs about $2.05.9 / gal for premium. ( some E-85 vendors
charge at a premium fuel rate of about $2.00/gal)

With the added octane of the E-85 you can actually splash blend it with mid grade gasoline with
out problems.
When I suspect it will be difficult to locate E-85 from the pump I just make a point of topping off
the tank before it drops below 3/4 full. This keeps the ethanol blend up to a high enough level to
avoid any drivability issues with my oversize injectors.

Is the WRX warrantied to run on E85 ?

NO it is not, but my experience shows its not a major issue either.

I think it is important to note that they don't recommend greater than 10% ethanol, ie they
warranty the car will run fine with up to 10% ethanol but greater than that your on your own. But
they do not say you should avoid higher blends of ethanol and other tests have shown modern
cars can run on upto about 30% blends with no problem.

They do specifically mention that methanol is not to be used over 5% concentration, and that is
due to corrosion issues with methanol, which is Much Much more prone to corrosion than
ethanol.

So far I have seen no "real" problems with the E-85. There are two minor CEL issues.

1 If your running over 35% on a stock ECU and fuel injectors you will push the fuel trims to full
rich and get a "too lean" CEL.

(edit 10/02/2005 --- emission CEL has not re-appeared in quite a long time. I am beginning to
suspect it was due to a loose vacuum hose resulting from the injector upgrade that got fixed as
I was working on other things )

As mentioned above any "damage" should be easily remedied, ie replacing a hose, or some O
rings, possibly changing to a different fuel pump. It is very difficult to predict long term
corrosion, or materials compatibility so I've decided to bite the bullet and be the test dummy and
see what if anything breaks.

Based on my tests, the short term conclusion is you can run concentrations of >10% fuel ethanol
for periods in execess of 1 year with no detectable damage. We'll just have to see how things go
in another year or so.

Hopefully this winter I'll get the time to pull the fuel pump and look over the interior of the fuel
tank to see how its doing.
What is the history of large scale conversions to high ethanol fuels

When Brazil began making a wholesale conversion to high ethanol fuels back in the late 70's
following the energy crisis, they made several studies on the ability of normal cars to run ethanol
blends. They found that the cars of that period could run up to about 22% blends on the stock
system with no problems, which is why they settled on a 20% blend as one of the fuels available.
The issue was one of control authority of the ECU to compensate for the leaner mixture. Some
could handle more than others. As I stated, it appears the stock ECU on the 2002 WRX is good
up till about a 33% - 35% blend.

During the 70's and 80's when oxygenated fuels and "gasahol" first saw wide use here in the U.S.
there WERE fuel component compatibility problems. My 1969 VW fuel lines really didn't like
the ethanol and began to leak like a sieve, some carburetor needle valves softened, some
carburetor floats would soak up the ethanol and get too heavy to function as a float. There were
lots of problems with clogged fuel filters on cars that had been running on gasoline only for
decades and had lots of varnish build up in the fuel system. The ethanol in gasohol was a very
efficient fuel system cleaner and all that crud got carried to the fuel filters. Once the fuel filters
were replaced those problems disappeared.

At that time All the auto manufactures moved to ethanol compatible fuel line components, ie. o
rings, rubber hose etc. They warranty that they are good to 10% but my experience shows they
are satisfactory to much higher concentrations. The VW showed its compatibility problem in a
matter of months after we went to ethanol blended oxygenated fuel here in Denver. Engineers
typically don't solve a compatibility problem by making the new component "sorta compatible"
they change compounds to materials that are not effected by the chemical in question.

The Denver area has been using ethanol oxygenated fuels (ranging from 5% - 10%)
concentration for over 30 years. Every modern car works just fine with these low ethanol
blended fuels. Rubber hoses and O rings last for the life of the car.

The electrical conductivity issue is not significant in the case of ethanol. It is detectable with a
dialectic constant tester. It DOES become a significant issue with methanol blends which is why
methanol blended fuel is so aggressively corrosive. The main issue with methanol is it
aggressively attacks certain metals like magnesium and zinc. One of the reasons everyone is
looking at ethanol is the 30+ years of successful use of high ethanol fuel blends in Brazil and low
ethanol blends here in the U.S. with essentially zero problems after they changed fuel line and O
ring and seal materials in the fuel system.
Dielectric Constants:

• Gasoline 2.2
• Ethanol 24
• Methanol 33.6
• Water 48 - 88

Keep in mind that absolutely pure water is a good enough insulator it is used to cool electronic
components. It does not become an effective conductor until is dissolves minerals that act as
charge carriers (electrolytes).

Will my O2 sensor work with E85 and high ethanol blends?

The O2 sensor is not an issue, all it cares about is if your at stoich combustion at low throttle
settings, it doesn't much care how you get there, so no need to change it.

Air-fuel ration: http://en.wikipedia.org/wiki/Air-fuel_ratio

Oxygen Sensor: http://en.wikipedia.org/wiki/Oxygen_sensor

What about the evaporative emissions system in my car?

The higher vapor pressure of ethanol gasoline blends is not ideal for the evaporative emissions
system and the vapor recovery canister. For full emissions compliance these will need to be
modified. Currently there are no kits available to upgrade this part of the emissions system.

As I mentioned above the only -- real -- problem so far is simply some nusince CELs.
"too lean" if your over 30%+ on the stock injectors

If your emissions test requires no CELs present, you simply need to go to a lower ethanol fuel
blend for a while to allow that nuisance CEL to clear, and the dash warning light to go off.
What would be needed for a complete conversion kit to run 100% ethanol fuel

For full conversion to alcohol fuels the change list typically looks something like:

1. Go to an compatable fuel pump (walbro 255 l/h pumps seem to work well with E85)
2. Replace fuel lines with alcohol compatible lines. (not needed on WRX based on current
experience)
3. Possibly replace filter (not needed on WRX based on current experience).
4. Replace injectors with large enough ones to feed proper fueling. (+30% flow over stock)
5. Replace injector/FPR o-rings with compatible o-rings... viton maybe? (not needed on
WRX based on current experience)
6. Add a spark/flash suppressor to the fuel tank inlet tube. (does not appear to be a
significant hazard based on current experimenter experience)
7. Ensure the fuel tank material is completely compatible with alcohol. (not needed on
WRX based on current experience)
8. Reflash or other engine management option for ideal tune to use E85's capability
(optional at low to moderate boost)

With ethanol on the WRX, you only need (according to my current experience base) is numbers
4, 8.

(edited to remove comments about emissions CEL which proved to be due to other causes and
has never returned)

What about fuel system corrosion?

Corrosion does not appear to be an issue with modern OBDII cars. They are all certified by the
manufactures to be safe to use on 10% ethanol fuel blends, and industry insiders say they are safe
for much higher percentages. You don't install components that are "sorta safe" with a chemical,
you put in a fuel hose etc. that is ethanol safe for concentrations well above what you expect to
use. Not to mention that folks have been talking for years about raising the ethanol level to 20%
or more.

Occasional use would be no problem at all based on my experience. I have never cut open the
fuel filter (still have the OEM filter in place) I'll open it when I replace it. I want to put enough
time on it to have conclusive evidence if there are problems.
(edit circa 2007 I cut open a fuel line and it looked like new, not changes or damage on stock
rubber fuel lines in 2002 WRX)
The engine is not an issue with either, WI using a water alcohol mix or straight alcohol injection.
In those systems alcohol and water are not used in significant quantity or for long duration. The
Buick GN folks and lots of folks in the DSM crowd have done it for literally decades with no
problems for the engine.

Many years ago there were studies that indicated engines that ran on alcohol ALONE as a fuel,
had issues with lubrication and valve seat wear. Keep in mind, those studies were done a long
time ago, when engine oils were much less sophisticated than they are now, and some engine
manufactures in the 1940's,1950' and 1960's made stupid engineering decisions and did not use
hard valve seat inserts like stellite in the cylinder heads. This resulted in valve seat recession
problems if you did not have lead additives in the fuel to protect the valve seats.

What about oil contamination?

Modern lubricants, especially the synthetic oils are much much different than the oils used
during those studies, and modern engines run at higher temperatures today which will quickly
boil any traces of alcohol out of the oil.
In cold weather I run an 180 deg thermostat to assist quick warmup I have left it in during this
past summer and so far the engine has no heating issues with the 180 thermostat on the E85 fuel.
(edit experience shows modern synthetic oils work very well with E85)

Is Ethanol less corrosive than Methanol?

Methanol is much more corrosive than ethanol. It attacks certain soft metals that are not much
used in modern fuel systems. Years ago, the carburetors were made of un-anodized aluminum
and if methanol fuel was used, you had major problems with electrolytic corrosion between the
aluminum and copper components used in the fuel system, since they were in continous contact.

That sort of corrosion only occurs when you have a current path between the dissimilar metals
AND, a conductive path through the fluid in the system.

In Brazil where they have run high ethanol fuels since 1939, they found that to convert older cars
designed for gasoline, long before ethanol blends were common, needed several changes to
convert the cars over. This led to changes in valve materials, piston rings choices, nickel plating
of the fuel tanks etc.

Modern cars in the U.S. are designed for use with ethanol up to 10% concentration in the fuel.
That has led to several changes in component materials over the last 30 years that the U.S. has
used ethanol enhanced fuels. All modern fuel lines and such are designed with the expectation
that some ethanol will be in the fuel.

My experiment is an intentional effort to push the envelope and see what happens. Over the last
2 years I have run high ethanol fuels ( normal pump fuel here in Colorado contains up to 10%
ethanol anyway) for months at a time.

At mixtures below 33% by volume of alcohol ( about 39% E-85 by volume) I had absolutely no
problems of any kind. At higher mixtures > 33% alcohol, I got a nuisance CEL for too lean
which could be eliminated by added gasoline to the mix or as I have recently increased the
injector size. (Increasing fuel pressure would also increase the effective size of the injectors).

The only issue with near 100% E-85, is the car starts a little harder in cold weather. When it gets
very cold I drop the mixture to about 80% or so to solve that by adding about 2 gallons of
gasoline to the tank of E-85.

I have done no oil analysis at this point, walbro 255l/h fuel pump and 550 injectors show no
signs of problems to date.

Larry - (last updated 6/10/08)

Part 2
Where can I buy E85?

Distribution is currently a problem, it is most available in the mid-western states, but there is a
concerted effort to get an ethanol fueling infrastructure built, it just takes time, and recognition
by the gasoline vendors that there is a market. Marathon oil has already added E85 to their
formal product line so things are moving in that direction.

Current high fuel prices can only help that process, as will lots of people calling up gas stations
and asking them when they plan on pumping E85 for FFV's.

E85 already contains inhibitors and such to minimize corrosion, and most of the problems with
ethanol usage were solved 20 years ago when most of the country moved to ethanol addition to
gasoline following the oil crisis in 1973 and high oil prices that continued up until 1981.
Where can I buy E85?

For information on distribution points check out:

http://afdcmap2.nrel.gov/locator/FindPane.asp

How about fuel mixtures for E85 what is needed ?

The one interesting thing to note is that the stock pig rich mixtures at WOT on the stock ECU of
the WRX are very close to max power mixtures on E85 blends. If you play with the blending
ratio to find your magic number you can take advantage of this by giving the car a fuel blend that
is max power rich at stock ECU WOT mixtures. *** just take your time and work up the blend
slowly ***

After a few folks have dyno runs on various blends we can probably pick out a "best starting
point" blend that will get you very close. We could also work out the other supporting mods to
make best use of E85 for minimum cost.

It basically takes about a +30% increase in fuel flow to get the same equivalent mixture with E85
that you had with gasoline.

For example an adjustable fuel pressure regulator can increase your effective injector flow to
partially compensate for the fuel changes.

As mentioned above, lean is still a bad thing, and with alcohol fuels due to the higher power
levels a serious leanout is more lethal. With conservative mixtures alcohol actually burns so cool
that on E85 FFV there is some concern about getting the cat to light off.

If you sneak up on a fuel blend slowly you should be just fine. I would venture to say for the near
stock folks, up to 30% is probably safe as long as you give the ECU a few miles to make the
major part of the fuel trim accomodation before you start beating on the car. You'll know when
the ECU starts to get happy as the midrange torque is noticably better.

I also strongly suspect that a well tuned car with no cats can probably pass the emissions sniff
test on high E85 blends.
How does ignition timing change on E85 ?

Timing on ethanol blends will not change very much. MBT timing for both gasoline and E85 are
very nearly the same at light to moderate engine loads. At high engine load the E85 will want
just slightly more advance. The big difference will be fuel/air mixture. The E85 will give
improved torque with much richer mixtures than gasoline. Both gasoline and E85 will give best
thermal effeciency at about 15% rich of stoich, so the equivalent of 12.78:1 on gasoline would be
about 8.5:1 on E85, but E85 will continue to give better torque numbers up to about +40% rich
of stoich or 7:1 mixtures, so on a utec you would want to richen up your WOT high load cells
and add a tweak of timing to get the most out of E85 from what I've read.

“Are you sure you don't mean that E85 will allow more advance?”
Just passing on what I've found in the various sources. Logically you are correct, but one source
says simply that MBT timing is the same for E85 and gasoline, and another report says at low
loads the E85 and gasoline like the same MBT timing but at high loads MBT timing for the E85
is slightly more advance.

I suspect this is due to them not running ideal max power mixtures but cannot confirm it. Burn
speed for E85 changes quite a bit with mixture, so if they were just a little bit lean or rich of ideal
the burn rate would be lower.

Lots and lots of variables not well covered in some of the sources and in general they are
focusing on emissions issues not max power torque so that would incline them to use less than
best power timing advance. In a couple of the reports they also had limited control authority over
timing and may not have explored the extremes very thoroughly.

What data is available on combustion speed for E85?

Just about a year ago (2003), I was finding conflicting information on alcohol fuel burn speeds
and contacted The National Renewable Energy Lab here in Denver, they did a search and could
only find a couple of references on it, one that showed ethanol fuels burn faster, so the literature
is not very rich in data on the subject. Most reports are not entirely applicable to our needs like
tests on lawnmower engines and alcohol fuels have some significant limitations

Still digging for info but that is my best information at the present time.
The report that mentioned the slight increased advance requirement on heavy load E85 fuels was:

"Final Report Control of Exhaust Emissions from Small Engines Using E-10 and E-85 Fuels"

http://www.michigan.gov/cis/0,1607,7...0064--,00.html
http://www.michigan.gov/documents/CI...E-_87915_7.pdf

On page 12 it says:

"Tests were conducted to assess the impact of MBT ignition timing on fuel economy and
exhaust emissions. The spark timing was varied until the least advanced timing was
achieved for the maximum torque for a given setting of the engine throttle. The A/F ratio
was adjusted to achieve near stoichiometric operation. ... " The MBT timing for E-85 fuel
was a few crank angle degrees advances compared to E-0 fuel when the engine was
delivering high loads, typically 100 - 75%. Not much difference was observed at lower
engine loads."

I'm assuming that this explains the timing change as they were using what would be closed loop
fuel mixtures on the WRX. So in load cells that would continue to use closed loop fueling you
would want to have slightly more ignition timing than you would with gasoline. In the case of
WOT open loop fueling where your running a max power rich fuel mixture, I very strongly
suspect the fuel burn speed would be noticably faster, and you would use less ignition advance
than on gasoline and reap the benefit of less negative work on the late stages of the compression /
early period of combustion prior to TDC.

In a power point presentation presented by the EPA at the SAE Government and Industry
Meeting in Washington D.C. on May 13, 2003 titled:

"Ethanol-Gasoline Blends: Fuel Economy and Emissions Benefits"

http://www.epa.gov/otaq/presentation...mjb-051303.pdf

On page 9 of the presentation is a nice chart comparing laminar burn speeds of gasoline and
Ethanol.

It shows the following burn speeds:

Mixture fuel/air---- 1:1 ------------ 1.1:1

Gasoline --------- 26 cm/sec ------- 30 cm/sec (max about 31 cm/sec)

Ethanol ---------- 41 cm/sec ------- 45 cm/sec (max value)

Clearly at +10% - +15% rich mixtures where ethanol shows max thermal efficiency it burns
significantly faster than gasoline. Which could be quite important to a very over square bore
engine like ours!!
What is the ideal fuel air mixture for E85?

If you are tuning with a wideband O2 sensor you will want to switch it to Lambda mode to get
valid fuel air mixtures with blended fuels. If you can only get gasoline AFR's, simply divide the
gasoline AFR's by 1.5 to get the true AFR for 100% E85. If running a partial mixture you can
make proportional changes based on the percentage of E85 in the mix.

Here are some comparisons of stoichiometric fuel mixtures for different fuel blends:

======== stoichiometric AFR ===== max power rich AFR

Gasoline ---------- 14.7:1 -------------------12.5
100% E-85 ------- 9.73-9.8:1 ------------- ~ 9:1 - 8:1
100% fuel ethanol - 9:1 ------------------- ~ 7.2:1

One source specifies that the proper fuel tune (stoich) for a VW Golf running 22% ethanol was
12.7:1 (this reference was probably from the Brazil tests)

Ethanol reaches max torque at richer mixtures than gasoline will.

Using a 99% ethanol mixture MEP increases with mixtures up to 40% excess fuel where with
gasoline MEP is reached near 20% excess fuel. Over all thermal efficiency for both gasoline and
ethanol is reached near 15% excess fuel.

E85 burns faster than gasoline at best mixtures so it is an inherently more effecient fuel. It also
produces more exhaust gas for a give weight of fuel air mix giving higher average cylinder
pressures inspite of lower EGT's. With streight E85 in a properly tuned car its good for about
+5% power / torque increase. I suspect on a turbocharged car the benefit is larger.

My source lists stoich for E85 as 9.8:1 ("Burn rates and emissions from ethanol gasoline blends")
http://www.engin.umd.umich.edu/resea...5-2004.doc.pdf
Here's a little reference chart I whipped up when I was working out my dyno numbers.

Code:

Fuel AFRst FARst Equivalence Lambda

---- ----- ----- Ratio -----
=======================--====================================

Gasoline stoich 14.7 0.068 1 1

Gasoline Max power rich 12.5 0.08 1.176 0.8503
Gasoline Max power lean 13.23 0.0755 1.111 0.900

=======================--====================================

E85 stoich 9.765 0.10235 1 1

E85 Max power rich 6.975 0.1434 1.40 0.7143
E85 Max power lean 8.4687 0.118 1.153 0.8673

=======================--====================================

E100 stoich 9.0 0.111 1 1

E100 Max power rich 6.429 0.155 1.4 0.714
E100 Max power lean 7.8 0.128 1.15 0.870

=======================--====================================

Detail discussion of E85 tuning can be found at specialty E85 web forums such as
www.e85forum.net/forum where they have assembled people with lots of first hand experience
tuning with E85 in high performance situations.

E85 fuel ethanol content seasonal changes

E85 fuel blends change content seasonally just like all gasolines blends change with the seasons.
This seasonal change in blend for E85 is primarily to improve cold weather starting, where the
changes in gasoline during cold weather is to improve cold weather starting. In the summer
months, gasoline must be blended to reduce vapor lock and to reduce evaporative emissions.

The recommended dates for changing E85 fuel blends are listed in a chart in the E85 handbook
on page 22, which is in the "E85 Fuel Specification" tab.
The Volatility class specifications are broken down on page 10.

Volatility class 1 --- minimum ethanol 79%

Volatility class 2 --- minimum ethanol 74%
Volatility class 3 --- minimum ethanol 70%
As you can see each region has a different start date and recommendation for seasonal blends
depending on local weather climate.

Here on the high plains east of the rockies in Colorado we run the class 1 fuel blend from mid
June -- mid Sept, run class 2 fuel from mid Sept -- mid Oct and run the class 3 fuel blend from
Mid Oct -- mid April, then back to the class 2 blend from mid April to mid June. The standard
only specifies a minimum ethanol content, vendors can run higher ethanol content if it is
economical. If it is cheaper for the fuel blender to add more he can. Ethanol content is bottom
line driven by local weather conditions, and cold starting problems for local drivers just a
gasoline blends are modified to give easier starting in cold weather.

In the Southern part of Texas they would never go to a class 3 blend, and in Wisconsin, they
would only have the class 1 blend for about 2 months in the summer. In Florida they would be on
class 1 almost all year long and in North Dakota and Wyoming and Montana, they would be on
class 3 almost all year long.

Power available on E85

Many sources make a big deal about E85 having less thermal energy per gallon that a gallon of
gas. They frequently draw the false conclusion that you cannot make more power on E85 than
you can on gasoline. E85 actually has a higher specific energy at stoichiometric fuel air mixtures
than gasoline, and at proper max power mixtures releases more thermal energy in the cylinder for
a given amount of air to burn. Since an internal combustion engines power output is primarily air
supply limited this means you can make 5% to nearly 30% more power on E85 than you can on
gasoline.

(edit 6/10/08) Current experience shows turbocharged cars like the WRX and DSM families can
run upwards of 30 psi boost on E85 tunes without knock. The only case I know of where knock
was logged was on a 500+ whp DSM running 35-39 psi boost!

Ethanol specific energy at stoichiometric fuel air mixtures is actually higher than gasoline
allowing a higher release of energy per lb of air burned than gasoline.

Typical gasoline Thermal energy 19,000 BTU/lb max power fuel air mixture 12.5:1
Typical E85 Thermal energy 13,475 BTU/lb max power fuel air mixture 6.975:1
Typical ethanol Thermal energy 12,500 BTU/lb 6.429:1
If you are consuming 100 lbs of air, lets see how much fuel energy you release for each of these
fuels using gasoline as the base 100% reference.

100/12.5 = 8 lbs of gasoline @ 19,000 BTU/lb = 152,000 BTU = 100%

100/6.975 = 14.337 lbs of E85 @ 13,475 BTU/lb = 193,189.9 BTU = 127% more heat energy
100/6.429 = 15.555 lbs of Ethanol @ 12,500 BTU/lb = 194431.5 BTU = 128.9% more heat energy

Typical fuel energy contents:

1 gallon of gasoline = 125,000 Btu

1 gallon of ethanol = 84,400 Btu
1 gallon methanol = 62,800 Btu
1 gallon of gasohol(10% ethanol, 90% gasoline) = 120,900 Btu
1 gallon of E-85(85% ethanol, 15% gasoline) = 90,500 Btu
1 gallon of kerosene or light distillate oil = 135,000 Btu
1 gallon of middle distillate or diesel fuel oil = 138,690 Btu

My long term average on gasoline was 24 mpg, my mileage on the same setup on E85 was 22
mpg, my current mileage on gasoline is about 22 current mileage on E85 is 18 mpg.

Current setup upgraded turbo-

• gasoline mileage 22 mpg = 5682 BTU/mi

• E85 90,500 Btu /gallon / 18 mpg = 5028 BTU/mile (daily driving with periodic high
boost acceleration)
• E85 90,500 BTU/gallon/ 19.3 mpg = 4689 BTU/mile (driving mostly off boost as daily
commuter)

Old setup stock turbo –

• gasoline mileage Gasoline 125,000 Btu/ gallon / 24 = 5208 BTU/mile

• My old setup, @ 92% of gasoline milage or 22 mpg
• E85 90,500 BTU/gallon/22 = 4114 BTU/mile

A 5% increase in engine power is common on NA engines with minimal conversion changes,

and much higher numbers are possible if compression or boost is used to take advantage of E85's
higher fuel octane.

Last update 6/10/08)

Part 3
Testing ethanol content with the water test

The proper way to do the water test is specified on page 27 of the handbook.

Handbook method:

ethanol_guidebook.pdf
2.1 + 1.94 x (total vol - lower meniscus) = hydrocarbon
ethanol = 100 - hydrocarbon

Ethanol bonds very strongly with water so the percentage is not a direct relationship as seen
above in the formula.

For example, If you add 10 ml of water to 100 ml of standard pump gasoline, and end up with
17ml of phase separation -- you have a full E10. (yes I know the math does not work but this is
due to ethanol's bonding properties) In this case 10 ml of water and 10 ml of alcohol bond to
form a mixture of 17 ml of water alcohol mix.

Choosing proper injector size for Subaru WRX E85 conversions

http://forums.nasioc.com/forums/show...1#post21693201

Simplified formula:

(𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑚𝑚𝑚𝑚𝑚𝑚 𝑎𝑎𝑎𝑎𝑎𝑎 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝐶𝐶𝐶𝐶𝐶𝐶 𝑎𝑎𝑎𝑎 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏) 𝑥𝑥 1.38

≅ 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑐𝑐𝑐𝑐/𝑚𝑚𝑚𝑚𝑚𝑚 𝑒𝑒𝑒𝑒𝑒𝑒ℎ 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 𝑜𝑜𝑜𝑜 𝐸𝐸85

(assumes a 4 cylinder engine with similar fuel needs to the Subaru WRX)

What is your test history to date (6/2005) with E85 ?

I just tallied up my usage over the last 2 years, (the first year I was very cautious and did not use
very much). I've put about 500 gallons of E85 and 100% fuel grade ethanol through the cars tank
--- that's some where around 40 fill ups, or around 9,000 miles or so on high ethanol (ie greater
than 10%) blend fuels.
I am intentionally running the E85 with no special fuel system modifications. In a few years or
when ever problems crop up, I'll pull things apart and see what has happened. So far I have
absolutely no evidence of ANY kind of degradation on my 2002 WRX. The only non-stock
component I'm running in the fuel system at the current time is a walbro 255 L/hr pump, so I
cannot absolutely tell you how the stock pump likes the stuff.

Update ( Jan-Nov 2007)

I made a full conversion over the winter bringing the car back on the road with decapitated OEM
injectors flow tested at 810 cc/min, an Aeromotive adjustable fuel pressure regulator which gives
me the ability to dial in any fuel pressure I want from 35 psi to 70 psi with the turn of an allen
wrench. A slightly improved ECUtek reflash, and the Mistubishi 16G turbo off a 96 Australian
WRX. At 17 psi I produce 281 AWHP compared to 240 AWHP I was hitting on 100 octane
Sunoco GT100 and 250 AWHP I could hit on Sunoco GT+ 104 octane race gasoline at higher
boost pressures.

I have successfully tuned a usable home made FFV setup by optimizing the tune for a 50/50
blend of E85 and gasoline. At moderate boost levels the ECU is capable of making the necessary
adjustments and open loop fueling is acceptable (see detail notes in the FFV segment below).

I have currently moved to a 100% E85 tune since I now have adequate availability of E85 and
can run it all the time with little difficulty. In a cost per mile basis my current tune is equivalent
to getting 30.5 mpg on pump premium (20 mpg on E85 at $2.19 vs pump premium at
$3.35/gallon). On a cost basis the car is consistently cheaper to drive on E85 than it is on
gasoline.

Why is E85 a better fuel ?

Ethanol and is a very turbo friendly fuel for many reasons.

1. It has a much higher evaporative cooling power than gasoline so the intake air charge in
the cylinder is significantly cooler that it is with a comparable mixture of gasoline --- that
means higher VE.
2. Its octane as blended in E85 is about 100, its blending octane when added to gasoline is
rated at 118, so it is a very cost effective octane booster.
3. Ethanol burns faster than gasoline but has a slightly longer ignition delay during the slow
burn phase of combustion so the engine does not do as much negative work fighting
rising cylinder pressures due to large ignition advances. The total ignition advance for
E85 is almost identical to the ideal advance for gasoline so it does not cause the ECU
problems when you mix them.
4. At proper mixture you actually are releasing more energy in the cylinder due to the
higher quantity of fuel you can burn. ( Ethanol can burn effeciently at much richer
mixtures than gasoline can) That means about a 5% increase in energy release all by
itself.
5. Peak combustion pressures are actually lower for ethanol than for gasoline but the
cylinder pressures stay higher longer, so you have more (longer) crank angle that is
usable by the engine. This lower peak cylinder pressure also helps with detonaton
control.
6. It will, at proper mixtures lower EGT's by around 200 deg F, but due to the higher
quantity of exhaust gas products it produces you do not lose any spool up (in fact I would
wager spool up is better).
7. It is much cheaper ( if you go to a station that is not trying to price gouge).
http://www.epa.gov/otaq/presentation...-isaf-no55.pdf

8.

How much will my miles per gallon of fuel drop with E85?

The only negative to E85 is that it gives a lower fuel milage on a gallon for gallon basis to
gasoline. The actual difference in energy content between straight gasoline and E85 is about
27%.

The drop in mileage is not as significant as you would think based on that difference due to the
higher efficiency of the ethanol as a high performance fuel. This winter I was getting about 92%
of the fuel milage I would get on gasoline on 100% E85.

The lower mileage is not really a big deal, ethanol has lower energy per gallon but your
reduction in milage is not nearly as large as that difference would imply. Due to the higher
torque,you use slightly smaller throttle openings to get the same level of performance, and due to
the greater quantity of combustion products (more moles of gas) per lb of fuel the engine
efficiency actually goes up slightly. My long term fuel mileage average is in the vicinity of 24.5
mpg, with pump gas, and with 75% ethanol blend, I was getting just over 23 mpg driven
normally. Recently I have been flogging the crap out of the car to sort out new boost controller
settings for my new turbo (went from a 13T to a 16G). Given I now have a larger turbo and all
that is hardly a noticeable fuel mileage drop. I have gotten around 300 -345 miles/tank on
straight gasoline when I was bone stock, and I expect to get from 280 - 310 miles per tank on the
E-85 based on my notes of fuel consumption and accounting for the unusually hard driving I
have been doing the last week working on the boost controller settings.
In very cold weather <20deg F I don't go above about 90% E85 to improve cold starting and
speed up engine warm up a bit. Other than that the car loves E85 and so does my wallet ----
$1.89/gallon (6/2005) for 100 octane fuel is hard to argue with. It only drops my fuel mileage a
small amount. I get 93.76% of my gasoline milage when driving conservatively in my WRX
with the larger injectors and high ethanol fuel blends.

Will a wide band O2 sensor accurately read fuel air mixtures with E85 blends ?

To get an accurate AFR reading you need to switch the meter to Lambda or equivalence ratio
setting rather than AFR. Most O2 sensors assume you are running gasoline and will report a
stoichiometric mixture as 14.7:1 which is the proper value for gasoline. E85 has a Stoichiometric
mixture of between 9.7 - 10:1 and a max power mixture of about 6.98-8.5:1 or so, where with
gasoline it is 12.5:1-to 13.1.

If you must use an O2 sensor that only reports gasoline AFR information simply divide the
numbers it reports by 1.47 - 1.50.

On gasoline, my ECU is supposed to give a mixture of 11.5:1 and on 100% E85 the dyno's wide
band reported an AFR of 11.6:1. That means that my true AFR on the E85 was about 7.8:1
which is right in the middle of max power mixtures for E85.

---Larry
The energy balance question

Some sources say that fuel ethanol requires more energy than it returns is this true?

No not really based on recent studies using very detailed analysis of actual energy inputs
currently used, in modern farming and state of the art ethanol plants the numbers come out as
below. Note that gasoline actually delivers less energy than was expended getting it into your
tank.

It has a negative energy balance. Which is exactly the claim the anti-ethanol lobby is blaming
fuel ethanol with as a good reason not to use it.

=====================
Fuel ethanol returns:
+31% (Wang - 2002)
+34% (Shapouri - 2002)
+21% (Graboski - 2002)
Fuel ethanol with state of the art production techniques can return +68% (Wang - 2002)

Gasoline energy balance (GREET V1.6)

Conventional gasoline -19%

Reformulated Gasoline -20%
MTBE -33%

===========================

http://www.mda.state.mn.us/ethanol/balance.html

A USDA study released in 2004 found that ethanol may net as much as 67% more energy than it
takes to produce. Argonne is one of the US Department of Energy's largest research centers.

http://www.transportation.anl.gov/pdfs/TA/349.pdf
http://www.agriculture.com/ag/story...l&catref=ag1001

Report on the new study :

http://www.ncga.com/public_policy/P...thanolStudy.pdf
http://journeytoforever.org/ethanol_energy.html
The cellulose based ethanol production will actually return nearly 100% energy because it
will use its own waste product stream for process energy.

Using the cellulose based process also eliminates the argument that more land will be required to
grow corn than is available. Fact is many crops can be used to brew ethanol.

A list of common production sources of the above include.

• Corn
• sugar cane
• Sugar beets
• wheat
• tapioca
• potatoes
• rice
• barley
• molasses
• fruit
• whey

It can also be commercially prepared from cellulose treated by enzymes. Sources include:

• sawdust
• waste paper
• scrap rags
• alge
• grasses (switch grass)
• crop waste

==========

We're not talking about the total substitution of ethanol for gasoline, but a concerted effort to
displace as much as possible. The U. S. is already the worlds second highest consumer of fuel
ethanol behind Brazil. Every barrel of fuel ethanol burned replaces 3/4 of a barrel of petroleum
on a energy content basis. At current oil prices nearing $70 per barrel that is a significant amount
of money payed out to U.S. suppliers for a commodity instead of to foreign suppliers. Given that
money rolls over 5-6 times in an economy before it get sequestered in a long term capital
investment, that means $350 dollars of net benefit to our economy for each barrel of fuel ethanol
consumed vs a negative $50 of loss of revenue to the overseas supplier.
Yes I know that much of that money gets passed through the hands of U.S. companies but the
end point is still the overseas oil producer.

Even more important, the energy inputs for fuel ethanol production are in large measure low
quality heat requirements that can be met with domestic sources of energy which we have in
great abundance, like Coal or waste heat from other processes.

How does fuel ethanol rank on Green house gas production ?

Using E85 you have a very favorable green house gas production picture. The CO2 produced
when burning the ethanol portion of the fuel is totally recovered by the next seasons crop. This is
the definition of a renewable fuel.

In short your "Net" green house gas production is only 15% of what it would be on straight
gasoline. That is the equivalent of getting 6.7 times your normal gas mileage.
How to make your Subaru a Flew Fuel Vehicle (FFV)
How to make your car (Subaru) a FFV at home with very little money or effort.

Well I have been sitting on my hands, and biting my lip now for a couple months waiting until I
had the data to backup what I am about to suggest.

I believe I have figured out how (at least on the Subaru WRX) to quickly and easily achieve a
home made FFV. The solution is so simple it makes my head hurt, but I wanted to check it out to
see if the ECU would do what I expected it to do. (sometimes those electrons do odd things you
don't anticipate)

First a review of the basics.

Pure E85 requires about 30***37; more fuel flow than pure gasoline to run normally.
The Stock ECU handles up to a 30% blend of E85 with absolutely no problems, not even
a CEL.
The Stock ECU will run just fine on a 50% blend of E85 but will throw a CEL because it
has used up its fuel trim authority of +/- 25% to get closed loop fueling back to Stoich
(Lambda 1.00).

E85 is a cool burning high octane fuel that will tolerate leaner mixtures than gasoline
under high load.

Gasoline likes rich mixtures under high load to prevent detonation if its octane is a bit
low.

On my first attempt to convert the car, I switched to the JDM STi 550 cc/min injectors and made
no other tuning changes. This increased the fuel flow by 25% (stock injectors are 440 cc/min).
The car ran beautifully on 100% E85, ran good on 50/50 E85 and gasoline but developed a small
hitch on transition to open loop fueling from closed loop fueling near 30% -40% E85..

The light dawns !!

At this point I realized that if you could setup a circumstance where the ECU was sitting on zero
fuel trims at a 50/50 mix of gasoline and E85 it had enough fuel trim authority to cover both
extremes of 100% gasoline or 100% E85.
At this point I planned out my next experiment ---- unfortunately it took me nearly a year to
assemble the parts I needed (not because they were necessary for the tests, but because I was
scratching for the funds, and had other plans down the road and only wanted to do the upgrade
once.)

For the last few weeks I have been driving in all sorts of weather (including -19 temps) with
100% E85 and now recently a 50/50 mix of E85 and gasoline.

Here's the magic combination!

All you need to do on a Subaru WRX (2002 -- I assume others will behave the same), is
increase your fuel low by 15% over stock. This will make the ECU have zero LTFT's and
STFT's on a 50/50 mix of gasoline and E85.

If you plan to run mostly on E85 you can set the neutral point a bit higher, if you expect to need
to spend a good fraction of your time on pure gasoline set the neutral point a bit below the 50/50
fuel blend.

The ECU will use 15% of its fuel trim authority, going positive as you move toward 100% E85
and going negative as you move toward 100% gasoline.

I have spent the last couple weeks driving around on -15% fuel trims and the car runs just fine.

The easiest way to add 15% to the fuel flow, is to up the fuel pressure from 43 psi to 57 psi.
This has two minor down sides. If you have an upgraded turbo (and fuel pump) you will run out
of injector at high load (not good --- don't go there). On the stock turbo you will also run out of
fuel pressure and flow from the stock fuel pump at high boost (need to add a high flow fuel
pump if you run over stock boost levels).

The slightly more expensive way to do it, if your interested in performance, is to install larger
fuel injectors, an upgraded fuel pump and either adjust fuel pressure or re-scale the injectors with
one of the aftermarket tuning solutions.

This is what I did, I have an Aeromotive adjustable fuel pressure regulator, 800 cc/min
decapitated injectors and an ECUtek reflash. Harvey at Super Rupair in Boulder has been worth
his weight in gold helping me get the reflash tweeked properly to prove to my self I am on the
right track. He's made nearly daily adjustments to the map as I tried out different things.
Thanks Harvey !!! (and all the guys at Super Rupair who have worked with me on this
experiment).

More tuning will follow as I work out the final setup I am looking for, but I wanted to get the
word out so some of the other E85 experimenters could do some peer review of my concept and
prove if it is a safe and reasonable solution for the average user.

Quote:

So have you logged +15***37; fuel trim being applied throughout WOT with 100% E85? And
-15% LTFT with 100% gasoline?

Not completely, still burning a high ethanol mix at the moment (last week was all at 50/50 mix),
trying to get some milage figures. I'm currently getting 24.3 mpg on the 50/50 mix which is
better than most folks can manage around town on gasoline.

I have run the 100% E85 and with no changes in the tune, on the 50/50 mix its applying a -15%
LTFT. I played with the fuel pressure a bit and found everything is consistant with my
expectations. The base line tune on E85 has near zero LTFT and it is now -15% on the 50/50
mix. I will get the injectors rescaled this coming week to bring the LTFT down to zero then we
can start some more controlled tests.

It will take several more weeks to get enough miles on the car to finish a comprehensive battery
of tests but I feel confident with the results seen so far, that there will be no surprises. On the
50/50 mix using the E85 baseline tune it is still showing a IAM of 16 across the board with no
knock correction at any rpm or load I have been able to hit.

The Wideband just got installed yesterday and confirmed the readings I'm seeing in the Delta
Dash logs, so I know I can trust the DD log numbers for AFR within a decimal or so. Waiting on
the arrival of a new EGT to finish up the instrumentation I need before I start pushing it.

I still want to lean it out a bit, both in open loop and on tip in. Harvey is still tweaking the MAF
scaling to get it exactly right for my intake setup. Then I will start beating on it a bit as the
weather warms up. Kind of tough to flog it really hard when its foggy and snowing and temps
jump up and down from -19 lows one day to 55 degree highs a couple days later.
Quote:

This is great news! What happens when you transition from a 100% Petorul, to 100% E85?
How long until the you run from -15% to +15% fuel trim?
As above have not done the 100% gasoline runs yet (since they are the most risky due to its
lower octane) I will do that last. The LTFT's change very quickly but take a while to stabilize on
a fixed value. I bumped the mixture up from 50/50 mix this morning to about a 75% E85/ 25%
gasoline mix and the trims went from the -15% on 50/50 mix to -7% trims within 2 city blocks.
The Wideband shows the ECU was hitting Lambda 0.99-1.01 before I pulled out of the gas
station.

More to follow, as I keep nibbling away at this. I want to get all these numbers using typical
daily driving so it takes a while to crank off the miles.

That is one of the reason I wanted to give a heads up to the other experimenters to get them
thinking and testing along the same lines.

FFV test#2 completed

sudden transition from 50/50 mix of gasoline and E85 to a tank of 100***37; gasoline.

I ran the car on a 50/50 mix of E85 and gasoline for a bit more than 2 full tanks (600 miles). I
tweaked the injector scaling so that on that 50/50 mix the LTFT's and the STFT's approached
zero ( By that I mean the sum of them was zero +/- 5%).

I then ran the tank down to about 1/2 -1 gallon into the reserve (fuel fill light on for 10-15 miles)
and then filled the tank with 91 octane Premium gasoline.

Upon starting STFT's immediately (within 10-15 seconds) had scrolled up to -25% then the
LTFT started to decrease.

LTFT's and STFT's settled in around the near zero area but were slightly higher than I expected 0
===> +8% on occasion.

The good news is that the switch to gasoline is a graceful failure mode, in that if the ECU does
not immediately apply full correction, the mixture goes rich.
Watching the Wide Band, cruise mixtures still settled in at Lambda 0.99 - 1.00 but you could see
it was bouncing around more before it settled.

About 10 miles after I put the new fuel in the tank I did a WOT blast and saw mixtures on the
Wide Band of 0.69 ( 10.143 AFR ) --- to ---- 0.72 (10.584 AFR) These are essentially the stock
AFR's !!!!!

Now I will watch things and see how the fuel trims settle in. There were no drivability issues
with the sudden switch of fuels other than an occasional pop from the tail pipe.

I'm going to let the car cool a bit and run it through a couple more driving cycles here in a bit.

Stay tuned for more of " How the ECU Turns "

( I know more controlled tests would be useful but I am trying to get a first pass "coarse
evaluation" before I start confirming the fine details. Next I will run this tank down to the fill
light and throw in some straight E85 and see what happens.)

--Larry

It takes time, each driving cycle it is closer.

On my second driving cycle (15 - 32 miles) the LTFT's hung around -11.8***37; ===> -14.1%.

I just finished my 3rd driving cycle since the change in fuels. Total of 55 miles.

In light throttle cruise it is showing a LTFT that varies from -14.1% ==> -15.7%, so it is closing
in on exactly that number.

Just out of curiosity Jon --- do you know what the ECU defines as a "driving cycle" ? Does the
engine simply need to be shut off, or does it need to cool enough to go through warm up
enrichment steps.

In the for what its worth department, the way the LTFT's are acting is I think the ECU may keep
a moving average of the LTFT somewhere for certain conditions and use that default value the
next time the engine is started.

The reason I speculate this, is on the first driving cycle most of the adjustment was made in the
STFT and the LTFT tried to stay near its historical average prior to the fuel change. After
shutting the car off, when it was restarted for the second driving cycle the LTFT immediately
jumped to a value that was near the average adjustment it needed on the first driving cycle until it
warmed up, then the LTFT started moving around again, and slowly got more negative as I
drove.

After shutting the engine off and starting the 3rd driving cycle the LTFT did the same thing, it
immedialtely went to a -15.7 value and stayed there until the engine warmed up, then both the
STFT and LTFT moved around a bit as the LTFT gradually settled near -14.1% most of the time.

We will see what it does on the 4th driving cycle, after the engine has some time to cool off.

How to make your Subaru a FFV Update II

Okay I finished enough testing to post a recipe for a home built WRX FFV.

I have satisfied myself that the following concept is sound as the car is behaving as I expect, and
so are the fuel trims.

You can run this in two different modes. You can let the fuel trim authority of the ECU do all the
work (we will call this "autotuning mode"), or you can make life easier on the ECU by a minor
tweak of the fuel pressure to help it get in the ball park quicker -- (we will call this "user assisted
tuning mode").

Autotuning works just fine, but takes about 4 driving cycles for the ECU to settle into final
LTFT's. It is a bit less than optimal tuning at the ragged edge so if your tune is a max power tune
I would not recommend this. For a happy daily driver at moderate boost levels (<17 psi) I think it
will be safe unless there is something amiss with the base tune on the ECU.

Step 1 -- Upgrade fuel pump to a walbro GSS 342 255 l/h pump or equivalent.
Upgrade fuel injectors to have at least 30***37; more max flow than stock.
Install an adjustable fuel pressure regulator and an under hood fuel gauge.

For autotune mode. Set the fuel pressure at 45 psi and get the ECU reflashed with an Ecutec or
equivalent reflash with a 50/50 mix of E85 and gasoline in the tank. Use the fuel you normally
have available and summer blend 85 if available (winter blend E85 is about 70% ethanol so if
tuned with the winter blend, richen the mixture a tad in the wide open throttle areas.)

On the 50/50 mix of gasoline and E85 at 45psi fuel pressure you should have the flash setup so
you get LTFT's that settle down near zero or slighly negative ( 0 to -5% ).
If you simply add 100% gasoline to the tank, the ECU will auto tune to -10% to -15% fuel trims.
You will get normal stoich AFR's while in closed loop but your WOT AFR's will be a bit rich
(around 10:1).

If you simply add 100% E85 to the car and let it auto tune, you will get normal stoich AFR's in
closed loop fueling and your open loop fueling will be a bit leaner than the tune at 50/50 fuel mix
by about 7%.

For user assisted auto tune mode, you simply adjust the fuel pressure slightly when you make a
major change in the fuel in the tank.

35 psi for 100% gasoline

45 psi for 50/50 gasoline and E85
55 psi for 100% E85.

If you adjust the fuel pressure as above, you will keep the ECU near 0 LTFT's and your WOT
AFR's will be only slightly effected by the fuel switch.

The adjustment of fuel pressure does not need to be precise. Simply make a ball park guess what
the blend in the tank is, and you will be close enough unless your running a high boost system
that needs the full octane of the E85. In that case, I assume you have access to some sort of
logging ability and or a wideband so you can tune accordingly or have a dedicated 100% E85
max performance map.

My setup (essential mods for the FFV conversion)

• Aeromotive adjustable fuel pressure regulator.

• 800 cc / min decapitated OEM injectors.
• Walbro GSS 342 255 l/h fuel pump.
• Ecutec reflash -- tuned at 50/50 mix of gasoline and E85 with -5% LTFT's after is settles
in (more than 4 driving cycles). WOT wide band is showing 0.82 - 0.84 lambda on the
50/50 gasoline, E85 fuel mix.

(non-essential mods)

• upgraded fuel system under the hood (fuel rails and new braided fuel lines)
More details to follow as I get more miles on the setup and get better info.
50/50 gasoline - E85 mix giving me 24 mpg, same milage on 100% gasoline. I will update the
fuel milage on 100% E85 when I finish this fuel tank, but am guessing it will turn out to be about
22 mpg.

Just a little update on the E85 conversion project.

I've been running on 100***37; E85 for several fuel tanks and the long term fuel trims at idle
have stabilized at a rock solid 14.8% so my attempt to get a good split is only 0.2% off of ideal.

In watching the fuel trims over time, I would suggest that folks attempting this watch their idle
(engine warm) long term fuel trim. It is very stable and does not move around as much as the
LTFT does at other throttle settings.

Currently fuel milage on the 100% E85 is running in the low 20 mpg range, compared to a long
term average of 24 mpg on straight gasoline before I began this last major conversion process.
That gives a milage reduction of about 15% - 18% depending on how I drive.

Given the fuel price paid today for E85 vs current fuel prices for 85 octane regular, I am running
a 105 octane fuel for the equivalent cost per mile as if I was buying mid grade gasoline.

(note I have been running significantly higher boost recently that I ever tried on the older
conversion with the stock turbo or the 16G before I had enough injector to feed it, so some of the
fuel milage loss is directly related to higher power levels)

I think there is more milage there but I have resisted the temptation to keep fiddling with the tune
for a while to get stable numbers as a baseline for future comparison.

By the way to all concerned, after several full tanks of E85 on the new setup I am still getting the
behavior I expected, with my idle fuel trims stabilized at +14.8***37; on 100% E85 and 55 psi
base fuel pressure.

Fuel milage is a bit lower than I wanted at high mid 18 mpg - low 20 mpg depending on how I
drive. This may have more to do with a heavy foot rather than a solid comment on the setup.

I will be dialing back the fuel pressure a bit and see if that helps the fuel milage by leaning out
the open loop a bit.

--Larry
Cold start problems due to summer/winter blend switch over.

By the way folks a brief update. I recently have had some cold start problems with my setup, and
it appears this is due to the switchover of the local E85 to the summer blend (85***37; alcohol)
from the winter blend (probably 70% -75% alcohol). The switch appears to have happened the
last week or so of March or the first week of April. I need to richen up my tune slightly in the
low rpm ranges near idle. The winter time base tune is a tad too lean for good cold starting on the
summer blend at temps near 45 deg and high humidity.

Quick fix is the same as for winter cold starting just chuck in a couple gallons of gasoline with
the 100% E85 and problem magically goes away.

The symptoms of too lean on start, is the car will not fire immediately on the first few cranks and
when it does fire it stumbles and coughs for a second or so before the idle cleans up. heavy
throttle application when cold will lead to lean back fire in the intake manifold. Problem goes
away as soon a engine gets up above about 100 deg F water temp.

Emissions test on E85

I have been trying to figure out how the system behaves in that first 3 -4 seconds after you start
the engine. I suspect that the problem is that when you shut the engine off the idle air control
valve goes to its home position (not sure if that is wide open or fully closed). Then right after the
engine starts the IACV attempts to get control of the idle speed, and the mixture is determined
entirely by the default mixture and post start enrichment since the front O2 sensor is not active
yet.

It is during that phase of the startup and post start run that it struggles.
Having grown up during the era when some cars still had manual chokes, and struggled for
months to get my automatic choke on my Hemi to work right, I immediately recognized what
was going on as all the symptoms matched with a car that did not have the choke set right. It just
took a while for the brain to realize the E85 blend may have changed on my last fill up. The
weather in that time period was quite variable so I would have shirt sleeve weather for a couple
days then a cold damp morning followed by a couple more nice warm days so it took a couple
weeks to get a good feel for it.

(I'm not sure the lean mixture back fires were bad enough to damage the intake if it had been
plastic but I know what you mean, I have had cars pop back like that and bend choke plates and
jamb them in the air horn of a carburator. I also know of folks that have blown the guts out of air
filters due to lean backfire).
On a totally unrelated topic, I'm a happy camper, I just ran the WRX through emissions and
passed with flying colors.

100***37; summer blend E85 with an extra 1.5 gallons of gasoline added to the summer blend,
on Harvey's baseline tune, I have a single Spintech high flow cat in the exhaust now.

For those of you that don't know the Colorado emissions test is a driving cycle,(I/M 240 test) on
an AWD dyno test so these are under driving condition tail pipe emissions on E85 .

http://www.aircarecolorado.com/im240tst.htm

Engine has 59K miles on it, with over 100 drag strip passes on it, no detectable oil consumption,
converter is new.

Code:
............ reading .... limit ...
HC GPM 0.3335 1.20000
CO GPM 3.9638 20.00000
CO2 GPM 414.3504 -- NA ---
NOx GPM 1.9216 3.00000

The National Center for Vehicle Emissions Control and Safety at Colorado State University has
done extensive testing of in-use vehicles and their converters. Part of the testing included the
development of converter feedgas charts for fuel-injected and carbureted vehicles (see CO, HC
and NOx feedgas charts below).

gpm = grams of pollutant per mile driven

Fuel injected 4 cyl

• HC ~= 1.0 - 2.0 gpm

• CO ~= 10 - 15 gpm
• NOx ~= 1.3 - 2.0 gpm

Based on those numbers my NOx is a tad high so I will richen up my tune a bit as HC & CO is
well under control.
Also a richer mixture is better for a max performance tune on E85. Once I get my water injection
setup re-installed that will also drop NOx a considerable amount.
A cautionary note for anyone playing with E85 seeking to copy my home made FFV
project.

After some testing today, I think I would have to issue a strong caution to using the "autotune
mode" if you are running any significant boost levels (over 8-12 psi).

As you know, I recently have been running a bit lean, and part of that was due to the recent
change over from winter to summer blend E85. Well I managed to scare myself today! I have
recently been exploring fuel economy issues with the E85, using the midrange fuel pressure
setting and let the system use LTFT's only to handle the E85, starting from a safe tune at a 50/50
gasoline-E85 blend. The result is, I made a few too many tweaks to lean the mixture out, working
entirely in the low boost range.

I first did a bit of logging and found that driving like a normal human being in daily traffic, I
rarely went over 2-3 psi boost in typical city traffic. On the few occasions that I did, I seldom
was on boost long enough to get over about 8 psi boost (this was in economy driving mode, ie
easy into the throttle avoiding WOT squirts).

I have been driving on the summer blend E85 now for a couple weeks so all the fuel trims had
fully stabilized and today was a nice cool day with some light sprinkles of rain, so I thought it
was a good time to do a couple of WOT runs and see what was happening under those
conditions.

I found that I was running lambda 1.02 ( 15:1 gasoline AFR) all the way up to 18 psi.
Eeeeeekk!

Needless to say it did not take me long to get out and crank up the fuel pressure to 55 psi. That
dropped my Lambda under WOT open loop down to 0.82.

The good news is, I was expecting it to be a bit too lean, so had been purposely avoiding
flogging on the car hard. I was also so lean I was over on the lean side of the high detonation
mixtures. No harm no foul, but just a red flag and a word to the wise, it is easy to get so carried
away "tweaking the system" that you go a bit too far!

All that said, I think it still has some possibilities for folks that are willing to run low boost or
NA setups strictly for the true FFV capabilities but I would not recommend it now, especially on
the summer blend E85 if you like to run the car hard.
Autotune would work fine if you also had a water injection setup however.

The other observation is that the single best thing you can do to increase your fuel milage is to
change your boost curve! I have a Blitz DSBC which has 4 settings (channels) and I can set the
boost gain independently on these channels. By running a very low gain on the first channel I can
create a situation where in day to day driving my boost seldom gets up above 3-6 psi and a full
WOT pull will only go up to about 1/2 of the max boost limit I have set. Then by simply twisting
the knob to channel 2 which has a higher gain on it I have a normal agressive boost profile.

Using the low gain setup I bumped my fuel milage from 18 mpg to almost 24 mpg with no other
changes. In short agressive quick spooling boost curves are your enemy if your looking for good
fuel milage.

IMPORTANT update on FFV auto tune

post 950 (page 38 very lean on open loop running Autotune if you turn down fuel pressure)

============
The Jan 2008 HotRod Magazine has an article on E85 that is quite positive on its use as a
performance fuel.
(although nothing new in the article that has not known for several years by E85 experimenters)

They have the results of a back to back dyno test of 100 octane race gasoline vs E85 on a 402
CID LS2 , 10.2:1 compression ratio, crate engine, done by Wheel to Wheel Powertrain in
Madison Heights Michigan.
Gasoline (100 octane race gas) 540 hp @ 6000 rpm, 509 ft-lb torque @ 5200
E85 ------------------------ 546 hp @ 6000 rpm, 524 ft-lb torque @ 5200

With a turbocharger kit installed fueled on E85 :

13 psi boost -- 850 hp @ 5900 rpm, 833 ft-lb torque @ 5000 rpm
no comparison to gasoline you can't run that kind of boost on it with this engine.

Quote:
"There's no sense even trying it with this boost and compression," Urban says."You just can't
do this with pump gas" With its knock-stifling 105 road octane, E85 is a pump fuel that
performs like race fuel. "I love this stuff," Urban says. "Its high-octane fuel for everyone, 105
you can buy on the road"
discussion regarding walbro fuel pumps and E85 .LINK

Fuel line sizing

The OEM fuel line is big enough for at least 600 hp.

Eaton has a chart on fluid line sizing for their Aeroquip lines. It shows that you only need the big
lines on low pressure systems or systems running under some vacuum. On EFI where fuel line
pressure is 43+ psi you can run a significantly smaller dia than on low pressure carburated
systems.

The chart is intended for motor oil or hydraulic oil but gasoline has much lower viscosity and
flow losses than oil.

See page 406 this doc.

http://www.hydraulicseaton.com.cn/gb...16_394-433.PDF

Max flow for a walbro 342 is 255 l/hr or 67.4 gal per hour = 437.9 lbs / hr or about 700 hp on
gasoline, or 560 hp on E85.

The OEM hard line is 8mm which is basically 5/16 inch ID line, which at the 15 ft/sec flow rate
approx gives you 13 gal/min. or 780 gallons per hour.
That is WAAAYYY more fuel than you need for 600 hp.

Seasonal fuel blends

Quote:

I believe in winter, the refiners change E85 to 75% alcohol to improve cold weather
performance.

Yes that is correct in some areas, the seasonal changes differ depending on where you are in the
country.

I thought I had posted this before, but will put it in the front of the FAQ so folks can find it too!
E85 fuel ethanol content seasonal changes

The recommended dates for changing E85 fuel blends are listed in a chart in the E85 handbook
on page 22, which is in the "E85 Fuel Specification" tab.
The Volatility class specifications are broken down on page 10.

Volatility class 1 --- minimum ethanol 79%

Volatility class 2 --- minimum ethanol 74%
Volatility class 3 --- minimum ethanol 70%

http://www.e85fuel.com/pdf/ethanol_guidebook.pdf

As you can see each region has a different start date and recommendation for seasonal blends
depending on local weather climate.

Here on the high plains east of the rockies in Colorado we run the class 1 fuel blend from mid
June -- mid Sept, run class 2 fuel from mid Sept -- mid Oct and run the class 3 fuel blend from
Mid Oct -- mid April, then back to the class 2 blend from mid April to mid June. In short here in
Colorado near Denver we are already on the winter blend, but there is latitude in the standard, so
the standard only specifies a minimum ethanol content. If it is cheaper for the fuel blender to add
more he can. Ethanol content is bottom line driven by local weather conditions, and cold starting
problems for local drivers just a gasoline blends are modified to give easier starting in cold
weather.

In the Southern part of Texas they would never go to a class 3 blend, and in Wisconsin, they
would only have the class 1 blend for about 2 months in the summer. In Florida they would be on
class 1 almost all year long and in North Dakota and Wyoming and Montana, they would be on
class 3 almost all year long.

Larry
Injector sizing

Original post - #1409

Quote:

hotrod you have you come up with a more concrete formula for calculating injector sizing

It depends a bit on how conservative you want to be but you can get a pretty quick estimate
doing it this way.

Fuel flow required is directly proportional to the total air flow, which obviously changes with
boost pressure etc.

On gasoline, you need about 1.6 - 1.8 cc/min static flow on each injector for each hp the engine
makes (4 injectors total)
You need about 2.0 - 2.3 cc/min static flow on each injector for each hp the engine makes on
E85 (4 injectors total)

1 HP approx equals 1.45 CFM = approx 0.108025 lbs air flow

Our engine needs about 0.59 lbs of gasoline per hp / hr. Each gallon equals 3785 cc, and gasoline
weighs about 6.2 lbs/ gallon, so that 0.59 lbs of gasoline is 0.095 gallons or 360 cc/min injector
flow on 4 injectors. That means you need 90 cc/ hr injector flow per hp or about 1.5 cc/min per
injector to make one hp on gasoline.

Since E85 requires about 30% more fuel that works out to about 1.95 cc/min of E85 to make a
hp.

Since you can approximate the maximum power you can make based on the airflow in CFM
divided by 1.45 that is a handy way to ball park your injector size.

For E85 take your turbochargers max flow in CFM at the boost you expect to run and divide by
1.45, then multiply by 2.0 and you will be pretty close in cc/min for each injector.
For gasoline it would be divided by 1.5 giving 403 cc/min on the stock turbo with gasoline.

• Stock turbo flows 390 cfm max flow at 14.7 psi

390/1.45 x 2.0 = 538 cc/min on E85
• 16G small max flow 505 cfm
505/1.45 x 2.0 = 696 cc/min on E85
• 20G max flow 695 cfm
695/1.45 x 2.0 = 959 cc/min

If you do not want to muck around with the formula it simplifies to turbo air flow in CFM at the
boost pressure you intend to run x 1.38 will give you the approximate cc/min static flow you
want to use for your injectors on a 4 cylinder engine.

simplified formula :
*assuming 4 cylinder turbo engine

(𝐶𝐶𝐶𝐶𝐶𝐶 𝑎𝑎𝑎𝑎 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏) 𝑥𝑥 1.38 ~ = 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑐𝑐𝑐𝑐/𝑚𝑚𝑚𝑚𝑚𝑚 𝑓𝑓𝑓𝑓𝑓𝑓 𝑒𝑒𝑒𝑒𝑒𝑒ℎ 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖
Transmission Information
Subaru Transmission Chart
http://www.gearhack.com/myink/ViewPage.php?file=docs/Subaru%20Transmission%20Chart

This chart came from a PDF document (attached) that I downloaded off the web a long time ago.
I believe it came from Transmission Information Spreadsheet.

Since that PDF chart is now out of date with newer Subaru models, I decided to put all the
information in this wiki. While I search for a replacement for my current transmission, I can fill
out this wiki with newer data that I find. The PDF spreadsheet is not printable, nor can you copy-
and-paste from it. However, you can print and copy-and-paste from this wiki chart.

I'm sure this chart will become outdated after I'm done with my own swap. But being part of this
wiki, anyone (including you) can come along and update this chart in the future. Good luck and
godspeed.

T.R. = Transfer Ratio

F.D. = Final Drive

6MT Transmission

Center Front
Application
Model Range Trans. Code 1st 2nd 3rd 4th 5th 6th Rev. T.R. F.D. Diff. Diff. Notes
Notes
Type Type
Aus. Impreza WRX Viscous A.P.
?? 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
STi MY02-04 (4kgf) Suretrac
JDM Impreza WRX
STi MY01-02
Viscous
(also Prodrive Style TY856WB1AA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 Open
(4kgf)
& STi Limited
MY02)
JDM Impreza WRX
STi & STi RA
MY01-02 Viscous A.P. RA STi w/
TY856WB1CA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
(also Prodrive Style (4kgf) Suretrac 17 in. wheel
& STi Limited
MY02)
JDM Impreza WRX DCCD A.P. RA STi w/
TY856WB1BA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
STi RA MY01 (45:55) Suretrac 16 in. wheel
JDM Impreza WRX
w/ external
STi RA Spec C Viscous A.P. w/ 17 in.
TY856WB2GA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 transmission
MY02 (4kgf) Suretrac wheel
cooler
(also S202)
JDM Impreza WRX w/ external
DCCD A.P. w/ 16 in.
STi RA Spec C TY856WB2DA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 transmission
(45:55) Suretrac wheel
MY02 cooler
JDM Impreza WRX Viscous
TY856WB3AA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 Open
STi MY03 (4kgf)
JDM Impreza WRX DCCD A.P. automatic DCCD
TY856WB3KA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
STi MY03 (35:65) Suretrac system
automatic DCCD
JDM Impreza WRX DCCD A.P.
TY856WB3JA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 on 17 in. wheel
STi Spec C MY03 (35:65) Suretrac
only
JDM Impreza WRX Viscous
TY856WB4AA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 Open
STi MY04 (4kgf)
JDM Impreza WRX
DCCD A.P. automatic DCCD
STi MY04 + V- TY856WB4KA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
(35:65) Suretrac system
Limited MY04
automatic DCCD
JDM Impreza WRX DCCD A.P.
TY856WB4JA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 on 17 in. wheel
STi Spec C MY04 (35:65) Suretrac
only
JDM Impreza WRX Viscous
TY856WB6CA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 Open
STI MY05 (4kgf)
JDM Impreza WRX DCCD A.P. automatic DCCD
TY856WB6KA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
STi MY05 (35:65) Suretrac system
automatic DCCD
JDM Impreza WRX DCCD Helical
TY856WB6JA 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900 on 17 in. wheel
STi Spec C MY05 (35:65) LSD
only
Viscous A.P.
JDM Legacy S401 ?? 3.636 2.375 1.761 1.346 1.062 0.842 3.545 1.000 3.900
(4kgf) Suretrac
JDM Legacy GT
Viscous
Spec B(Tuned by TY856WBDAA 3.636 2.235 1.521 1.137 0.971 0.756 3.545 1.100 3.900 Open
(4kgf)
STI) MY06
JDM Legacy GT
Viscous
Spec B(Tuned by TY856WBDAD 3.636 2.235 1.521 1.137 0.971 0.756 3.545 1.100 3.900 Open
(4kgf)
STI) MY07

US Impreza WRX DCCD A.P. automatic DCCD

TY856WH3MA 3.636 2.375 1.761 1.346 0.971 0.756 3.545 1.000 3.900
STi MY04 (35:65) Suretrac system
US Impreza WRX DCCD A.P. automatic DCCD
TY856WH4MA 3.636 2.375 1.761 1.346 0.971 0.756 3.545 1.000 3.900
STi MY04 late (35:65) Suretrac system
automatic DCCD
US Impreza WRX DCCD Helical system / front
TY856WW6MA 3.636 2.375 1.761 1.346 0.971 0.756 3.545 1.000 3.900
STi MY05 (35:65) LSD axle design
change
US Impreza WRX
1.100 3.900 DCCD
STi MY06-07
UK/Europe/NZ
Viscous A.P.
Impreza WRX STi TY856WN*** 3.636 2.375 1.761 1.346 0.971 0.756 3.545 1.100 3.900
(4kgf) Suretrac
MY02-04
UK Impreza WRX
DCCD automatic DCCD
STi (incl. Spec-D) TY856WG7KA 3.636 2.375 1.521 1.137 0.891 0.707 3.545 1.100 3.900 Helical
(41:59) system
MY07
Rear Diff.
EUDM Impreza DCCD Helical
TY856UW1MA 3.636 2.375 1.521 1.137 0.971 0.756 3.545 1.100 3.900 Type:R180 3.545
WRX STi MY06-07 (41:59) LSD
- Torsen
5MT Transmission (Pull Clutch)

Center Front
Application
Model Range Trans. Code 1st 2nd 3rd 4th 5th 6th Rev. T.R. F.D. Diff. Diff. Notes
Notes
Type Type
JDM Legacy
RS & GT TY752VB2DA 3.454 2.062 1.448 1.088 0.825 4.111
(turbo) MY93
JDM Legacy
RS & GT Viscous
TY752VBAAA 3.454 2.111 1.448 1.088 0.825 4.111 Open
(turbo) MY94- (4kgf)
95
JDM Legacy
Viscous
RS & GT TY752VBBAA 3.454 2.062 1.448 1.088 0.825 3.9 Open
(4kgf)
(turbo) MY96
JDM Legacy
RS & GT
TY752VBCBA 4.111
(turbo) MY97-
98
JDM Legacy
GT B-Spec
MY97-98 TY752VBCAA 4.444
(also GT B-
Spec Limited)
JDM Legacy
RSK & GT-B TY754VBAAA 4.444
MY99-00
JDM Legacy Viscous Helical
TY754VBACA 4.444 w/ LSD option
RSK MY00 (4kgf) LSD
JDM Legacy
Viscous
RSK & GT-B TY754VBBBA 4.111 Open
(4kgf)
MY01
JDM Legacy
Viscous Helical
RSK & GT-B TY754VBBCA w/ LSD option
(4kgf) LSD
MY01
JDM Forester Viscous
TY753VB1AA 4.444 Open
Turbo MY98 (4kgf)
JDM Forester
Viscous
Turbo MY99- TY755VB1AA 3.454 1.947 1.366 0.972 0.738 3.333 1.000 4.444 Open
(4kgf)
00
JDM Forester
Viscous
Turbo MY01- TY755VB2AA 4.444 Open
(4kgf)
02
JDM Impreza Viscous
TY752VB3AA 3.454 2.062 1.448 1.088 0.825 3.416 1.000 4.111 Open
WRX MY93 (4kgf)
JDM Impreza
Viscous
WRX RA TY752VB3BA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 4.111 Open
(4kgf)
MY93
JDM Impreza Double-cone
WRX, WRX Viscous 2nd synchro
TY752VB3CA 3.454 2.062 1.448 1.088 0.825 3.416 1.000 4.111 Open
SA, WRX STi (4kgf) added (all turbo
MY94 Impreza)
Double-cone
JDM Impreza
Viscous 2nd synchro
WRX RA TY752VB3DA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 4.111 Open
(4kgf) added (all turbo
MY94
Impreza)
JDM Impreza
Viscous
WRX & WRX TY752VB3FA 3.454 2.062 1.448 1.088 0.825 3.416 1.000 4.111 Open
(4kgf)
STi MY95
JDM Impreza
Viscous
WRX RA TY752VB3EA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 3.900 Open
(4kgf)
MY95
JDM Impreza
DCCD DCCD fitted off
WRX STi RA TY752VB3EA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 3.900 Open
(35:65) production line
MY95
JDM Impreza Viscous
TY752VB4AA 3.454 2.062 1.448 1.088 0.825 3.416 1.000 4.111 Open
WRX & WRX (4kgf)
STi Ver.II
MY96
(also STi 555
& WRX V-
Limited Ver.II
MY96)
JDM Impreza Double-cone 3rd
Viscous
WRX RA TY752VB4BA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 3.900 Open synchro added
(4kgf)
MY96 (RA only)
JDM Impreza
WRX STi RA
Ver.II MY96 Double-cone 3rd
DCCD
(also STi RA TY752VB4CA 3.454 2.333 1.750 1.354 0.972 3.416 1.000 3.900 Open synchro added
(35:65)
555 & RA V- (RA only)
Limited Ver.II
MY96)
JDM Impreza
WRX & WRX
Double-cone 3rd
STi Ver.III/IV
Viscous synchro added
MY97-98 TY752VBCAA 3.166 1.882 1.296 0.972 0.738 3.416 1.000 4.444 Open
(4kgf) (all turbo
(also STi V-
Impreza)
Limited Ver.III
MY97)
JDM Impreza
WRX RA Widened
MY97-98 Viscous 1st,2nd,3rd gears
TY752VB5BA 3.083 2.062 1.545 1.151 0.825 3.416 1.000 4.444 Open
(also WRX R (4kgf) (EJ20K models
V-Limited only)
MY98)
JDM Impreza
WRX STi R &
Widened
RA Ver.III/IV
DCCD 1st,2nd,3rd gears
MY97-98 TY752VB5CA 3.083 2.062 1.545 1.151 0.825 3.416 1.000 4.444 Open
(35:65) (EJ20K models
(also STi Type
only)
R V-Limited
Ver.IV MY98)
Double-cone 3rd
JDM Impreza
Viscous synchro added
WRX 5-Door TY752VBCBA 3.454 2.062 1.448 1.088 0.825 3.416 1.000 4.111 Open
(4kgf) (all turbo
MY97-98
Impreza)
JDM Impreza
WRX STi RA DCCD Helical
TY752VB6EA 3.083 2.062 1.545 1.151 0.825 3.416 1.000 4.444
V-Limited (35:65) LSD
Ver.IV MY98
JDM Impreza DCCD Specially treated
TY752VB6SZ 3.083 2.062 1.545 1.151 0.825 3.416 1.000 4.444 Open
22B (35:65) gear components
Bellhousing
JDM Impreza attachment
Viscous
WRX 5-Door TY754VB1AA 3.454 2.062 1.448 1.088 0.825 3.333 1.000 4.111 Open changed from 4
(4kgf)
MY99-00 to 8 points (all
models)
Bellhousing
JDM Impreza
attachment
WRX RA & Viscous
TY754VB1BA 3.083 2.062 1.545 1.151 0.825 3.333 1.000 4.444 Open changed from 4
RA Ltd. (4kgf)
to 8 points (all
MY99-00
models)
Bellhousing
JDM Impreza
attachment
WRX & WRX Viscous
TY754VBAAA 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open changed from 4
STi Ver.V/VI (4kgf)
to 8 points (all
MY99-00
models)
JDM Impreza
WRX STi R &
RA Ver. V/VI Helical FLSD
MY99-00 DCCD avail. 12/99 on
TY754VB1CA 3.083 2.062 1.545 1.151 0.825 3.333 1.000 4.444 Open
(also STi RA (35:65) regular STi RA
Limited Ver. (optional?)
V/VI MY99-
00)
JDM Impreza
DCCD Helical
WRX STi RA TY754VB1EA 3.083 2.062 1.545 1.151 0.825 3.333 1.000 4.444
(35:65) LSD
Ltd. Ver. V/VI
MY99-00
(also S201)
JDM Impreza
WRX NB, Viscous
TY754VBBAA 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open
NBR, 20K (4kgf)
MY01-02
Reverse
engagement &
JDM Impreza Viscous
TY754VB4AA 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open 3rd/4th synchro
WRX MY03 (4kgf)
revised (see US
info)
JDM Legacy
2.0GT MY04 Viscous
TY757VBAAB 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.111 Open
(also US (4kgf)
Legacy 2.5GT)
JDM Legacy
Viscous This is Push not
2.0GT spec B TY757VBABB 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open
(4kgf) Pull
MY04
JDM Legacy Viscous dual cone 1st
TY757VBBAB 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.111 Open
2.0GT MY05 (4kgf) synchro
JDM Legacy
Viscous dual cone 1st
2.0GT spec B TY757VBBBB 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open
(4kgf) synchro
MY05

Rev. pop-out
US Impreza Viscous addressed 10/01
TY754VN2AA 3.454 1.947 1.366 0.972 0.738 3.333 1.100 3.900 Open
WRX MY02 (4kgf) (main shaft, rev.
idler, rev. hub)
1st, 2nd, 3rd
US Impreza Viscous
TY754VN2BA 3.454 1.947 1.366 0.972 0.738 3.333 1.100 3.900 Open gears widened to
WRX MY03 (4kgf)
RA width
3rd synchro stop
US Impreza Viscous changed (3/4
TY754VV4AA 3.454 1.947 1.366 0.972 0.738 3.333 1.100 3.900 Open
WRX MY04 (4kgf) gear, synchro
assy)
US Impreza 1st synchro
Viscous
WRX late TY754VV5AA 3.454 1.947 1.366 0.972 0.738 3.333 1.100 3.900 Open changed to dual
(4kgf)
MY04 cone type?
US Impreza Viscous
TY754VZ6AA 3.454 1.947 1.366 0.972 0.738 3.333 1.100 3.900 Open
WRX MY05 (4kgf)
US Forester Viscous Gears as per
TY755VH4AA 3.454 1.947 1.366 0.972 0.738 3.333 1.000 4.444 Open
XT MY04 (4kgf) WRX revisions
US Forester Viscous
TY755VW5AA 3.454 1.947 1.366 0.972 0.738 3.333 1.000 4.444 Open
XT MY05 (4kgf)
US Baja Turbo Viscous Gears as per
TY754VHEAA 3.454 1.947 1.366 0.972 0.738 3.333 1.000 4.444 Open
MY04 (4kgf) WRX revisions
US Legacy
Viscous
Turbo GT TY757VBAAB 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.111 Open
(4kgf)
MY05
US Legacy
Turbo Outback TY757VWAAB
MY05

UK/Europe
Impreza Turbo
2000 / GT Viscous
TY752VN3BA 3.454 1.947 1.366 0.972 0.738 3.416 1.000 4.111 Open
MY94-95 (4kgf)
(also Series
McRae MY95)
Aus. Impreza Viscous
?? 3.454 1.947 1.366 0.972 0.738 3.416 ?? 3.900 Open
WRX MY94 (4kgf)
UK/Europe
Impreza Turbo Viscous
TY752VN4BA 3.454 1.947 1.366 0.972 0.738 3.416 1.000 4.111 Open
2000 / GT (4kgf)
MY96
UK/Europe
Impreza Turbo
TY752VN5BA
2000 / GT
MY97
UK/Europe
Impreza Turbo Viscous
TY752VN6AA 3.454 1.947 1.366 0.972 0.738 3.416 1.100 3.54 Open
2000 / GT (4kgf)
MY98
UK/Europe
Impreza Turbo Viscous
TY754VN1AA 3.454 1.947 1.366 0.972 0.738 3.416 1.100 3.54 Open
2000 / GT (4kgf)
MY99
UK Prodrive Viscous
?? 3.166 1.882 1.296 0.972 0.738 3.333 1.000 4.444 Open
P1 MY99 (4kgf)
5MT Transmission (Push Clutch)

Center Diff. Front Diff. Application

Model Range Trans. Code 1st 2nd 3rd 4th 5th 6th Rev. T.R. F.D. Notes
Type Type Notes
US Impreza (FWD)
TY752VX3AA 3.636 1.950 1.344 0.972 0.780 N/A 3.583 1.000 4.111 N/A Open
MY93-94
US Impreza (AWD) Viscous
TY752VX3AA 3.545 1.947 1.366 0.972 0.780 N/A 3.416 1.000 4.111 Open
MY93 (4kgf)
US Impreza (AWD) Viscous
TY752VX3CA 3.545 1.947 1.366 0.972 0.780 N/A 3.416 1.000 4.111 Open
MY94 (4kgf)
US Impreza (AWD) Viscous
TY752VT3AA 3.545 2.111 1.448 1.088 0.825 N/A 3.416 1.000 3.900 Open
MY95 (4kgf)
US Impreza EJ18E Viscous
TY752VT4AA 3.545 2.111 1.448 1.088 0.825 N/A 3.416 1.000 3.900 Open
(AWD) MY96 (4kgf)
US Impreza EJ22E Viscous
TY752VABBA 3.545 1.947 1.366 0.972 0.780 N/A 3.330 1.000 3.900 Open
(AWD) MY96 (4kgf)
US Impreza EJ18E Viscous
TY752VT5AA 3.545 2.111 1.448 1.088 0.825 N/A 3.416 1.000 3.900 Open
(AWD) MY97 (4kgf)
US Impreza EJ22 (AWD) Viscous
TY752VA5AA 3.545 1.947 1.366 0.972 0.780 N/A 3.333 1.000 3.900 Open
MY97-98 (4kgf)
Viscous
US Impreza 2.5RS MY98 TY752V2DAA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Impreza EJ22 MY99- Viscous
TY754VA1AA 3.545 1.947 1.366 0.972 0.780 N/A 3.333 1.000 3.900 Open
01 (4kgf)
US Impreza 2.5RS Viscous
TY754VCAAB 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
MY99-01 (4kgf)
Viscous
US Impreza RS MY02-03 TY754VFBBA 3.545 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Impreza TS MY02-03 TY754VC2AA 3.545 2.062 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
(4kgf)
US Impreza Outback Viscous
TY754VC2BA 3.545 2.062 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
MY02-03 (4kgf)
Viscous
US Impreza RS MY04 TY754VC4CA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Impreza TS MY04 TY754VC4AA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
(4kgf)
US Impreza Outback Viscous
TY754VC4BA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
MY04 (4kgf)
Viscous
US Forester MY97-98 TY753VJ1AA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Forester MY99 TY755VC1AA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Forester MY00-01 TY755VC1BA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Forester MY02 TY755VC2AA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Forester MY03 TY755VC3AA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Forester (non-turbo) Viscous
TY755VC4AA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
MY04 (4kgf)
US Legacy Outback Viscous
TY752VAAAA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 4.111 Open
MY94 (4kgf)
Viscous
US Legacy MY94 TY752VAABA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 3.900 Open
(4kgf)
Viscous
US Legacy MY95-96 TY752VABAA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 3.900 Open
(4kgf)
US Legacy Outback Viscous
TY752VABBA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 4.111 Open
MY95-96 (4kgf)
Viscous
US Legacy GT MY97 TY752VACBA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 4.111 Open
(4kgf)
US Legacy Outback Viscous
TY752VACCA 3.545 2.111 1.448 1.088 0.871 N/A 3.416 1.000 4.111 Open
MY97 (4kgf)
Viscous
US Legacy MY97-98 TY752VACAA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 3.900 Open
(4kgf)
Viscous
US Legacy GT MY98 TY752V2DAA 3.545 2.111 1.448 1.088 0.780 N/A 3.416 1.000 4.111 Open
(4kgf)
US Legacy Outback Viscous
TY752V2DCA 3.545 2.111 1.448 1.088 0.871 N/A 3.416 1.000 4.111 Open
MY98 (4kgf)
Viscous
US Legacy MY99 TY754VAAAA 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
(4kgf)
Viscous
US Legacy GT MY99 TY754VCAAB 3.545 2.111 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Legacy Outback Viscous
TY754VCACB 3.545 2.111 1.448 1.088 0.871 N/A 3.333 1.000 4.111 Open
MY99 (4kgf)
US Legacy L & Brighton Viscous
TY754VCAAA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
MY00 (4kgf)
US Legacy Outback Viscous
TY754VCACA 3.454 2.062 1.448 1.088 0.871 N/A 3.333 1.000 4.111 Open
MY00 (4kgf)
Viscous
US Legacy GT MY00 TY754VCADA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Legacy L & Brighton Viscous
TY754VCBAA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 3.900 Open
MY01-02 (4kgf)
Viscous
US Legacy GT MY01-02 TY754VCBDA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Legacy Outback Viscous
TY754VCBCA 3.454 2.062 1.448 1.088 0.871 N/A 3.333 1.000 3.900 Open
MY01-02 (4kgf)
Viscous
US Legacy L MY03-04 TY754VCDAA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
Viscous
US Legacy GT MY03-04 TY754VCDDA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 4.111 Open
(4kgf)
US Legacy Outback & Viscous
TY754VCDCA 3.454 2.062 1.448 1.088 0.871 N/A 3.333 1.000 4.111 Open
Baja MY03-04 (4kgf)
Viscous
US Legacy & Baja MY05 TY757VCACA 3.454 2.062 1.448 1.088 0.871 N/A 3.333 1.000 4.111 Open
(4kgf)
US Legacy Outback Viscous
TY757VWAAB 3.454 2.062 1.448 1.088 0.871 N/A 3.333 1.000 4.111 Open
MY05 (4kgf)
Viscous
EU Legacy 2.0 MY95-97 TY752XRAAA 3.545 2.111 1.448 1.088 0.871 N/A 3.416 1.000 3.9 Open
(4kgf)
EU Legacy 2.2 sedan Viscous
TY752VG1AA 3.545 2.111 1.448 1.088 0.871 N/A 3.416 1.000 3.9 Open
MY90 (4kgf)
Viscous
EU Legacy 2.0 MY95 TY752XRBAA N/A 1.000 3.9 Open
(4kgf)
EU AU Legacy OBK 99- Viscous
TY754XKAAA 3.545 2.111 1.448 1.088 0.825 N/A 3.333 1.000 4.11 Open
2002 (4kgf)
Viscous
US Impreza MY05 TY754VC6AA 3.454 2.062 1.448 1.088 0.780 N/A 3.333 1.000 3.90 Open
(4kgf)
How To Break-in Your Newly Built Engine
v1.03 - https://www.cobbtuning.com/support/how-to-break-in-your-newly-built-engine-v1-03/

1. Make sure that all engine assembly is performed in a clean, contaminant free, and
conditioned (temperature and humidity controlled) environment. All engine hardware
must be perfectly clean, free of contaminants, and properly lubricated prior to
installation. Verify that all intake plumbing is clean and free of debris.
2. Check that all of the proper fluids are at the proper levels in the proper areas of the
engine, especially engine oil and filter.
3. Verify that the timing belt has been properly installed and the camshafts/camshaft gears
are properly timed.
4. Again, check that all of the proper fluids are at the proper levels in the proper areas of the
engine, especially engine oil and filter.
5. Remove the spark plugs and disconnect the fuel injectors, turn engine over 3-4 times for
20 seconds each time to build oil pressure. If you have an oil pressure gauge you can see
the pressure go up after a few cranks, this insures you have oil throughout the engine and
peripherals (turbo, camshaft phasing mechanisms, etc.) prior to start-up. You can also
remove the oil drain line from the turbocharger to make sure the turbo is primed with oil
as well. Ideally, you will want to install the oil pressure gauge as far from the oil pump as
possible so you can see you are getting pressurized engine oil in every part of the engine.
6. Install spark plugs, start the engine up (on a known, good ECU calibration) with more
people than you need…have them watch under vehicle, in engine bay, out exhaust etc.
listening for mechanical noises and watching for obvious fluid leaks.
7. Watch oil pressure gauge the whole time, keep an eye on the engine watching for any
fluid leaks. Allow the engine to come to temperature (coolant) and watch oil pressure to
come down to a stable level at idle. This will happen after the RPM come down from
cold start and the oil thins out from the heat.
8. Easily drive the engine around (on a known, good ECU calibration) for 50-100 miles then
change engine oil and filter. Do your best vary the engine RPM and decelerate the engine
under vacuum as frequently as you accelerate the engine, you want your engine to break-
in evenly. The filter is critical to replace since it is going to contain several thousands of
particles of metal and coating particulates from the ring/cylinder wall and bearing/friction
surface break-in. Take the oil pan out in broad day light and look for a glittery material to
be in the oil (make sure the oil container is absolutely clean prior to allowing the engine
oil to drain into the container). No large debris should be present in the oil. Never use
synthetic engine oil during this process, use a straight weight, non-synthetic, non-
detergent engine oil. You want friction at this point; use an SAE30 oil, non-synthetic, no
detergents, etc.
0-100 miles, run engine very lightly, no more that 1-2 psi boost, no more than 28% TPS and
3500 RPM, change engine oil and filter. Do your best to vary the engine RPM and decelerate the
engine under vacuum as frequently as you accelerate the engine, you want your engine to break-
in evenly.

100-500 miles, run engine very lightly, no more that 3-5 psi boost, no more than 40% TPS and
4500 RPM Change engine oil and filter. Look for more glittery material in oil. Do your best to
vary the engine RPM and decelerate the engine under vacuum as frequently as you accelerate the
engine; down shift the vehicle to come to a stop rather than using just the brakes.

500-1000 miles, run engine more aggressively, no more that 6-8 psi boost, no more than 50%
TPS and 5000 RPM Change engine oil and filter. Look for less glittery material in oil. Datalog
the engine, look for knock events, etc.

Slowly raise the rev limit throughout break-in all the way to redline just after 1000 miles. Please
be sure to vary the RPM range throughout engine break-in…you want to do what you can to
promote even break-in of the engine hardware. Ideally, the engine will optimally perform
throughout the RPM range. Engine ring sealing can be tested throughout the engine break-in
process with a compression gauge and cylinder leak-down test kit (CLT); you can test at TDC
and BDC. Be sure to perform these tests on a warm motor with the throttle blade fully open. You
would want your measured compression to be within 20psi of each other on all cylinders and the
CLT to 7% or less across all cylinders. We hope this helps.
The NASIOC Oil Selection Guide
by gpshumway of NASIOC forums

1. Introduction
Standard Disclaimer:
This guide is not an official Subaru document, and does not represent direct technical advice
from the author, it is opinion and for information only. If you have an engine failure after reading
this, don't blame me or anyone else at NASIOC.

TL;DR Readers:
Readers with A.D.D. are advised to skip to the engine specific recommendations and specific
oils sections at the end of this guide, but be warned, forum members and moderators will not
treat kindly new threads asking questions which are answered here.

Ron (Unabomber) has done a wonderful job with all of his FAQs including the one on oil.
Despite Ron’s best efforts we continue to see frequent “best oil” and “how good is this oil?”
threads. With the advent of the FA and FB series of engines from Subaru and their substantially
different design from the older EJ engines, it seems time to provide some updated information
and a basis for oil selection in different applications and driving conditions. In his FAQ Ron
offers his opinion as to which oils are “best” if you were to press him on the subject. While there
are very good reasons for his selections, Ron did not enumerate them in his FAQ. This guide is
not intended to supplant Ron’s information, but augment it by providing a detailed discussion of
motor oil and its properties. It is the intent of this guide to provide the background information
required for a reasonably competent layperson to select a “good” oil for their application. This
process consists mostly of understanding what categories exist for motor oil and selecting an oil
from the right category, the performance differences between oils in the same category are
generally so small as to be immeasurable outside a laboratory. Because of this it is impossible for
laypersons to determine which oil is “best”. Even professionals with the tools and knowledge to
do such analysis generally don't have the time or resources. If you take nothing else from this
guide, it should be to stop focusing on brand and “best” so you can think more about category
and “good”.

This guide is a combination of information gleaned from various internet sources, industry
organizations and companies involved with making and testing motor oil. I am not a lubrication
engineer or tribologist, merely a technically savvy enthusiast. Much of this guide consists of my
opinion which is based upon the imperfect information available to the general public. As such,
reasonable people will undoubtedly disagree with my assessments. That’s fine, I’m certainly
willing to learn from others and hopefully fill some of the large void of information surrounding
motor oil. Your comments are welcome and appreciated, but please keep it civil and remember
that much of the available information is imperfect at best. Please feel free to point out any
inaccuracies or glaring omissions.

We will start by discussing the basics of engine lubrication and the physical properties of motor
oil, move to the specifications which control those properties and define the categories modern
oils fit into. We'll briefly discuss used oil analysis as a tool for confirming and fine tuning a
selection, and finally we'll examine what Subaru recommends for their cars and use all that
information to select the correct category of oil for our unique situation. This guide will
generally illustrate how to select oils based upon technical requirements, and will only obliquely
address concerns over warranty coverage. Subaru's recommendations for oil will be the starting
point for selection, but some reading between the lines of the simplistically written owner's
manuals will be required. For most users technical and warranty requirements coincide nicely,
but for many owners of performance Subarus there is some cause for concern. We will address
these concerns in a specific section on warranty requirements.

2 Engine Lubrication Basics

In order to select good oil for our application we must first understand its function within the
engine. Modern four stroke automotive piston engines have an ***8220;active***8221;
lubrication system consisting of a positive displacement oil pump and passageways throughout
the engine which deliver oil from the pump to the engine’s functional elements. This oil system
is an ancillary system in the engine, a vital one for sure, but not essential to the engine’s primary
function. It is entirely possible to design an engine without an oil pump, two-stroke outboard
motors and four-stroke lawnmowers are two examples. An oil’s ability to be drawn from the pan
and pumped through the engine are primary concerns of the specifications governing motor oil
which we’ll discuss later.

Oil’s first and foremost job is to prevent wear within the engine, it does this in two ways, by
providing a hydrodynamic film which completely separates moving parts from each other, and
by providing chemical anti-wear compounds which prevent wear between parts in contact. These
compounds take the form of additives dissolved in the oil and are largely independent of the oil’s
viscosity (thickness), which determines its hydrodynamic properties. Which anti-wear
mechanism is functioning depends on the state of lubrication at a particular interface inside the
engine. Hydrodynamic lubrication has already been mentioned, the two other states are boundary
and mixed-film. In boundary lubrication the surfaces are in contact and the anti-wear additives in
the oil are primarily responsible for preventing wear. Mixed film is - you guessed it- the
transition condition between boundary and hydrodynamic.

You can read more here: http://www.stle.org/resources/lubelearn/lubrication/

Contrary to popular belief, oil pressure supplied by the oil pump does not provide hydrodynamic
lubrication in the engine. The hydrodynamic wedge which separates a rotating shaft from its
journal bearing is a result only of the oil’s viscosity and the rotational speed of the shaft. As long
as the oil pump supplies enough oil to keep the bearing full of oil, no additional pressure is
necessary. Think of a water-skier or a hydroplaning tire, the water in each case is not supplied
under pressure, yet both the skier and the tire are suspended above the solid surface below.

Why do we worry about oil pressure if it doesn't indicate lubrication quality then? Well, proper
oil pressure is an easy to measure indication of a healthy engine and oil system, if it falls outside
of normal parameters, it indicates that something is wrong. If it drops precipitously, something is
very wrong. Oil pressure is a result of the oil’s viscosity, flow rate, and the size of the orifices
which it is being forced through, and viscosity is highly dependent on temperature, so slightly
low oil pressure can be an indication that the oil is too hot. Oil pressure also drops as the
clearances in the engine increase through wear, particularly the clearances within the oil pump
itself. Oil pressure which drops precipitously indicates a failure of the lubrication system, most
often the pickup becomes uncovered (usually through high-g cornering) and the system draws air
instead of oil, but the pressure can also drop quickly due to a sudden mechanical failure like a
broken oil pump drive, or an oil gallery plug that comes out. If the 2psi oil pressure idiot light in
your Subaru turns on, you have only a couple of seconds to turn the engine off or face
catastrophic failure. Note: don't confuse the oil pressure idiot light with the oil level idiot light on
newer cars. The implications of the level light are not nearly as severe.

While not essential to hydrodynamic lubrication, oil pressure may be necessary to activate
variable valve timing mechanisms or other functions within the engine, in this case the oil is
functioning as a hydraulic fluid, not a lubricant, which brings up the topic of secondary
functions. In addition to lubrication, the oil also must also be: a coolant, a corrosion inhibitor, a
detergent, an acid neutralizer and sometimes a hydraulic fluid. Most of these functions are not
handled solely by the oil itself, but by compounds dissolved in it as part of the additive package.

Main crankshaft bearings, big end connecting rod bearings and camshaft bearings are designed
for hydrodynamic lubrication, but many interfaces within the engine are reciprocating, not
rotating, and therefore do not generate the consistent velocity required for hydrodynamic
lubrication. These interfaces experience mixed-film or boundary lubrication conditions, and rely
on the oil's anti-wear additives to prevent wear. Despite the seemingly undesirable lubrication
condition, these components last for hundreds of thousands of miles in a modern engine.
Engineers think about states of lubrication as regions of the Stribeck curve (fig 1), named for the
engineer who invented it. The vertical axis is friction and the horizontal is either speed (assuming
constant viscosity) or viscosity (assuming constant speed). Because speed is a function of the
engine's bearing sizes and RPM, set by the designers, we'll concern ourselves only with Stribeck
curves with viscosity as the independent variable. When selecting an oil for your Subaru, the
shape of the curve at the transition between hydrodynamic and mixed-film regimes are important
to keep in mind.

Figure 1 The Stribeck Curve

1 - Boundary lubrication
2 - Mixed film lubrication
3 - Hydrodynamic lubrication
3 Motor Oil Properties
When selecting oil we're mostly selecting an oil with the best physical properties for our
application, which of course requires defining those properties. Motor oil behaves differently
under variations of temperature and mechanical stress, so a wide variety of tests are used to fully
characterize its physical properties. Governing specifications place limits on these properties
which an oil must meet, and vehicle manufacturers call out these specifications in owner's
manuals as requirements of proper maintenance.

3.1 Viscosity

We've already talked a little about viscosity, and most car enthusiasts have some idea what it is;
explicitly, viscosity is the scientifically measured “thickness” of a fluid. The stereotypical
molasses in January has a very high viscosity, while water has a low viscosity. Oil viscosity is
measured in several ways, dynamic, kinematic, and at varying temperatures. This is done
because the viscosity properties of oil depend greatly on temperature and the physical stress
imposed. Ideal “Newtonian” fluids have very predictable viscosity properties which can be
calculated easily, but motor oil is a non-Newtonian fluid so engineers measure its viscosity under
varying conditions in order to develop a full picture of how it will perform in service. Motor oil
is generally a shear-thinning non-Newtonian fluid, meaning it has a lower viscosity in fast
moving parts than in slow moving parts. This Wikipedia entry gives an overview of different
types of viscosity and what makes a fluid Newtonain. http://en.wikipedia.org/wiki/Viscosity

Dynamic Viscosity
Dynamic viscosity is the most basic measure of a fluid’s thickness, it indicates the fluid’s
resistance to shearing. The theoretical model for dynamic viscosity is two large plates of material
separated by a thin film of fluid, one plate is stationary, the other moving. The amount of force
required to keep the moving plate at a particular velocity determines the dynamic viscosity. A
detailed explanation of dynamic viscosity is not in the scope of this guide, but the above
Wikipedia entry gives an explanation.

Kinematic Viscosity
Kinematic viscosity is a fluid’s dynamic viscosity divided by its density, it is convenient to
measure in gravity driven test fixtures. Motor oil’s kinematic viscosity is commonly measured at
100°C (KV100) and 40°C (KV40), the former is the basis for the SAE grades of motor oil we’re
all used to seeing (e.g. 5w30). In motor oil, kinematic viscosity is highly dependent on
temperature.
Viscosity Index (VI)
The viscosity index is a derivative of the oil’s KV100 and KV40 set against an arbitrary scale. It
gives an indication of an oil’s viscosity variation with temperature at moderate and high
temperatures. It does not characterize oil viscosity behavior at extremely low temperatures, nor
under high rates of shear. An oil with a higher viscosity index will show smaller variations in
viscosity with variations in temperature than an oil with a low viscosity index. Knowing the
KV100 and KV40 (and hence the VI) allows us to use the standard mathematical model to
approximate an oils kinematic viscosity across a range of temperatures from about 32°F to
300°F. Below 31°F the formation of crystals can cause the oil to increase viscosity much faster
than the equations would predict, and extrapolating beyond 300°F is unwise.

An online calculator for viscosity index is available here:

http://www.widman.biz/English/Calculators/VI.html

High Temperature-High Shear Viscosity (HT/HS)

HT/HS is a measure of dynamic viscosity, it represents the oil’s thickness within the vital
hydrodynamic bearings of an engine. It is the most important measure of how well the oil will
perform as a hydrodynamic lubricant. HT/HS is measured at 150°C (302°F) which is a worst-
case temperature for most engines. The higher the HT/HS, the stronger the hydrodynamic film in
the engine’s bearings. Thus, the correct HT/HS is more important than the KV100 (SAE Grade)
when selecting oil. If the oil formulator publishes their HT/HS it's best to ignore the grade and
KV100 and base your choice on the HT/HS. As we will see later, different industry specs
categorizing motor oil are based in large part on HT/HS and those specifications will tell us
much about the oil's HT/HS even if the formulator doesn't publish a specific value.

Low Temperature Cranking Viscosity (CCS Viscosity)

Cranking viscosity is measured in a Cold Cranking Simulator (CCS) at a temperature
corresponding to the oil’s SAE Grade. The CCS measures the oil’s resistance to being pumped
through the engine when very cold. Oils with lower CCS viscosities require less power from the
starter to achieve sufficient RPM to start the engine - the lower the CCS viscosity, the easier the
engine will start at very cold temperatures. Depending on the SAE grade, CCS is measured
between -31 °F (0w) and 14°F (25w).

Low Temperature Pumping Viscosity (MRV Viscosity)

Pumping viscosity is measured in a mini-rotary viscometer (MRV), it indicates the oil’s
resistance to being drawn into the oil pump when very cold. Oils with lower MRV viscosity are
more easily drawn from the oil pan into the pump. This is important because oil which is too
hard to draw into the pump will cause cavitation and the pump will ingest air instead of oil. Air
makes a lousy hydrodynamic lubricant, so failed main bearings are the inevitable result of
cavitation. The lower the MRV viscosity the greater the safety margin against cavitation.
Pour Point
Pour point is a sort of viscosity. It is the lowest temperature at which the oil will flow under
prescribed conditions. You can think of this as the “freezing” temperature for oil as 0°C is the
freezing temperature of water. Water is a homogeneous substance composed only of H2O
molecules, so it has a definite freezing temperature (32°F), oil is a heterogeneous soup of
molecules varying in weight and composition, so it doesn’t turn solid at a definite temperature. It
never the less starts forming crystal structures at some temperature at which point further cooling
causes the viscosity to increase much faster. This point of accelerating viscosity increase is what
the pour point intends to measure. Pour point has largely been supplanted by MRV and CCS as a
measure of oil’s extreme cold performance, but many manufacturers still publish pour points for
their oils any way.

This Mobil 1 video shows a non-scientific pour point test:

https://www.youtube.com/watch?v=YnagT6wKfmY

Note this test is conducted with 5w30 oil at -40°C, which is much colder than 5w30 is intended
to be used in an engine.

Temporary Shear
When oil is subjected to high shear stress, as when sandwiched between the crankshaft and main
bearings, oil exhibits lower viscosity than would be predicted by measuring that viscosity in a
slow moving fixture. Once the oil leaves the high shear area and returns to the pan, its viscosity
returns to normal. This characteristic is known as temporary shear thinning, and the oil's
resistance to it is an important performance characteristic. The ratio between an oil's low shear
viscosity and its high shear viscosity is known as its temporary shear ratio. Oil formulators don't
publish the temporary shear ratio for their oils, and very few publish specifications which would
allow you to calculate it, but many oils have available specifications which allow one calculate
an approximation.

Shear Stability
Often confused with temporary shear is permanent shear. Some constituents of oil, particularly
Viscosity Index Improvers (VIIs), change their chemical makeup when subjected to repeated
cycles of physical shearing. The physical stress of shearing “clips” the longer molecules into
shorter ones, thus permanently reducing the oil's viscosity. In many oils designed to improve fuel
economy this effect can be quite pronounced, often reducing the oil's viscosity by 10% or more
and resulting in a reduction in viscosity grade from 30 to 20 or 40 to 30. An oil's resistance to
permanent shear is known as its shear stability. Motor oil certification specifications set
minimum levels of shear stability with the European ACEA specs being more stringent than the
American API/ILSAC specs.
3.2 Other Properties

Specific Gravity
This is the density of the oil relative to water (1=water), it is published by many manufacturers
on their product data sheets. A normal user has no use for it, but advanced users might use it in a
calculation to estimate the temporary shear ratio of an oil. Those calculations are beyond the
scope of this guide.

Flash Point
Flash point is the lowest temperature at which the oil will generate sufficient vapor to catch fire
if exposed to an ignition source. The flash point of virgin oil is compared to that of used oil in an
oil analysis. Contaminants in the oil like unburned fuel lower an oil’s flash point.

Blackstone Labs Explains Flashpoint here: (http://www.blackstone-labs.com/what-...lash-

point.php)

Volitility (NOACK)
The NOACK test measures the percentage of oil lost to evaporation at 250°C, a temperature
consistent with the highest temperatures seen at the top piston ring. Lighter constituent elements
in oil tend to evaporate first, leaving the remaining oil thicker, which can affect its cold weather
properties during extended use. Evaporated elements of the oil can also affect combustion when
burned. NOACK is given as a percentage of the oil which evaporates during the test. All other
things equal, lower NOACK oils will show less consumption and more stable viscosity.

Total Base Number (TBN)

TBN is a measure of the oils ability to neutralize acidic contaminants. Acidic byproducts of
combustion enter the oil via piston ring blow-by and must be neutralized before they damage
engine components or the oil’s base stock. The higher the total base number, the more acid the
oil can neutralize before being changed. Unfortunately, depletion of the TBN can be quite non-
linear, especially with new technology oils using non-traditional additives, so an oil with a higher
TBN may not allow for longer oil change intervals than one with a lower TBN.
4 Motor Oil Components - What’s in my oil?
he average user is neither a chemical engineer nor a tribologist, and therefore the actual chemical
makeup of motor oil is not particularly useful to them. So why write this section? Primarily to
debunk a lot of the misinformation surrounding what is good or bad about particular components
of motor oil. Motor oil is composed of two basic components, a base stock and an additive
package. Base stocks can be guessed reasonably accurately using the oil***8217;s Material
Safety Data Sheet (MSDS), but this is largely irrelevant to the end user because what we really
care about is the oil***8217;s ability to protect the engine, that ability is heavily influenced by
the oil***8217;s additive package, the constituents of which are closely guarded trade secrets.
As such, you will not find this information in the public space and even if you could, it would
require the knowledge of an experienced tribologist to evaluate. If you***8217;re such a person,
you should be writing this guide, not reading it. Nevertheless, we will briefly discuss the various
constituents of oil as they are useful to understand in certain circumstances.

Base Stock
The base stock of makes up 80% or more of an oil***8217;s total volume, they are classified by
the American Petroleum Institute (API) into five groups. The groups are defined by the types of
molecules which make up the base oil and by physical properties, they are designated by roman
numerals I-V. Groups I, II and III are hydrocarbon base stocks, group IV is poly-alpha-olefins
and group V is ***8220;everything else***8221; which in the context of motor oil usually
means esters. The exact chemical definitions of olefins, esters, etc is beyond the scope of this
guide.

Which groups are synthetic? Funny you should ask, ***8220;synthetic***8221; is a marketing
term, not a technical one. To the extent it could be considered technical it would mean that the
base stock was synthesized from component elements. In other words
***8220;synthetic***8221; is a process, not a physical property, and we***8217;re trying to
select finished oils with the right physical properties, we don't really care about the process used
to make them. One could theoretically synthesize a really crappy base stock with poor physical
properties, but nobody does because it***8217;s cheaper to make crappy base stocks by refining
crude oil. That said, we can make some reasonable generalizations.

Traditionally groups I and II are considered conventional petroleum oils while groups IV and V
are synthetic, group III is the subject of great debate, especially among tribologists with a degree
from Internet U. The latest group III base stocks made by the gas-to-liquids process certainly
qualify as synthetic. The process involves sticking methane molecules (CH4) together until you
get a hydrocarbon of the desired molecular weight, a textbook definition of synthesis. The
resulting physical properties are excellent and every bit a match for the properties of traditional
PAO or ester base stocks. The only area where PAO and ester stocks still hold an advantage over
group III is at extremely cold temperatures irrelevant to most users, think Prudhoe Bay oil field
workers and Antarctic researchers. Pennzoil products with ***8220;Pure Plus***8221; marketed
by Shell oil and some Mobil 1 products (Mobil***8217;s not talking) are examples of gas-to-
liquids Group III synthetic oils.

For a detailed look at base stock performance, see these (somewhat old) articles in Machinery
Lubrication:
http://www.machinerylubrication.com/...oil-technology

Viscosity Index Improvers (VIIs)

As their name suggests, these additives improve the viscosity index of the base stock, reducing
temperature's effect on the oil's viscosity. They are added to oil to help it meet cold start
requirements. Unfortunately they carry negative side effects, they lower the oil's temporary shear
ratio and they also tend to shear permanently, lowering the formulation's overall shear stability.
This graph from Chevron/Oronite for their Paratone line of of VIIs shows how their effectiveness
drops at high shear rates (temporary shear):

Figure 2 - https://www.oronite.com/paratone/shearrates.aspx
Also from Chevron/Oronite, this graphic shows temporary and permanent shear.

Figure 3 - https://www.oronite.com/paratone/shearloss.aspx

With modern synthetic base stocks it's possible to create 5w and 10w oils without any VIIs at all.
The larger the ***8220;spread***8221;, the harder this is to achieve, so a 10w30 oil is easier to
formulate without VIIs than a 5w40.

Pour Point Modifiers

Pour point modifiers work to prevent crystal structure formation at temperatures approaching the
oil's pour point, they have little effect at moderate and warm temperatures. They work in concert
with VIIs to help an oil meet its cold start requirements, but do not carry all the negative side
effects of VIIs.

Anti-wear Additives
Anti-wear additives are compounds dissolved in motor oil which form a sacrificial coating at the
surface of moving parts which prevents wear of the base material. Contact with oil continuously
refreshes this sacrificial layer, minimizing wear of engine parts in boundary and mixed film
lubrication regimes. Traditional anti-wear additives are compounds of zinc, phosphorous and
molybdenum. Newer anti-wear additives are compounds of antimony, boron, titanium, ceramics,
and even organo-metallic compounds. The exact chemical forms and properties of these are
beyond the scope of this guide, but it***8217;s important to understand that they all require
temperature to function. Their ineffectiveness at low temperatures is the source of the common
knowledge that most engine wear occurs from cold starts. Many users believe the more anti-wear
additives, the better, but increasing their concentration beyond a certain point does not reduce
wear, and can lead to a reduction in the effectiveness of detergent additives and subsequently to
sludge deposits.

http://www.bobistheoilguy.com/forums...mony_Additives
http://www.bobistheoilguy.com/forums...pics/1040691/1
Friction Modifiers
Friction modifiers are used in oils designed to improve fuel economy. Normally friction and
wear are closely linked, but these additives don't really reduce wear. Some are multifunction
additives like Molybdenum, which in high concentrations acts as a friction modifier. They can
also interfere with the function of certain components, like the wet clutches in motorcycles, so
they are specifically forbidden in certain oils.

http://www.infineum.com/Documents/Cr...ogy%202009.pdf

Anti-Oxidant Additives
Anti-oxidant additives help prevent the oil’s base stock from oxidizing and becoming thicker
with use, which would hurt cold weather properties. Anti-wear compounds of Zinc and
Phosphorous also function as anti-oxidants, their anti-oxidation function is the reason some oils
have high concentrations of these additives. Modern base stocks are much more resistant to
oxidation than their forebears, so less of these additives are required today.
http://www.bobistheoilguy.com/forums...ltifunctional_
http://www.bobistheoilguy.com/forums...ati#Post729355
http://www.chemeng.titech.ac.jp/~mma...tpindex_e.html

Detergents
These compounds prevent byproducts of combustion and degraded engine oil from turning into
sludge deposits inside the engine. They break up and suspend the harmful compounds until the
next oil change. Detergents are commonly compounds of Magnesium or Calcium. Detergents
can compete with anti-wear additives for the surface of important components, so balancing
them in the overall formula is important for best performance. When oil is ingested into the
engine, the detergents can have a negative impact on the octane of the intake charge (they
promote detonation). This is the reason oils marketed as pure racing oils have low levels of
detergent and must be changed frequently. Detergents are often multifunctional serving also to
neutralize acids.

Acid Neutralization Additives

The TBN property of motor oil is a result of acid neutralization additives. Some byproducts of
combustion are acidic and those which end up in the sump through piston ring blow-by would
attack the engine's metal components if not neutralized. The high degree of control afforded by
modern fuel injection systems and the purity of modern fuels (particularly low Sulphur) have
made high levels of these additives less necessary. Some acid neutralizers also function as
detergents.
Anti-Foaming agents
Some oils contain additives which help prevent the formation of small bubbles via the stirring
action of the moving engine parts. Oil that’s been whipped into meringue by the crankshaft
makes a lousy lubricant. These additives are often silicon compounds.
5. Specifications Governing Oil
To ensure they will meet the needs of a modern engine, today's oils must meet testing and
qualification criteria laid out in a myriad of industry specifications overseen by several industry
groups. These specifications often specify minimum requirements, but no maximum, which
makes it possible for several specifications to be met by the same oil. Understanding how various
industry specifications overlap and combine to define how an oil will perform is the key to
selecting the right oil for your application without being distracted by oil formulator's marketing
messages.

5.1 SAE J300 (Viscosity Grade)

Society of Automotive Engineers (SAE) specification J300 is the source of the familiar grade
designations we’re all used to seeing on motor oil (e.g. 5w30). All higher level industry
specifications call upon J300 to define viscosity grades. Oil grades with a “w” in the middle are
referred to as multi-grade oils, their low temperature performance is superior to the older straight
grade (e.g. SAE 30) oils, which became very thick at low temperatures. The number before the
“w” is the low temperature grade (w = winter) while the number after the “w” is the oil’s high
temperature grade, sometimes referred to as the “weight” (e.g. 30 weight oil). Multi-grade oils
must meet requirements for CCS viscosity, MRV viscosity, kinematic viscosity at 100°C
(KV100) and HT/HS viscosity. Straight grade oils have to meet either MRV and CCS
requirements, or HT/HS requirements, but not both. This handy table from Infineum summarizes
the requirements of SAE J300. The table is from their larger oil classifications guide.

Figure 4 - http://www.infineum.com/media/80723/...tions-2013.pdf

We'll ignore straight grades here as they would be dangerous to use in many circumstances, and
modern base stocks and additives make multi-grade oils ubiquitous and high quality. SAE J300
was first written in the 1930s, a time when straight grades were often the best choice. You will
notice that what defines a particular high temperature grade is KV100 in a particular range along
with a minimum HT/HS viscosity (High-Shear-Rate Viscosity from the table). So, while the
SAE grades provide useful general information about the oil’s viscosity properties, it’s possible
for two very similar oils to be a different grade, a thick 30 grade can be very similar to a thin 40
grade. J300 splits oil grades into two general categories, “light duty” oils with HT/HS below 3.5,
and “heavy duty” oils with HT/HS above 3.5, placing the split between 30 and 40 grade, but
other specifications (ACEA) define a category of heavy duty 30 grade oils with HT/HS above
3.5.

The other thing to notice is the temperature requirements for the w number portion of multi-
grade oils. The actual viscosity requirements for the w number are not that important, they’re
arranged so an average car will start safely at the associated temperature, the temperatures
therefore are the important bit. What they tell us is how cold a given grade of oil can be used
safely. The cranking viscosity temperature represents a conservative minimum use temperature.
So, 10w oils can be used down to -25°C (-13°F), 5w oils down to -30°C (-22°F), and 0w oils
down to -35°C (-31°F – Brrr!). Also notice that the pumping viscosity requirements are always
set 5°C lower than the cranking requirements, meaning the oil is always capable of being drawn
into the oil pump if the starter is capable of turning the engine over, providing safety margin
against cavitation.

There are many important properties not defined in J300 however: chemical properties,
compatibility with seals and catalytic converters, and requirements for used engine oil. It is
strictly a set of viscosity requirements for new oil. Those additional requirements are left to other
specifications.

A note on grade designations; J300 specifies a dash be used in multi-grade oils (e.g. 5w-30) in
this guide I will omit the dash for the reason that many forum search engines treat a dash as a
space, and often refuse to search for terms shorter than three characters. So, searching for “5w-
30” can be a major pain because the search engine treats it as two separate two letter words,
“5w30” is easier to search for.

5.2 API Specifications

The American Petroleum Institute (API) defines a bewildering battery of requirements which
ensure API certified oils meet the needs of a modern engine, including chemistry limits for
beneficial and harmful constituents, tests for valvetrain wear, high temperature deposits, sludge
formation, compatibility with seals and a whole host of other requirements. For many engines an
oil of the appropriate SAE grade meeting the current API specifications is all that's needed to
ensure durable operation. There are other specifications which we will see later have more
stringent requirements which owners of high-performance cars may choose to seek in an effort to
use the best oil possible, but any oil meeting the appropriate API specification and SAE grade
will be a safe and effective lubricant.

The API specs are split into two groups, the service class (“S” series) for gasoline engines and
the commercial ("C” series) for diesel engines. The API coordinates their efforts with the
International Lubricant Standards Acceptance Committee (ILSAC) such that current versions of
the API gasoline standards with their optional Resource Conserving designation (formerly
Energy Conserving) are nearly equivalent to the current ILSAC standards. (i.e. API SN Resource
Conserving = ILSAC GF-5) The Resource Conserving designation applies only to multi grade
oils between 0w20 and 10w30 (and soon 0w16).

API Resource Conserving (API RC) requirements for fuel economy improvement are the
strictest of any industry specification and vary by grade. 0w20 and 5w20 oils must make the
greatest improvement, 0w30 and 5w30 less improvement, and 10w30 the least. Because the
parasitic drag of the oil is directly proportional to its viscosity, API RC oils are often formulated
toward the thin end of their viscosity grade, or designed to shear with use. The API requirements
for shear stability are particularly weak for 30 grade oils, allowing them to shear to 8.5 cSt in the
diesel fuel injector test. Because of this, API RC oils can be ill suited for some high performance
applications like turbo EJ Subarus unless they carry additional shear stability requirements like
those from the ACEA (see below).

The latest API service classes are SN (revision N) and CJ-4, with versions going back to SJ and
CH maintained as current. The specifications are generally written to be backward compatible,
so that an API SN oil can be used in cars which specified API SM, SL or even earlier in the
owner's manual. This is true for performance requirements, but some chemical requirements
have changed over time. With the introduction of API SM, reductions in the allowable amounts
of zinc and phosphorous in 30 grade and thinner oils were implemented for emissions system
durability reasons. Due to high levels of these additives, some Light Duty oils maintain their
certification at API SL.

The API maintains a certification and licensing program which ensures quality and allows oil
formulators to use the trademark “Donut” and “Starburst” symbols on their oil packaging. The
Donut is printed on the back label of oil containers and is used for any current specification, the
Starburst is reserved only for the latest S-series gasoline spec with Resource Conserving
extension, and is placed on the front of the container. Under the API's licensing requirements
formulators of engine oil may not market oil produced to non-current specs, but many un-ethical
formulators do any way. The Petroleum Quality Institute of America (PQIA) runs a constant
campaign to expose dangerously substandard oil for the benefit of consumers.
Figure 5 - The API “Donut” - http://api.org/~/media/files/certifi...glish_2013.pdf

All oils which carry the Starburst are Resource Conserving ILSAC certified oils conforming to
the latest specifications at the time of manufacture. You may find some old oil on the shelf, but
generally anything with a Starburst is a safe, current oil. Subaru bases the oil recommendations
in their manual on the API Donut/Starburst requirements.

Figure 6 – the API/ILSAC “Starburst” (From the 2010 Impreza OM)

http://www.pqiamerica.com/apiserviceclass.htm
http://www.api.org/certification-pro...ust-fluid.aspx
5.3 ILSAC Specifications

The International Lubricant Standards Acceptance Committee (ILSAC) is a consortium of auto

makers from the USA and Japan who write specifications for engine oil. Their requirements
coordinate with and extend the basic API “S” series requirements, particularly in the area of fuel
economy, but also in areas such as deposit formation. Unlike basic API 'S' oils which can be any
SAE grade, ILSAC requirements apply only to Light Duty multi-grade oils from 0w20 to 10w30
(and soon to the new 0w16), Heavy Duty oils 40 grade and higher are not included in the ILSAC
specs. The current specification is GF-5, and coordinates with API SN + Resource Conserving.
All ILSAC GF-5 oils are API SN + Resource Conserving, but it's theoretically possible to make
an oil which meets API SN + Resource Conserving, but does not meet ILSAC GF-5. In practice
all commercially available oils which meet SN+RC also meet GF-5.

Additive maker Infineum publishes a handy chart which summarizes the API and ILSAC
requirements (PDF)
http://www.infineum.com/media/80723/...tions-2013.pdf

5.4 ACEA Specifications

The Association des Constructeurs Européens d'Automobiles (ACEA) or in English, the

European Automobile Manufacturers Association is the European equivalent of a combined API
and ILSAC. They publish a comprehensive group of oil standards covering all categories of
service for both gasoline and diesel engines. For many aspects of performance their standards are
more stringent than the API/ILSAC specs. ACEA specs for engine performance are the most
stringent of the commonly cited specifications.

The A/B categories are for passenger gasoline and light duty diesel engines, the E categories are
for heavy duty diesels (think semi-trucks and earth movers), and the C categories place particular
emphasis on emissions system compatibility in both gasoline and diesel engines. The C category
is a class known as “Low-SAPS”, oils which have strict limits on content of Sulphated Ash,
Phosphorous and Sulfur.

ACEA certifications claimed by formulators are supposed to include the year of issue (i.e. ACEA
A5/B5-10 for the 2010 issue of the A5/B5 requirements) or if no year is denoted, they are to
conform to the latest issue at the time the claim is made by the formulator. Unfortunately many
formulators make claims which do not conform to the ACEA's guidelines seemingly without
repercussions. ACEA requirements have always been very stringent, but have become even more
so since 2008, so oils which carry the latest 2010 and 2012 certifications must be very high
quality indeed.
A/B series
All A/B series oils are intended for use in both gasoline and diesel engines at long to very long
service intervals. All sub-categories require 20 and 30 grade oils to stay within their original
grade when subject to a shear stability test consisting of being squirted 30 times through a diesel
fuel injector. These oils have moderate to high levels of SAPS additives and are suitable for very
long drain intervals, some European manufacturers specify up to 20,000 miles on these oils.

A1/B1: For engines designed to use low viscosity Light Duty oils with a HT/HS viscosity of 2.6
- 2.8 for 20 grade and 2.9 - 3.5 for all other viscosity grades. 40 grade A1/B1 oils are the only
ACEA oils allowed to thin out of their original grade in the shear stability test. With the 2012
revision of the ACEA standards A1/B1 wear standards are now the same as A5/B5 and A3/B4,
slightly more stringent than in 2010. A1/B1 and A3/B3 standards for piston cleanliness and
sludge are less stringent than for A3/B4 and A5/B5.

A3/B3: For high performance engines engines requiring HT/HS over 3.5. A3/B3 specifications
for engine wear are slightly less stringent than A3/B4, A1/B1 or A5/B5, but still stricter than
API/ILSAC. A1/B1 and A3/B3 standards for piston cleanliness and sludge are less stringent than
for A3/B4 and A5/B5.

A3/B4: For high performance engines requiring HT/HS over 3.5. A5/B5, A1/B1 and A3/B4
specifications for engine wear are identical and slightly more stringent than for A3/B3. A3/B4
and A5/B5 standards for piston cleanliness and sludge are more stringent than those in A1/B1
and A3/B3.

A5/B5: For gasoline engines and car & light van diesel engines designed to be capable of using
low friction low viscosity oils with a HT/HS viscosity between 2.9 and 3.5 mPa.s. Similar to
A1/B1 but without provisions for 20 grade oils and with increased requirements for piston
cleanliness and sludge resistance. A1/B1, A5/B5 and A3/B4 specifications for engine wear are
identical and slightly more stringent than for and A3/B3. A3/B4 and A5/B5 standards for piston
cleanliness and sludge are more stringent than those in A1/B1 and A3/B3.

E- Series
E series oils are intended for heavy duty diesel engines in machinery like over the road trucks
and earth moving equipment. The distinction between these oils and the passenger car oriented
A/B series has blurred over the years as governments have tightened the emissions requirements
for heavy diesels to be more in line with passenger cars. This series is relevant to Subaru owners
because several oils meet specs both for gasoline passenger cars and HD diesels, and therefore
must meet the strictest requirements for both. With the exception of the E4 category which uses
the A/B series shear stability test, all E-series oils must pass a shear stability test which is three
times longer than for the A/B series (90 cycles vs 30). The two relevant E series categories for
Subaru owners are E7 and E9, they are nearly interchangeable from the perspective of a gasoline
engine, E7 might allow for longer oil change intervals, but E9 has lower levels of additives
which might be harmful to emissions equipment. Either will have excellent shear stability.

C-Series
The C-series is a newer class of Low-SAPS oils with strict limits on ash, phosphorous and sulfur.
The resulting oils are very similar to the A/B series, but must be changed at shorter intervals. The
existence of this series is driven by the advent of diesel particulate filters which are very
sensitive to clogging with burned oil byproducts. Where the A/B series can last up to 20,000
miles in gasoline engines, C-series oils normally max out at 10,000. In regards to engine
performance, C1 and C2 are similar to A5/B5, while C3 and C4 are similar to A3/B3/B4.

http://www.acea.be/publications/arti...sequences-2012

For most specifications, Lubrizol's “Relative Performance Tool” is a great way to compare
various aspects of performance:
http://lubrizol.com/EngineOilAdditiv...l/default.html

5.5 Manufacturer Specifications

In addition to industry groups, many manufacturers write their own specifications for motor oil
to be used in their engines. Unlike the industry specifications, the underlying requirements of
manufacturer specifications are usually not available to the general public, it is therefore difficult
to use such specifications as a criteria for oil selection unless your particular car requires an oil
which meets a manufacturer specification. What we can assume from the presence of
manufacturers specifications is that none of the available industry specifications was stringent
enough to meet the manufacturer’s requirements. So, if we see an oil which meets both the
industry specs required by Subaru and some manufacturers specs, we can assume that oil meets
requirements more stringent than industry specs. Whether those extra capabilities are relevant to
our application is an open question. Many European manufacturer specifications are derivations
and extensions of ACEA categories. North American and Asian manufacturer specifications tend
to be extensions of ILSAC and API specifications.
6. Categories
These categories are based on those defined by the ACEA for use in gasoline and diesel engines
of either light or heavy duty nature. As Subaru doesn't make diesel engines for North America,
we will concern ourselves mostly with the gasoline categories, and only refer to diesel categories
if their requirements are relevant in a gasoline engine. It's important to understand that many of
these categories are not mutually exclusive, it's possible to meet the requirements of more than
one category at a time. For instance, EJ turbo engine favorite Shell Rotella T6 5w40 oil is
designed and marketed for heavy duty diesel engines in the ACEA E9 category, but also meets
API SM requirements for use in gasoline engines. Generally speaking all oils within a category
will perform similarly, but these categories being arbitrary, some oils which push the borders of
their category may be a wiser choice under certain circumstances.

Light and Heavy Duty Engine oils

Years ago engines were clearly delineated between the ***8220;heavy duty***8221; engines in
commercial vehicles like semi-trucks and earth moving equipment and the ***8220;light
duty***8221; engines in passenger cars and light trucks. The engines in these vehicles generally
used different oil, light duty vehicles used Light Duty oils (30 grade or lower) and heavy duty
vehicles used Heavy Duty oils (40 grade or higher). There were some exceptions like high
performance cars with flat-tappet engines which used HD oil, but the rule generally held. Today
there are many applications where the oil does not match the vehicle, many light vehicles use
heavy oil and some heavy vehicles use light oil. So while the names have lost their descriptive
value for the application, they still describe the oil's relative viscosity, Heavy Duty oils are
thicker than Light Duty ones. The confusion comes with 30 grade oils which can be either Light
or Heavy Duty.

The key characteristic of Light Duty oils is a HT/HS viscosity under 3.5. Light Duty oils come in
SAE grades 30 and below, including all multi-grades. The focus of these oils is generally on
maximizing fuel economy in passenger car and light truck applications. A common sub-category
of Light Duty oils is API Resource Conserving/ILSAC oils, which can be thought of as
***8220;super-light***8221; duty oils. These must meet additional fuel economy requirements
that non-RC oils of the same grade do not. All Resource Conserving/ILSAC oils are Light Duty
oils, but not all Light Duty oils are Resource Conserving, 20 grade oils marketed as
***8220;high mileage***8221; are a common example. The majority of oils found on the
shelves of your local retailer will be Light Duty. Specifications which address Light Duty oils are
API S with or without Resource Conserving extension, ILSAC GF-x, ACEA A1/B1, ACEA
A5/B5, ACEA C1 and ACEA C2.

Heavy duty oils are found in straight and multi-grades starting at 30 grade and going up. In
passenger cars HD oils tend to be specified by European manufacturers. Their defining
characteristic is a HT/HS viscosity above 3.5. SAE J300 only requires HT/HS greater than 3.5 in
40 grade and above oils, but ACEA A3/B3, A3/B4, C3, C4, E7 and E9 require 30 grade oils to
have HT/HS greater than 3.5. Many OEM specifications from European manufacturers also
specify 30 grade oils with HT/HS>3.5. SAE J300 requires 15w40 and thicker oils to have HT/HS
of 3.7 or greater.

So, when we combine Light and Heavy Duty categories with available grades we get six oil
categories which are relevant to North American Subaru owners, in order from lowest to highest
HT/HS:

Resource Conserving 20 grade (Light Duty)

0w20 and 5w20 oils with API RC/ILSAC certification, may also Meet ACEA A1/B1 and OEM
specifications. These oils have virgin HT/HS between 2.6 and 2.8.

Non-Resource Conserving 20 grade (Light Duty)

20, 0w20 and 5w20 oils which do not meet API RC/ILSAC specs, but may meet API S, and/or
ACEA A1/B1. These oils have virgin HT/HS between 2.6 and 2.8.

Resource Conserving 30 grade (Light Duty)

0w30, 5w30, and 10w30 oils with API RC/ILSAC certification, may also meet ACEA A1/B1,
A5/B5, C1 or C2. These oils have virgin HT/HS between 2.9 and 3.4.

Light Duty 30 grade (Non-Resource Conserving)

Any 30 grade oil which does not meet API RC/ILSAC requirements, but should meet API S
when used in gasoline engines and may also meet ACEA A1/B1, A5/B5, C1 or C2. These oils
have virgin HT/HS between 2.9 and 3.4.

Heavy Duty 30 grade

Any 30 grade oil with an unused HT/HS greater than 3.5. These oils cannot meet API RC/ILSAC
requirements, but should meet API S when used in gasoline engines and may also meet ACEA
A3/B3, A3/B4, C3, C4, E7 or E9.

40 grade (Heavy Duty)

All 40 grade oils are Heavy Duty and must have HT/HS greater than 3.5. Straight 40 grade and
15w40 oils are required by SAE J300 to exceed 3.7. Many high quality 0w40, 5w40 and 10w40
oils exceed 3.7 as well. Any 40 grade oil used in a gasoline engine should carry API S and/or
ACEA A/B or C certification.
Low-SAPS engine oils
These oils are found in any commonly available viscosity grade from 0w20 up to 5w40, they can
be Light Duty or Heavy Duty, their defining characteristic is low levels of Sulphated Ash,
Phosphorous and Sulfur (SAPS). Other than generally requiring shorter oil change intervals,
these oils perform equally to their higher-SAPS counterparts. ACEA C-series oils are the most
common Low(est)-SAPS oils, but many oils are described by their formulators as Low-SAPS
even without an ACEA C designation. With the exception of Sulphated Ash, API Resource
Conserving and ILSAC oils contain similar levels of Phosphorous and Sulfur, and can usually be
considered ***8220;Low-SAPS***8221; even though they're not normally referred to as such.

Industry professionals and enthusiasts have noted that Low-SAPS oils tend to reduce intake
valve deposits on direct injected engines, which may make these oils interesting to owners of
newer, direct injected Subarus.

http://www.lubrizol.com/EngineOilAdd.../LowerSAPS.pdf
7. How Oils are Marketed
It would be nice if formulators would label their oils with straightforward categories on the front
label, but they don't. They use nebulous marketing terms like synthetic and high-mileage.
Nevertheless, we can gain some useful information from the way oils are marketed.

Conventional Oils
Usually composed mostly of Group I and II base stock, but may contain substantial amounts of
Group III, any presence of Group IV and V is as an additive. These oils are the bottom of each
formulator's range and are thus produced to a price point. They are usually not labeled as
conventional, that is implied by the lack of terms like blend or synthetic. They don't generally
meet the toughest requirements of ACEA standards, but do meet some OEM standards. They
tend to have lower TBN values and thus require shorter change intervals than synthetic or blend
oils, but in many non-demanding applications they can provide excellent service for the life of
the car. Conventional oils can be Light Duty, Heavy Duty, or Low-SAPS, though Low-SAPS
conventional oils are rare. Being the value level in a formulator's product range, one can assume
the additive packages are designed to meet the minimum requirements of the specifications as
cheaply as possible.

Synthetic Oils
Composed of Group III-V base stocks. These oils occupy the upper range of formulator's product
lines. Most moderately priced synthetics today are Group III base stock, but high quality Group
III performs on-par or better than more expensive base stocks for all but the most demanding
applications. They often have high TBN levels, but not always. The long-term shear and
oxidation stability of synthetic base stocks is necessary but not sufficient for very long change
intervals (10,000+ mi). There are many synthetic oils formulated with the very best base stocks
which for various reasons are not formulated for long drain intervals. These oils tend to be race,
street/track or Low-SAPS oils. Being the more expensive oils in a product line it can be assumed
these oils get the best additives in addition to superior base stocks. Synthetic oils are available as
Light Duty, Heavy Duty and low-SAPS varieties.

Formulators often have several tiers of synthetic oil. Shell Oil Products has Quaker State
Ultimate Durability, Pennzoil Platinum, and Pennzoil Ultra Platinum in that order. Mobil has
Mobil Super Synthetic and Mobil 1, Castrol has Edge and Edge Extended Performance. Upper
tier oils tend to meet more stringent ACEA specifications. Some upper tier oils are long-drain
oils which don't necessarily perform better at short intervals than their short-drain counterparts.
Unless you're aiming for oil change intervals longer than 10,000 miles there's no reason to use
Mobil 1 Extended Performance over regular Mobil 1 for instance. Since many late model Subaru
owners will be choosing these oils, we'll detail some of the major players later.
Long-Drain Oils
These oils are marketed specifically as having the ability to do extremely long drain intervals.
They are also always marketed as synthetics. Common examples are Amsoil Signature Series,
Mobil 1 Extended Performance, and Castrol Edge Extended. They are marketed as being capable
of drain intervals of 15,000 miles or more. It's important to note that many oils not specifically
marketed as long drain oils are never the less capable of very long drain intervals, particularly
those conforming to ACEA A5/B5, A3/B4, E7, and E9.

Synthetic Blend Oils

These oils are marketed by formulators as a blend of synthetic and conventional base stocks.
Typically a blend of Group II and Group III, but there's no standardization for how much of
each. One can generally assume that the additive package in these oils falls between synthetics
and conventional oils. Whether blend oils are a good value is up for debate. Blends are available
as Light Duty, Heavy Duty and Low-SAPS varieties.

High Mileage Oils

High mileage oils are marketed as synthetics, blends, and conventional oils. Their defining
characteristic is higher levels of detergents and seal conditioners than conventional oils. Some
have higher levels of anti-wear additives, often too high to conform to the latest API/ILSAC
specs. Some meet API/ILSAC fuel economy requirements, some don't. The stated purpose of
these oils according to formulators is to clean out sludge from neglected engines and to soften
and revitalize aged seals, thus reducing leakage. Their additional anti-wear additives probably
indicates higher levels of wear performance than conventional oils, but it is unlikely they
outperform top-tear synthetics in that regard. Detergent effectiveness is also unlikely to be better
than top-tier synthetics. High mileage oils tend to be a bit thicker than their low-mileage
counterparts, often pushing the upper limit of their viscosity grade. It is generally safe to use
these oils in new engines as long as the other specs are appropriate.
Racing Oils
Many national brands produce a lineup of oils labeled as racing oils, these include Red Line's
racing series, Mobil 1 Racing and Amsoil Dominator. These oils have low levels of detergents
and TBN which make them unsuitable for use in street cars. They are designed purely for racing
applications where a long race is 500 miles and the engine never has a cold start. Many owners
are attracted by the high levels of anti-wear, anti-oxidant and friction modification additives in
these oils and erroneously presume them to be better for their modified street car. Unless your
car is a bona fide trailered-to-the-track race car, DON'T DO IT. The additive packages of these
oils are designed for the short drain intervals and narrow operating environment of racing, use in
a street car can lead to deposits and other problems.

Street/Track Oils
These oils are designed for use in street cars which see heavy track day use. They maintain some
levels of detergent additives and TBN for use on the street but shift their focus as much as
possible toward track use. They may be appropriate for heavy use in a street car, but are
generally expensive and unnecessary. Examples include Red Line (regular series), Valvoline
VR1 and Motul 300V.
8. Oil analysis – Closing the Loop
Where the manufacturers data sheets contain “typical” values for a particular type of oil when
new, used oil analysis (UOA) measures many of those factors after it’s been in use for some
time. UOAs are accomplished by taking a sample of oil from your engine, and sending it to a
laboratory who offers the service. Particularly retentive enthusiasts will also send an unused
sample of their oil for analysis so they can have baseline values for their particular batch of oil
rather than the “typical” values listed on the product data sheet, this is known as Virgin Oil
Analysis (VOA).

Companies offering UOA services include:

Blackstone Labs (http://www.blackstone-labs.com/)

Polaris Labs/Oil Analizers (http://polarislabs1.com/ or oiatesting.com)
Bently Tribology Services (http://www.cashmanequipment.com/bently/index.php)
Dyson Analysis (http://www.dysonanalysis.com/home.html)
Predictive Maintenance Services (http://www.theoillab.com/)
Wearcheck (also in Canada) (https://wearcheck.com/)
ALS Tribology (http://www.alsglobal.com/Our-Service...-Analysis.aspx)

Oil analysis was originally developed for over-the-road trucks which take several gallons of oil
and can have large variance in oil change interval depending on if they're doing city deliveries or
long hauls. In that context, it makes sense to spend $20 on analysis to determine if you need to
replace $100 worth of oil, and UOA is well suited to this purpose. In a street car which only
takes $25 worth of oil, we're generally not using UOA to minimize maintenance costs, we're
using it to optimize oil choice, and for this purpose it is less well suited. For Subaru enthusiasts
UOA is useful in a few ways, in decreasing order we can:

1. Confirm that the oil's virgin characteristics hold through the change interval.
UOA will measure the oil's KV100 which we can compare to the virgin value. Most baseline
UOA packages don't measure the KV40, but labs will measure this for an additional fee. With
both KV100 and KV40 we can calculate the used oil's viscosity index. Care should be taken
though as the accuracy of the measurements is poor, and the calculated VI can vary wildly as a
result.

2. Confirm that the change interval is appropriate for the car's operating conditions.
UOA will usually measure the TBN (sometimes it's an extra charge), and sometimes the Total
Acid Number (TAN), Oxidation, and Nitration of the used oil. The exact condemnation limits for
these parameters is a topic of debate, but following the guidance of your analysis lab will
generally tell you if your change interval is appropriate for your operating conditions.
3. Have some idea that the engine is healthy.
Small amounts of metals including iron and aluminum become entrained in the oil as the engine
wears, sudden spikes in some metals, or concentrations wildly outside the norm can indicate
problems. The presence of other substances like fuel and coolant can also indicate problems.
Care should be taken not to over-interpret these indicators as variances in usage can cause
substantial variation in observed values without indicating a problem.

4. Satisfy some of our curiosity about the oil's additive package.

Lots of “internet tribologists” love to look at the additive elements which show up in UOA and
proclaim one oil as “better” than another. This is folly as the UOA methods show only elements,
not the compounds which make them up and also don't show all the potential elements. A UOA
can tell us the concentration of molybdenum in the oil, but it can't tell us if that moly is tri-
nuclear moly-DTC or moly-disulfide, two compounds which behave differently. Still, looking at
these elements can sometimes be useful. For example, some oils use silicon compounds as anti-
foaming agents. If you know you're using such an oil, the presence appropriate concentrations of
silicon in UOA doesn't indicate poor air filtration.

This guide will not go into detail on how to interpret a UOA report because the labs who offer
UOA explain it well on their websites. What we will cover is a basic overview of UOA, along
with some of its limitations which are often not so clear. The information in the typical UOA
report falls into four categories: elements present, physical properties, chemical properties, and
contaminants present.

8.1 Elements

Elements are measured in parts per million and represent just what the name implies, chemical
elements dissolved or suspended in the oil. They have three sources, the virgin oil, the engine
itself, and the air and fuel flowing through the engine. It’s important to note that the report tells
you nothing about what form these elements are present in, it measures only fundamental
chemical elements, not molecules. Because of this we can only speculate as to the reason for
their presence. For some elements this is straightforward, but others present problems. Elements
commonly misinterpreted in UOA:

Iron
Iron comes from wear of iron or steel engine parts, iron compounds are not used in oil additives
and are not present in air or fuel. The detection methods used in UOA only measure a narrow
range of particle sizes for iron, so it's impossible to say whether small variations in iron
concentration actually represent differences in oil wear performance. Iron concentration tracks
with drain interval, so we can expect to see more iron in a 7,000 mile sample than in a 3,000 mile
sample. Typical concentrations in Subaru engines vary from 10 ppm at short change intervals up
to about 20 ppm for heavy use and long intervals.
Aluminum
Aluminum comes from wear of aluminum engine parts, particularly piston skirts, but sometimes
journal bearings. Small amounts should be of no concern, but large amounts (over about 15 ppm
in Subaru engines) may indicate poor air filtration, especially if accompanied by elevated levels
of silicon (from ingested sand).

Silicon
Silicon has several potential sources. Anti-foaming additives in the oil and RTV sealant used to
assemble or service the engine are benign, small particles of sand ingested through a poor quality
or damaged air filter will increase wear. If an oil formulation is known to contain silicon, or if
the engine is new or had a gasket replaced recently, the presence of elevated silicon should be of
no concern. If neither of those explanation holds, and especially if aluminum is elevated
(indicating piston wear) poor air filtration should be suspected. Some silicon from tiny ingested
dirt particles is to be expected, and increases with change interval mileage. For typical change
intervals and OEM air filtration, silicon typically shows 10-20 ppm in Subaru engines.

8.2 Physical Properties

KV100
KV100 is possibly the most important parameter to Subaru enthusiasts measured by UOA. It
helps us know if the oil we selected based on its virgin properties actually retains those
properties in service. Ideally the lab would measure HT/HS, and many offer that as an add-on
service costing hundreds of dollars, but that's obviously cost prohibitive, so we use KV100 as a
proxy. Luckily, automotive engineers have studied the relationship between KV100 and HT/HS,
and it's relatively predictable. HT/HS is generally reduced by half the amount that KV100 is
reduced due to shearing. So, an oil who's KV100 shears 12% in service will see its HT/HS
reduced by 6%.

In practice most Group III synthetic oils shear 10-15% in use, Ester/PAO oils somewhat less,
conventional oils somewhat more. If your oil shears more than this or if your usage results in
thinner oil through fuel dilution, a thicker oil or one more resistant to fuel dilution is a wise
choice.

http://www.astmtmc.cmu.edu/docs/dies...52501ATT12.PDF

Flashpoint
Oil's flashpoint is reduced primarily by the presence of fuel, but can also be reduced through
shearing and other mechanisms.
8.3 Chemical Properties

TBN
This is the oil's remaining ability to neutralize acidic contaminants, and is the primary
determinant of whether the change interval can be extended. Different labs recommend different
minimum limits for TBN. The most popular lab, Blackstone, recommends 1.0, second most
popular Polaris/OAI, recommends 2.0. Newer additive packages exhibit non-linear depletion of
TBN which older formulations did not. This is particuarly true for low-SAPS oils which often
start with a seemingly low TBN under 7.0, but can usually handle moderately long change
intervals any way.

8.4 Contaminants

Fuel
The presence of fuel in oil will lower the flashpoint and viscosity through damage to base stocks
and additives, substantially shortening the oil's service life. Properly sampling the oil after
heating it thoroughly should remove short term cold-start fuel from the oil but many engines
regularly run rich and substantial fuel dilution over the long term can be the result. Labs use
various methods to estimate fuel content, Blackstone in particular estimates it using flashpoint
which can lead to large variances in the reading.

Water
The presence of water in oil is often an indication of poor sampling procedure. The oil sample
should be taken after the car has been warm sufficiently long for all water in the oil to be boiled
off. Substantial water content in a sample taken hot might indicate a mechanical problem with
the engine.
9 Selecting an Oil for your Subaru
After all that we finally come to the meat of this guide. In basic terms, we'll start with the
recommendations in the owners manual, reverse engineer and read between the lines a bit, and
then use that informed speculation to modify the recommendations based on our unique
operating conditions.

To this point you will notice that the specifications governing engine oil focus on typical
conditions (KV100) or worst case conditions (MRV, HT/HS), but don't say much about actual
operating conditions. Since the specifications are written for cars in general, not those of a
particular manufacturer let alone a particular usage scenario, this is understandable. So now it's
time to think about the actual operating conditions inside our engines, and ask ourselves what
Subaru intended.

9.1 Subaru Oil Requirements

The oils most prominently recommended by Subaru for normal use in various engines are:

EJ series naturally aspirated – API RC/ILSAC 5w30

EJ series turbo – API RC/ILSAC 5w30 (synthetic from 2011)
FB series naturally aspirated – API RC/ILSAC 0w20
FA series naturally aspirated – API RC/ILSAC 0w20
FA series turbo – API RC/ILSAC 5w30

Prior to 2011 Subaru owner's manuals contained specific recommendations up to 50 grade for
heavy use in high temperatures, they continue to contain vague language about heavier oil for
high temperatures, but no specific grades or temperatures are mentioned. Why have they
removed reference to heavier grade oils? We we don't know for sure, but we can speculate fairly
effectively. Subaru's recommendations place higher emphasis on fuel economy and emissions
system durability than an owner might. In order to use API RC oil in the EPA fuel-economy
tests, the EPA requires manufacturers to either list it as the only grade in the owner's manual or
to list it as the “preferred” grade. Subaru used to list API RC oils as “preferred”, but now
mentions them exclusively. The other reason to emphasize API RC oil is emissions system
durability. Subaru is required by law to cover important emissions system components for at
least 80,000 miles (150,000 on PZEV models), this applies even if the vehicle burns oil. API RC
oils have lower levels of potentially harmful additives which “poison” emissions equipment
when burned. Combine those two regulatory requirements and we have a pretty good idea why
the recommendations for high viscosity oils with robust additive packages went away. Is Subaru
screwing owners over? Generally no, Subaru doesn't like warranty claims any more than owners
do, so in most circumstances the recommendations provide excellent service in normal operation,
the one possible exception is the turbo EJ engine (see - epa.gov/obd/pubs/420f09048.pdf )
Extrapolating from Subaru's recommendations

So, Subaru makes recommendations for “normal” use, but what are your usage conditions? How
are they different from Subaru's baseline? How do those conditions affect oil choice? Answering
those questions will require some informed speculation and will leave substantial ambiguity. If
you're towing a trailer into the mountains of Denali National Park in January or track driving at
Road Atlanta in the summer, the “best” oil is likely different from what's in the manual. Ideally,
we could measure oil temperature under various conditions to inform our choice, but until now
most Subarus were not equipped with oil temperature sensors, and most owners didn't invest in
aftermarket ones. Luckily, newer FA series engines (FB too?) are equipped with an oil
temperature sensor which can be read using a scan tool, owners of such cars who are truly
interested can monitor their oil temperatures and make a more informed choice. Most owners
though will have to make due with generalizations about how the car is used.

What can we assume about how Subaru designed the hydrodynamic interfaces of the engine?
Well, we can assume they used the Stribeck curve, remember that? Remember point 'a' the
transition between mixed-film and hydrodynamic lubrication which happens at a certain
viscosity? Notice how the friction increases quickly as we move into mixed-film lubrication and
much more slowly as we move farther into the hydrodynamic regime? Being mixed-film
lubrication, increased friction also indicates increased wear, while increased friction in the
hydrodynamic regime reflects only increased hydrodynamic drag without increased wear. It is
therefore much preferred to stay in the hydrodynamic regime even if it means slightly increased
friction, and Subaru's engineers know that. We can therefore assume that Subaru designed the
engine so under “worst-case” conditions the operating point of the engine will be at point 'a' and
all other times will be to the right. We can further assume that this condition will be at maximum
oil temperature, which in terms of piston cooling will be sub-optimal. The engine will be
designed to survive these conditions, but not thrive under them. This is a simplistic, but
instructive view.
Figure 7 - The Stribeck Curve

1 - Boundary lubrication
2 - Mixed film lubrication
3 - Hydrodynamic lubrication

So what exactly is “worst-case” operation? Subaru has given us some clues in the owners
manuals by indicating that thicker oils are recommended for heavy use and high temperatures.
From that clue we can infer that the viscosity of the oil recommended for the car at the
temperatures seen during moderately heavy street use puts us a bit to the right of point 'a' on the
Stribeck curve, for a little bit of safety margin. Such use would be twisty-road carving on a hot
day (say 90° F) in a car that is not heavily loaded and not climbing a mountain. Any decrease in
viscosity, either through shearing or higher oil temperatures, risks using up the safety margin and
moving toward mixed-film lubrication.

In reality it's not quite as bad as it seems. Notice the dashed line to the left of point 'a'? That's the
actual friction curve of real motor oil. The idealized Stribeck curve is for a pure hydrodynamic
lubricant, one without anti-wear or friction modification additives, an actual oil will not see the
large increase in friction in the transition to mixed-film lubrication because it contains anti-wear
and friction modification additives. So a small amount of mixed-film operation will not result in
bearing death, but being diligent owners we want to avoid the situation any way.
So, if you use your car only moderately hard, the oil recommended in the owner's manual should
provide good service, but the definition of “moderately hard” is ambiguous, and the
consequences of poor selection are asymmetric; an oil which is too thick costs you a little fuel
economy and a couple horsepower, an oil which is too thin causes bearing wear or possibly a
spun bearing. Given that, selecting an oil which is slightly thicker than we really need is entirely
rational in pursuit of a bit of safety margin, but there's a problem.

At the same temperature, a thicker oil will provide a stronger hydrodynamic film, but it will also
run hotter under the same operating conditions. The additional work done by the engine to move
the thicker oil and spin rotating parts through it ends up as heat in the oil. A 40 grade oil won't
run so much hotter that it's operating viscosity is lower than a 30 grade, but we can't simply
compare KV100 values and assume they represent operating viscosity. There's a delicate
balancing act here, oil that's too thin risks bearing wear, oil that's to thick promotes detonation by
keeping the pistons hotter.

That means the range of optimum oil temperature is quite narrow, while the range of acceptable
oil temperature is much wider. To keep water and fuel out of the oil we want the oil temperature
above 212°F (the boiling point of water), but we want the pistons as cool as possible, so
optimum oil temperature is in the range of 215-230°F. Operating temperatures down to 170°F or
so occur in low load running in cold weather and generally aren't a problem. Temperatures up to
270°F occur in heavy use in high temperatures and usually aren't a problem either. This thread
from FT86 club takes advantage of the oil temperature sensor in the 86 twins to give us some
decent information on typical oil temperatures:
http://www.ft86club.com/forums/showthread.php?t=29965

Taken together, these factors indicate that wise oil selection will be at most two categories
thicker than recommended by Subaru. For engines which recommend 0w20, a HD 30 grade is
about the thickest wise choice, for 5w30 engines, a 40 grade is about the thickest wise choice. 40
grade oil in a 0w20 car or 50 grade in a 5w30 car could be used in the event of infrequent very
heavy use, but is definitely sub-optimal. An examination of viscosity-temperature behavior will
show us why.
Here's a viscosity chart with typical oils from the relevant categories zoomed in on the normal
operating temperature range. We can see that stepping up one category results in an oil which
has the same viscosity about 20°F (~10°C) warmer.

Figure 8 – Operating Temperature Viscosities

If Subaru recommends 0w20 for your engine, the bearings should be very happy with 7.5 cSt oil,
if we put RC 5w30 into the engine, the oil temperature can rise to 240°F before the oil viscosity
drops below 7.5 cSt, now we're at the upper limit of optimal oil temperature. With HD 5w30 the
7.5 cSt temperature rises to 260°F, basically the upper limit of acceptable. In other words, using
a 40 grade oil implies we will be operating the engine at unacceptably high oil temperatures. And
since using a heavier oil encourages higher operating temperatures, we're compounding the
problem. So, if you think you need 5w40 in your car that calls for 0w20, what you really need is
an oil cooler, possibly a big one.

The same logic goes for a car that calls for 5w30, the bearings should be very happy with oil at
10 cSt, stepping all the way up to 50 grade implies oil temperatures in excess of 260 °F, well
outside the optimal range. So why did Subaru formerly recommend up to 50 grade oils for heavy
use and high temperatures? Simple, it's a band-aid for the small fraction of owners who use their
cars that heavily. Subaru's choice was either add an oil cooler to every car they made, which
costs them money, or put the heavy oil recommendations in the manual, which costs them
nothing.
The above chart is viscosity for new oil, but what about used oil? The lines in the chart simply
move down as the oil shears, but they retain their shape. Since we don't know Subaru's exact
design point any way, the conclusions about oil choice hold whether we're looking at a graph of
used oil or new.

We've talked a lot about viscosity here, but what about other requirements? The API/ILSAC
specs called for by Subaru ensure acceptable performance in areas like compatibility with seals,
detergency, and TBN. Note that many oils, especially top-tier synthetics, exceed the minimum
requirements by a substantial margin, and can therefore be expected to handle substantially
longer change intervals or more severe service than the minimum-spec oils Subaru must assume
when making recommendations.

9.2 Usage Scenarios

Let's take a moment to categorize use scenarios. Your use scenario is the worst case you will
encounter during a particular oil change interval.

Normal Street Driving

Just like it sounds, commuting, highway trips, etc. Even the occasional fast blast down a twisty
road qualifies as normal street use with an unmodified car. You simply can't apply full throttle
long enough on the street to raise the oil temperature high enough for it to qualify as heavy use
which might require thicker oil. An owner who lives in a particularly warm climate (e.g.
Phoenix, AZ) might consider hard driving on a twisty road as heavy use. High proportions of city
driving can be cause for shortened oil change intervals, but normal street driving generally
allows normal change intervals as recommended by Subaru, or even longer if confirmed by
UOA.

Heavy Use – Trailer Towing and Track Driving

Heavy use involves long periods of operation under heavy load, either due to high speeds on the
track or heavy weight and/or steep grades on the road. Heavy use combined with high ambient
temperatures could be considered “extra heavy”. The result is higher oil temperatures and
consequently thinner oil. If you use your car in this way, stepping up to the next higher category
and/or grade of oil is wise, at the very least select an oil within the lower category which displays
the highest shear and thermal stability. Owners who regularly use their cars heavily should
consider an oil cooler, especially if the car did not come equipped with one from the factory.

Track driving also subjects the engine to high g-loads which can cause large amounts of oil to be
ingested through the PCV system, potentially encouraging knock. Autocross is a unique
situation, the duration of heavy load is generally not enough to result in high oil temperatures,
but the g-loading can cause the engine to ingest oil through the PCV. Owners who regularly
track and/or autocross may consider adding an air-oil separator to the engine to minimize oil
ingestion, or at least using a street/track oil designed to minimize negative effects when ingested.

Heavy use often necessitates shorter oil change intervals due to fuel dilution, thermal stress, and
shearing.

Extreme Cold
With the easy availability and low cost of modern synthetic oils, few compromises in oil
performance are required to handle occasional temperatures down to -22°F, the limit of a 5wXX
multi-grade oil, but for extremely cold temperatures a 0w multi-grade oil is wise. For our
purposes, extreme cold is frequent temperatures below -13°F. Remember the definition of the w
number grades, 10w oils are good down to -13°F (-25°C), 5w oils are good down to -22°F (-
30°C) and 0w oils are good down to -31°F (-35°C). There are some benefits to using an oil one
grade better than the temperatures you'll encounter, so 5w oils will show small benefit over 10w
oils up to about -4°F, and 0w oils will show small benefit over 5w oils up to about -13°F. Few
places in the lower 48 states see any benefit from 0w oils, but they can occasionally be useful in
states bordering Canada. Canadian and Alaskan owners are wise to use 0w oils during the winter
months.

Extreme cold can shorten optimum oil change intervals, especially if the car is used for short
trips where the oil never reaches optimum temperature.

Modifications
Modifications can run the gamut from simple cat-back exhausts, to custom built engines with
giant turbos, making generalization about oil in modified engines difficult. Almost without
exception modifications which increase horsepower will also increase oil temperatures, making
thicker oil necessary in modified cars when it wouldn't be in unmodified cars. Owners of
modified cars who intend to use them heavily are wise to add a high capacity aftermarket oil
cooler, even if the car was equipped with one from the factory. Doing so will of course keep oil
temperatures down, but will also keep the pistons cooler, which helps prevent knock. Modified
cars are often tuned to run rich, which can increase fuel dilution and shorten optimum change
intervals.
9.3 Recommendations for specific cars/engines

9.3.1 EJ series Naturally Aspirated Engines

In North America, all Subaru four cylinder engines from 1991 until 2011 were variants of the EJ
series. Those without forced induction are not demanding of oil and will generally function well
on the Light Duty 5w30 oils recommended by Subaru. Naturally aspirated EJ engines were
discontinued at the end of the 2011 model year in the Impreza but were produced through 2012
in the Legacy/Outback.

Street driving
Light Duty 30 grade conventional or synthetic, 5w30 is most common but 0w30 and 10w30 will
also perform well in appropriate temperatures. Owner's manuals call for Resource
Conserving/ILSAC 5w30. Conventional oils perform fine at moderate change intervals up to
5,000 miles, synthetics can be used to 10,000 miles or more, but should be confirmed with used
oil analysis. Extending change intervals beyond the OEM recommendations could risk warranty
coverage.

Heavy Use
At a minimum a 30 grade Light Duty synthetic is a wise choice. If you don't expect to see
temperatures below 0°F, 10w30 synthetic oil is often more shear stable and resistant to heat than
other Light Duty oils and can provide a small safety margin for heavy use. For very heavy use
like uphill trailer towing or track driving in high temperatures, a Heavy Duty 30 grade or 40
grade oil is wise.

Modified Engines
Normal bolt-on modifications combined with a near-OEM redline should not necessitate altering
the oil choice otherwise made for operating conditions. API RC/ILSAC 5w30 for the street and
slightly heavier oil for heavy use should still be adequate. Aftermarket forced induction systems
have the potential to make the engine very demanding on oil as do “race-built” naturally
aspirated engines. In such a case, consult with your engine builder, or at the very least follow
recommendations for a modified turbo EJ engine (below).

Extreme Cold
A Light Duty 0w30 synthetic oil should be sufficient for a naturally aspirated EJ engine in
extreme cold, even if used heavily, no need to use a thicker HD oil.
9.3.2 Turbocharged EJ Series Engines

Subaru introduced the turbo EJ engine in the 1991 Legacy turbo, and began selling them in
volume with the 2002 WRX. The engine's lone remaining application is in the 2015+ WRX STI.
This engine is the most problematic Subaru engine for oil selection because of the sub-optimal
recommendations made by Subaru (particularly after 2010) combined with aggressive tuning of
the engines for fuel economy and emissions. A full explanation of the issues is beyond this
guide, but the primary issue is spun bearings due to contact between the crankshaft and the
bearing (lubrication failure) the proximate cause is detonation which can be worsened by oil
ingested through the PCV. Additionally oil which is thin either because it's very hot or because
of poor shear stability is less capable of protecting the bearings. It is therefore reasonable to use a
thicker oil in a turbo EJ to reduce, but not eliminate, the chance of a spun bearing.

Street Driving
The Resource Conserving 5w30 recommended in the owner's manual (synthetic from 2011) is
just barely adequate for street driving, especially in the 2008-2014 WRX which does not have an
oil cooler. A quality, shear-stable RC 30 grade synthetic conforming to ACEA A5/B5
specifications is recommended as a minimum. Heavy Duty 30 and 40 grade oils are a reasonable
upgrade.

Heavy Use
Turbo EJ owners who use their car heavily are wise to use a HD 30 grade oil at a minimum, and
HD 40 grade oils can provide additional safety margin. Installation of an oil cooler in cars not so
equipped from the factory is wise for heavy use, a high capacity aftermarket cooler can provide
additional safety margin.

Modified Engines
Owners of modified cars are wise to use Heavy Duty 30 grade oil at a minimum for street use.
Modified cars receiving heavy use should have an oil cooler installed in addition to the use of
HD oil. For occasional heavy use, modified cars can get away with shear stable HD 40 grade or
50 grade oil conforming to ACEA A3/B4, or a combination of E7 or E9 and a gasoline
specification like API SM.
9.3.3 Naturally Aspirated FB Series Engines

With the exception of the BRZ sports car, all naturally aspirated Subaru four cylinder engines
since 2013 are of the FB series, it has replaced the EJ series as Subaru's main naturally aspirated
engine. The FB25 was introduced to North America in the 2011 Forester, the FB20 followed in
the 2012 Impreza, and in 2013 the FB25 replaced the EJ25 in the Legacy/Outback. All versions
of this engine recommend 0w20 synthetic oil in all operating conditions in North America.
Generally, performance with such oil has been good, but many owners of early cars have
reported high oil consumption. Some engines were subject to a manufacturing defect where the
gaps between piston rings lined up and fostered excessive oil consumption. Other engines seem
to consume substantial oil, but not enough to indicate they have the piston ring defect. Subaru
has revised their definition of excessive oil consumption to be more than a quart every 600 miles
(previously a quart every 1,000 miles), which strikes many owners as ludicrously excessive.
Those owners may choose to use thicker oil simply to reduce the consumption.

Normal Street Driving

Synthetic API RC/ILSAC 0w20 has proven suitable for FB engines on the street. Top-tier
national brands and long drain versions can be extended well beyond the factory recommended
6,000 mile change interval. (7500 mi for 2011-2014) Such use is not wise during warranty, and
should be accompanied by UOA.

Heavy Use
Subaru of America (SOA) does not specify heavier oil for heavy use in the FB engines, but
stepping up to an API RC/ILSAC 30 grade oil for heavy use will provide some margin of safety.
Doing so poses some risk of denial of a warranty claim.

Modified Engines
Normal bolt-on modifications should not alter the oil recommendations for normal or heavy use.
Modified FB series engines are rare at the time of writing, so feedback on them is minimal,
choose with caution.

Naturally Aspirated, Direct Injected FA Series Engines

The FA20 is used only in the BRZ sports car (and its Scion FRS twin), it is similar to the FB20
but has direct injection and a higher redline, both of which indicate it might be a bit harder on
oil. The FA20 shares its D4-S fuel injection system with some Lexus models and has both direct
and port fuel injectors. Judging by Lexus experience this engine is less likely to show fuel
dilution and intake valve deposits than many direct injected engines. Still more prone to fuel
dilution than port injected naturally aspirated engines.
Normal Street Driving
Synthetic API RC/ILSAC 0w20 has proven suitable for FA20 engines on the street. Extension of
drain intervals past the factory recommended 6,000 miles is possible, but use caution as the
direct injection system can sometimes show fuel dilution that the port-injected FB series does not
experience. Subaru allowed 7,500 mi change intervals in the FA20 during the 2013 and 2014
model years, then shortened the recommendations to 6,000 mi for 2015. Use of UOA is wise if
extending intervals.

Heavy Use
Subaru recommends heavier oil of an unspecified grade for heavy use in FA engines, Stepping
up to an API RC/ILSAC 30 grade oil for heavy use will provide some margin of safety. If
ambient temperatures are above 0°F, use of a quality API RC/ILSAC 10w30 synthetic should be
sufficient for heavy use.

Modified Engines
Normal bolt-on modifications combined with a near-OEM redline should not necessitate altering
the oil choice otherwise made for operating conditions. API RC/ILSAC 0w20 for the street and
slightly heavier oil for heavy use should be adequate. Aftermarket forced induction systems have
the potential to make the engine very demanding on oil as do “race-built” naturally aspirated
engines. In such a case, consult with your engine builder, or at the very least follow
recommendations for a modified turbo FA engine (below).
9.3.4 Turbocharged, Direct Injected FA series Engines

The FA20 DIT was introduced in the 2014 Forester XT and replaced the EJ25 in the 2015 WRX.
Direct injection is new to Subaru and carries with it a couple potential problems highlighted by
other manufacturers, namely fuel dilution of the oil and carbon deposits on intake valves. Early
UOA results in FA20 DIT engines indicate fuel dilution is a substantial problem, and Subaru's
shortening of the change interval from 7,500 to 6,000 miles for 2015 provides some evidence
they're aware of the problem. No reports of carbon buildup have surfaced as of the time of
writing, but it is a problem owners should watch. The engines are still fairly new, but the
incidence of spun bearings in the FA turbo engines appears substantially less than their EJ series
forbears, so the necessity of using thicker oil in these engines seems less than in the EJ. The
FA20 DIT already uses a thicker oil than its naturally aspirated brother, the FA20 (5w30 vs
0w20). The potential for carbon deposits, though may make a low-SAPS European oil (ACEA
C3 or C4) a good choice, with the extra HT/HS safety margin coming as a bonus.

Normal Street Driving

Synthetic API RC/ILSAC 5w30 has proven suitable for FA turbo engines on the street.

Heavy Use
Subaru does not specify heavier oil for heavy use in the FA20 DIT, but stepping up to a Heavy
Duty 30 grade or 40 grade oil for heavy use will provide some margin of safety. Doing so poses
some risk of denial of a warranty claim.

Modified Engines
Owners of modified cars are wise to use HD 30 grade oils at a minimum for street use. Modified
cars receiving heavy use should have a high capacity oil cooler installed in addition to the use of
HD oil. For occasional heavy use, modified cars can get away with shear stable HD 40 grade oil
conforming to ACEA A3/B4, or a combination of E7 or E9 and a gasoline specification like API
SM.
9.4 Subaru Oil Requirements – Warranty Redux

A lot of discussion on the forum centers around what oil is required to maintain warranty
coverage. The reason for so much debate about warranty requirements is Subaru’s confusing and
often self-contradictory recommendations in the owner’s manual, which are often just that –
recommendations without clear requirements. Generally between 2002 and 2010 Subaru
recommended conventional API RC/ILSAC 5w30 oil. During this period the manuals also
mention thicker grades for hot weather and heavy use. In 2011 the recommendations changed
substantially with the introduction of 0w20 oil for the FB engine and the removal of specific
grades heavier than 5w30 for the EJ engine. Since any Subaru older than 2009 is beyond the
5yr/60k powertrain warranty, we will ignore the older manual recommendations. Owners of
older cars (or newer cars with more miles) can select oil based purely on technical properties
without regard for Subaru’s warranty requirements.

Prior to the synthetic recommendations, the 2009-2010 Impreza manual is typical. The
recommendations are the same for both turbo and naturally aspirated cars.

Figure 9 – 2010 Owner's Manual Oil Recoommendations

http://techinfo.subaru.com/search/index.html
Page 11-12 recommends oils which conform to API SM + Resource Conserving, or ILSAC GF-
4, and shows a table which includes 5w40 grade oil for any temperature. Great, but 5w40 cannot
be Resource Conserving or ILSAC GF-4, the best you can do in a 5w40 grade oil is plain API
SM. The next page makes recommendations for severe driving conditions which require only
API SM or SL, but fails to mention 5w40 as an acceptable grade, listing only 30, 40, 10w50,
20w40 and 20w50! To pile stupidity on top of confusion, there’s no advice on what weather
conditions the severe service grades are suitable for. You certainly wouldn’t want to use straight
40 grade at 0°F.

So, if I tow a trailer in a Minnesota winter (the author does), which oil should I be using? If I
have an engine failure while using a 5w40 oil, will Subaru deny my warranty claim because the
oil isn’t Resource Conserving? None of this is clear.

As bad as the 2009-2010 manual is, it at least has a chart allowing 5w40 oil. The 2011-2015
manuals start the oil section by advising you to use “Subaru approved” oil, but then tell you to
contact your dealer to find out what that oil is. Brilliant. So my dealer, who’s dumber than a box
of rocks, gets to decide which oil is approved? Presumably the only “approved” oil is one that
comes in a Subaru bottle, at great expense. An alternate engine oil is apparently OK if the
approved oil isn't available. The alternate oil is apparently synthetic 5w30, but then the manual
contradicts itself in the very next sentence saying conventional 5w30 is OK if synthetic is
unavailable. Huh??? How do we define available? Anything is available through the internet if
we're willing to wait a week. The very next paragraph mentions using 5w40 conventional oil for
replenishment only. NASIOC community members have looked hard for any 5w40 oil marketed
as conventional and come up empty, and any way, what’s wrong with 5w40 synthetic? Why
even mention the oil being conventional? That implies synthetic is somehow not acceptable.
Well done, Subaru.

As bad as the EJ engine recommendations have become, the FB series engines really take the
cake. They make the same “approved oil” pitch as the EJ manuals, give an alternate oil as
synthetic 0w20, then allow a non-existant conventional 0w20, and finally recommend
conventional 5w30 or 5w40 for replenishment, completely ignoring the more appropriate 5w20
conventional and synthetic options. Brilliant, Subaru, just brilliant. No wonder owners are
confused.
The BRZ manual contains the lone remaining reference to heavy use. It says “An oil with a
higher viscosity ... may be better suited if the vehicle is operated at high speeds, or under
extreme load conditions.” but doesn't give any guidance on what constitutes high speeds or
extreme loads, nor does it detail exactly what “heavier oil” means. Do we use 30 grade? Or even
40? While the manual language frees the owner to use almost any oil in heavy use, a recalcitrant
dealer can use the ambiguity as an excuse to deny warranty coverage and force the owner to seek
redress in court. Courts are generally more expensive than engines, so few owners actually do
this.

So what's the concerned owner to do? The strictest interpretation of the current owner's manual
recommendations is to use “Subaru Approved” oil from the dealer. Those who lie awake at night
terrified of a non-warrantied engine failure should simply have the dealer change the oil, confirm
they're actually using the Subaru oil and save the receipts. There, wasn't that easy? Aren't you
glad you spent all that time reading this guide? Owners who are this risk averse shouldn't be
participating in any activities which Subaru could consider abuse, which is also grounds for
denying warranty claims. That means no autocross and no track days.

One step up the risk ladder is using an oil from the exact same category as the Subaru brand oil.
That means using an API RC/ILSAC synthetic oil in 0w20 or 5w30 grade depending on your
engine. This strategy is very low risk, but only provides documentary protection against a denied
warranty claim, the oil itself doesn't protect the engine any better than the Subaru oil during
heavy use.

BRZ/FRS owners with less risk aversion who use their cars heavily can step up to a 30 grade oil
under the “heavy use” recommendations in the manual, but 2011+ EJ owners are in a bit of a
bind. Using a Heavy Duty 5w30 oil which meets API SM (or newer) is the lowest risk strategy,
unfortunately such oils are rare and expensive. Currently all such oils are designed for European
OEM applications and cost $10+ per quart, but for the semi-nervous owner they're the best
choice. See oil recommendations below.

A note on modifications – if you've modified your car, you shouldn't be counting on warranty
coverage any way, and you should select the oil which best suits your operating conditions
regardless of Subaru's recommendations. That's the ethical thing to do, modding a car is a pay-to-
play activity.
10 Favorite oils in various categories
While an effort has been made to ensure the information in this section is correct at the time of
writing, oil makers change their formulations constantly, sometimes changing the category of an
oil in the process. For instance, during a recent reformulation Valvoline Maxlife Full Synthetic
5w30 changed from being a Heavy Duty oil conforming to ACEA A3, to a Light Duty resource
conserving oil conforming to ILSAC GF-5. Check current product data from the manufacturer if
in doubt, links to product data have been included where available.

10.1 Light Duty 5w30 Conventional oils

All RC conventional oils from national brands perform similarly, there's no need to detail the
small differences between them here. If you want better oil, spend a few bucks and buy a good
blend or synthetic. Good blends like Valvoline Maxlife and Pennzoil Gold and lower tier
synthetics like Quaker State Ultimate Durability are only slightly more expensive than
conventional oils. If you must split hairs to differentiate between conventional oils, choose one
meeting Ford M2C and/or GM dexos specs.

10.2 Resource Conserving 0w20 synthetic oils.

Pennzoil Platinum 0w20 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5
Excellent oil from Shell Oil Products US (SOPUS), but no ACEA or dexos rating. SOPUS no
longer publishes HT/HS for Pennzoil Platinum

Pennzoil Ultra Platinum 0w20 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A1/B1, GM dexos1.
Excellent oil from Shell Oil Products US (SOPUS) Has ACEA A1/B1 and dexos, but also higher
NOACK than Pennzoil Platinum, right on the ACEA limit of 13%. SOPUS no longer publishes
HT/HS for Pennzoil Ultra Platinum.

Mobil 1 Advanced Fuel Economy 0w20 ***8211; HT/HS 2.7

API SN +RC/ILSAC GF-5, ACEA A1/B1, GM dexos1, Ford M2C.
Excellent oil from Exxon Mobil, the only 0w20 claiming adherence to Ford's WSS-M2C947-A
spec for 5w20 oil, which implies it meets the stricter turbocharger deposit specifications (TEOST
33C) for 5w20. UOA has shown fairly poor shear stability for M1 AFE 0w20, however. As of
5/28/15 Mobil's web product page and PDS no longer list ACEA A1/B1 for AFE 0w20,
indicating a possible recent formulation change.
Castrol Edge 0w20 ***8211; HT/HS ??
API SN +RC/ILSAC GF-5, ACEA A5/B5, OEM approvals
Excellent oil from BP. Latest version has titanium additive trickled down from older
***8220;Edge with Titanium FST***8221; formulation (now called Edge Extended). Castrol
doesn't publish real HT/HS or many other specs like MRV and CCS for their products, only spec
minimums, which makes detailed evaluation hard.

Castrol Edge Extended 0w20 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A5/B5, OEM approvals
Excellent oil from BP. Broadly similar to Mobil 1 Extended Performance. Guaranteed by Castrol
for up to 15,000 miles. Now that the titanium additive package has trickled down to 'regular'
Edge, there's little reason to use the 'Extended' version for change intervals of 10,000 miles or
less. Castrol doesn't publish real HT/HS or many other specs like MRV and CCS for their
products, only spec minimums, which makes detailed evaluation hard.

Subaru (Idemitsu) 0w20 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5
The Subaru branded oil made by Idemitsu Lubricants of Japan. Generally similar to top-tier
domestic brands like Pennzoil and Mobil 1. No ACEA rating, but for an oil marketed in the US
as an OEM lubricant, that's hardly expected. Product data is hard to come by for this oil.

10.3 Resource Conserving Light Duty 30 grade Synthetic oils

Amsoil OE and XL series 5w30 ***8211; HT/HS 3.3

API SN +RC/ILSAC GF-5
The difference between the XL and OE series is their TBN, XL oils are designed for longer drain
intervals, while OE oils are designed for OEM recommended change intervals. Other than TBN
the base stocks and additive packages are the same. The 5w30 grade of these oils have some of
the highest HT/HS viscosities within the RC 5w30 category, and are thus interesting to turbo
Subaru owners highly concerned with warranty coverage, who wish to conform to the strictest
interpretation of the owner's manual recommendations.

Amsoil Signature Series 5w30 ***8211; HT/HS 3.1

No Formal Certifications
Mentioned here only because it dos NOT carry API or ILSAC certification, and therefore could
pose a risk to warranty coverage. Amsoil uses tricky ***8220;recommended for***8221;
language in their marketing which implies conformance to API and ILSAC standards, but
Signature Series is not officially certified. It is excellent oil, but its claim to fame is very long
drain intervals, similar to Mobil 1 Extended Performance and Castrol Edge Extended. Unless
you're extending your drain intervals past 10,000 miles, there are better ways to spend $10+ per
quart on motor oil.

Pennzoil Platinum 5w30 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A5/B5, + OEM approvals
Excellent oil from SOPUS. Shell no longer publishes the HT/HS viscosity for Pennzoil Platinum.

Pennzoil Ultra Platinum 5w30 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A5/B5, + OEM approvals
Generally similar to Pennzoil Platinum but with increased levels of cleaning additives and higher
TBN, but also with higher NOACK. Shell no longer publishes the HT/HS viscosity for Pennzoil
Ultra Platinum.

Mobil 1 5w30 ***8211; HT/HS 3.1

API SN +RC/ILSAC GF-5, ACEA A1/B1?, OEM approvals
Mobil 1 5w30 has a bit of a bad reputation on NASIOC because of the sub-par performance of
the early API SM formulation, the latest API SN formulation performs as well as any ILSAC
5w30. ExxonMobil warrants Mobil1 for use up to 10,000 miles. As of 5/28/15 Mobil 1's web
information page no longer lists any ACEA specs for M1 5w30 (non-EP) the PDS lists
A1/B1, indicating a possible recent formulation change. Formerly M1 5w30 was ACEA
A5/B5

Mobil 1 Extended Performance 5w30 ***8211; HT/HS 3.0

API SN +RC/ILSAC GF-5, ACEA A1/B1, OEM approvals
Generally similar to 'standard' Mobil 1 with slightly lower HT/HS. ExxonMobil warrants M1 EP
for use up to 15,000 miles, but for change intervals less than 10,000 miles there is no reason to
use M1 EP in place of 'standard' Mobil 1.

Castrol Edge 5w30***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A5/B5, OEM approvals
Excellent oil from BP. Latest version has titanium additive trickled down from older
***8220;Edge with Titanium FST***8221; formulation (now called Edge Extended). Castrol
doesn't publish real HT/HS or many other specs like MRV and CCS for their products, only spec
minimums, which makes detailed evaluation hard.

Castrol Edge Extended 5w30***8211; HT/HS ??

API SN +RC/ILSAC GF-5, ACEA A5/B5, OEM approvals
Excellent oil from BP. Broadly similar to Mobil 1 Extended Performance. Guaranteed by Castrol
for up to 15,000 miles. Now that the titanium additive package has trickled down to 'regular'
Edge, there's little reason to use the 'Extended' version for change intervals of 10,000 miles or
less. Castrol doesn't publish real HT/HS or many other specs like MRV and CCS for their
products, only spec minimums, which makes detailed evaluation hard.

Subaru (Idemitsu) 5w30 ***8211; HT/HS ??

API SN +RC/ILSAC GF-5
The Subaru branded oil made by Idemitsu Lubricants of Japan. Generally similar to top-tier
domestic brands like Pennzoil and Mobil 1. No ACEA rating, but for an oil marketed in the US
as an OEM lubricant, that's hardly expected. Product data is hard to come by for this oil, the
MSDS linked above was posted by user ***8220;Toombs***8221; in 2013 and may not be
current.

Mobil 1 Advanced Fuel Economy 0w30 ***8211; HT/HS 3.0

API SN +RC/ILSAC GF-5, ACEA A5/B5, OEM approvals
The only ILSAC 0w30 oil readily available in the USA. Performance is generally similar to high
quality 5w30 synthetic oils except at very low temperatures where M1 AFE excels. UOA has
shown M1 AFE 0w30 to be more shear stable than its 0w20 sibling. Alaskan and Canadian
owners of Subarus calling for 5w30, this is your winter oil. Canadian owners can also use the
equivalent from Petro Canada which is generally less expensive. Because this oil is a different
grade than called for in the owner's manual there is a slight chance of warranty issues when using
it.

10.4 Light Duty 30 grade Synthetic oils (non Resource Conserving)

Mobil 1 High Mileage 5w30 ***8211; HT/HS 3.3

API SL, ACEA A5/B5
A high quality oil with high levels of zinc and phosphorous additives which necessitates
certification to the older API SL standard. A good choice for owners of older Subarus, but could
pose a threat to warranty coverage in newer cars which call for use of API SM or newer oil.

Motul 8100 Eco-Energy 5w30 ***8211; HT/HS 3.4

API SL, ACEA A5/B5
A high quality oil with high levels of zinc and phosphorous additives which necessitates
certification to the older API SL standard. A good choice for owners of older naturally aspirated
Subarus which call for 5w30, but could pose a threat to warranty coverage in newer cars which
call for use of API SM or newer oil. Difficult to find and expensive, there is little reason to use
8100 Eco-Energy rather than functionally similar Mobil 1 High Mileage. Some Subaru dealers
offer Eco-Energy as their premium fill.
10.5 Heavy Duty 30 grade Synthetic oils

Mobil 1 ESP Formula 5w30 ***8211; HT/HS 3.58

ACEA C2, ACEA C3, API SN (engine protection only), OEM approvals
A Low-SAPS oil designed for use in European passenger car diesel engines with particulate
filters, but also for use in gasoline engines (VW 504). One of very few HD 30 grade oils which
claim any conformance to API SN or SM standards. An excellent choice for owners of newer
turbo Subarus who want to use an API SM or SN 5w30 for warranty reasons, but want the extra
bearing protection afforded by a Heavy Duty oil. Builder specifications and ACEA C3 ensure
good shear stability which has been confirmed through UOA. M1 ESP is not suitable for long
drain intervals due to it's Low-SAPS, low TBN nature. Owners of direct injected European cars
have noted that M1 ESP substantially reduces intake valve deposits compared to higher SAPS
oils. Available at PepBoys and some auto parts stores catering to independent mechanics like
CarQuest and NAPA. Generally quite expensive at around $10/qt.

Castrol Edge Professional European OE 5w30 ***8211; HT/HS 3.5?

API SM, ILSAC GF-4 (engine protection only), ACEA C3, OEM approvals
One of very few HD 30 grade oils which claim any conformance to API SN or SM standards. An
excellent choice for owners of newer turbo Subarus who want to use an API SM or SN 5w30 for
warranty reasons, but who want the extra bearing protection afforded by a Heavy Duty oil.
Builder specifications and ACEA C3 ensure good shear stability which has been confirmed
through UOA. Similar in many ways to M1 ESP 5w30, but capable of longer drain intervals due
to being a High(er)-SAPS oil. Sold through European new car dealers (VW and BMW) and some
specialty auto parts houses. Generally very expensive. Castrol doesn't publish an honest HT/HS
for OE 5w30, only spec minimums.

Valvoline Synpower MST 5w30 ***8211; HT/HS 3.5?

API SN, ILSAC GF-5(emission system only), ACEA A3/B3/B4 & C3, OEM approvals
One of very few HD 30 grade oils which claim any conformance to API SN or SM standards. An
excellent choice for owners of newer turbo Subarus who want to use an API SM or SN 5w30 for
warranty reasons, but who want the extra bearing protection afforded by a Heavy Duty oil.
Builder specifications and ACEA specs ensure good shear stability. Similar to M1 ESP 5w30.
Difficult to find, but sold through some European new car dealers and some specialty auto parts
houses. Generally very expensive. Valvoline doesn't publish HT/HS for MST 5w30, 3.5 is the
spec minimum.

Motul 8100 X-Clean 5w30 ***8211; HT/HS 3.5

API SM, ACEA C3
A Low-SAPS oil designed for use in European passenger car diesel engines with particulate
filters, but also for use in gasoline engines. One of very few HD 30 grade oils which claim any
conformance to API SN or SM standards. An excellent choice for owners of newer turbo
Subarus who want to use an API SM or SN 5w30 for warranty reasons, but want the extra
bearing protection afforded by a Heavy Duty oil. Builder specifications and ACEA C3 ensure
good shear stability. 8100 X-Clean is not suitable for long drain intervals due to it's Low-SAPS,
low TBN nature. Some Subaru dealers offer X-Clean as their premium fill.

Castrol Edge 0w30 (a.k.a. German Castrol or GC) HT/HS ***8211; 3.5?
API SL, ACEA A3/B4.
The only reasonably available Heavy Duty 0w30 oil in the United States. GC is an excellent
choice for heavy use in cold climates. High levels of zinc and phosphorous additives necessitate
certification to the older API SL standard. Available at AutoZone stores. Castrol doesn't publish
real HT/HS or many other specs like MRV and CCS for their products, only spec minimums,
which makes detailed evaluation hard.

Red Line 5w30 ***8211; HT/HS 3.7

No formal certifications
A street/track oil composed of PAO and Ester base stocks and a very robust additive package
high in zinc, phosphorous and especially moly. The moly acts as a friction modifier, giving this
oil very low friction for its viscosity. The PAO/Ester base stocks combined with a lack of VII
additives give this oil one of the best temporary shear ratios available. It is also extraordinarily
shear stable, often showing no shear at all in UOA. These characteristics make it possibly the
best oil available for heavy use in high temperatures, but performance improvements over 5w40
mixed fleet oils are dubious and the cost is double. Similar cold weather performance to 5w40
synthetics, not up to the best Light Duty synthetics. Despite a high initial TBN, rapid depletion
means this oil is not suitable for extended drain intervals.

Motul 300V 5w30 ***8211; HT/HS 3.5

No formal certifications
Another street/track oil similar in purpose and composition to Red Line, but even more
expensive. Given the greater availability and lower cost of Red Line, there's little reason to use
300V. Mentioned here only because it often comes up as the ***8220;ultimate***8221; oil in
forum discussions. Low TBN requires short change intervals.

10.6 Heavy Duty 40 Grade oils.

Shell Rotella T6 5w40 (RT6 for short) ***8211; HT/HS 3.8?

API SM, API CJ-4, ACEA E9
A long time forum favorite for turbocharged Subarus, which mostly means the EJ series.
Marketed for HD diesel applications, but designed for use in gasoline engines too as evidenced
by the API SM certification. Known as a mixed fleet oil because commercial vehicle operators
with both light duty gasoline vehicles and heavy duty trucks can stock one oil for their entire
fleet. Excellent shear stability ensured by ACEA E9. Robust enough for even highly modified
engines in heavy use. If your engine needs a more robust oil than RT6, you should be looking for
mechanical issues and/or installing an oil cooler. Cold weather performance is very good, but not
quite as good as the best Light Duty 5w30 synthetics, adequate for the vast majority of users in
the lower 48 states. RT6 is available at Walmart for about $5/qt, an excellent value. Shell does
not publish HT/HS for RT6, but 3.8 is a reasonable minimum guess based on grade and
specifications.

Mobil 1 Turbo Diesel Truck 5w40 (TDT for short) ***8211; HT/HS 3.8
API SM, API CJ-4, ACEA E7
Another mixed fleet oil marketed for diesel trucks but designed to work well in gasoline engines.
All comments for Shell Rotella T6 apply to Mobil 1 TDT. Slightly more expensive than RT6,
which accounts for its lower popularity.

Chevron Delo Synthetic 5w40 ***8211; HT/HS 3.8?

API SM, API CJ-4, ACEA E7
Another mixed fleet oil marketed for diesel trucks but designed to work well in gasoline engines.
All comments for Shell Rotella T6 apply to Delo Synthetic 5w40. Slightly less available than
RT6, but now in most Walmart stores. Chevron does not publish HT/HS for Delo 5w40, but 3.8
is a reasonable guess based on grade and specifications.

Mobil 1 0w40 ***8211; HT/HS 3.8

API SN, ACEA A3/B4, + OEM approvals
The most available 0w Heavy Duty oil in the United States. Carries many European OEM
certifications. Suitable for year round heavy use in cold or warm climates. Good shear stability
ensured by ACEA A3/B4, but not as good as ACEA E7 and E9 oils. Owners of direct injected
European cars have noted higher levels of intake valve deposits with M1 0w40 than with Low-
SAPS alternatives.

A note about 50 grade oils

Older Subaru owner's manuals mention 50 grade oil for heavy use. The benefits of using such
oils over the available HD 5w40 synthetic options are dubious. The available oils all have
substantial limitations, either low shear stability (Castrol 5w50, Eneos 0w50) or poor cold start
performance (M1 15w50).

Acknowledgements:

I'd like to thank Ron (Unabomber) for his blessing to post this, and Dennis (bluesubie) for all his
help editing.