Subaru WHP numbers- A serious Discussion

Here you go Mike. This is a pretty good read on Suby manifold and header choices.
http://www.iwsti.com/forums/engine-power-performance/50286-manifold-header-dyno-comparo-results.html

 

It was nice of those guys to do all those tests, the thing that I don't like about it is that they used the factory ecu. To do it correctly you really need to eliminate all variables and the only way to do that is to do it with a standalone ecu. The factory ecu retains too much control.

Just from what I have seen the Perrin headers look like they will make good power on a stock turbo setup, or maybe a small bolt on like a green. After that I think the primaries are probally too small.

Matt

 

Also headers may not show as much gain on a stock turbo car because the turbo is the actual restriction.

 

Mo Powa

 

I thought it was just that most people that buy Subaru's want good driveability and a good powerband instead of all out high hp numbers, hence the reason of a performance sedan. Or maybe that was just me. Otherwise why didn't people just go with a Supra instead? I would like to see some high hp/tq Subaru's, but I personally only want a little more than stock.

 

You don't have to sacrafice powerband to get big numbers. A gt35r STI still has plenty of powerband but it makes 500whp.

My GT30R drives better than any stage 2 car I've ever driven. Pople obsess over spool, if you drive a car that doesn't "spool" until 4000rpms you will rethink the way you feel about it.

Matt

 

Well I think for why the power with same boost is diffrent on subaru's compared to hondas is mostly due to the compresson of the motors!

As far as power you dont need a shit ton of power to be quick! In KY we have a hybrid that dyno's 349hp pump gas and 460ish on ras gas and runs 10.80s all day long @ 128!

I say just keep it plan and simple to make good power!

 

That's true, I totally forgot the full ball bearing turbo's. I just remember riding in my buddies 18g sti and 6 spd swap, and it felt like it had to wait a little longer for the pull than on my stock wrx. When it did pull, it was pretty hard. I guess it's just a difference in driving styles though, he liked longer straights and such while I like the quick turn turn turn turn turn and I don't need all that power while he did.

Maybe you could stick a 30R on my car and I might see the light

 

Not necessarily true...

I have not seen 'impressive' numbers from higher compression builds vs stock-ish compression builds.

Also, from my personal experience...

EA82T. The 1.8 SOHC, 8V 1985~1994 turbocharged SUBARU engine found in Leone/DL/GL/GL-10/XT/RX.

My GL-10 wagon doesnt count since I did all my R&D and engine blowing up with that car.

I ran one single stock engine for a long time in my RX, stock compression was 7.7:1. Made 165hp/225tq to the wheels on a Dynojet dyno. Stock ECU. Only mods were: STi TMIC, TWE header, TWE downpipe, Greddy Profect B spec II bost controller and a custom intake. Ran 14.5 PSI to redline. Full boost by 2700 rpm. it currently holds the world record for power on a stock turbo/stock engine/stock ECU EA82T vehicle.

It was reliable (as reliable as an EA82T could be), easy to drive, predictable and made nice power. Usable powerband from 2500 to 6500 RPM. Stock redline was 6500. Eventually the stock composite headgasket blew...I replaced it. The engine still lives in JWX's RX. Total mileage on that engine is about 260,000 miles (I aquired it at 163,000 miles). Has seen several track days and drag strip runs. It still lives today and runs strong.

Next, and tried under 3 different setups...EA82T with the N/A bottom end. Resulted in 9.5:1 compression. I did replicate the setup mentioned above...but it never made good power. 135/145 to the ground on the stock ECU. Same on the MegaSquirt but that was running more boost and light detonation. I was unable to run more than 12psi on pump gas.

Next was a few different Air-water IC setups. Neither resulted in good power.

Next I upgraded the injectors and turbo. It was a little faster...however, at this point the rings were giving out and compression numbers dropped below 110 psi. Low tolerance for this engine was 145psi.

Over all the only real gain from the 9.5:1 was off boost throttle response. It was a little more friendly to drive below the spool areas...idle and up to 2400 rpm. Otherwise, it sucked and felt like it had no balls even with added boost/fuel/timing and even on the stock ECU.

The only other good thing about it was that the N/A shortblocks are literally free when you find one. Turbos...not so much. Both blocks have identical internals and block contruction...only different being a dished piston on the turbo vs the flat top on the N/A.

I wouldnt build a high compression turbo engine again. Highest I would go is 8.5:1.

I realize that the EA82 cannot be rightfully compared to the EJ series as it is an entirely different engine. But, we are talking SUBARU engines here and thats my 2cents on High Compression-Turbo engines.

 

great thread, btw...

desi already hit on this point, but i can tell you the answer to most of your questions in one simple response (from experience). IT ALL COMES DOWN TO $$$$$!

ALL subaru owners should know it ISN'T cheap to mod a subaru PROPERLY. mike, you of all people should know. how much is an average H22 block? hell i've found entire swaps (block, transmissions, wiring harnesses, etc. etc) for under $2000 (used of course). but you can't buy a new 2.5l SHORT BLOCK for less than $1500 anywhere now. so people are afraid of pushing the envelope, or buying cheap parts, or using ANYONE other than a reputable subaru tuner for fear of being out a ton of money. and like someone else said, a good percentage of us came from the world of Honda's or Nissan's where parts are nice and cheap.

personal example: i bought a USED '06 WRX transmission for $2200 bucks and then had it installed bringing the total to over $3000. I can buy a USED tranny for my acura (honda F22 motor, 5spd tranny), for $300 bucks and install it myself, because IMHO hondas are ALOT easier to work on with basic tools than the intricate subaru layouts. that translates to being able to blow through 10!!!! honda trannies before i am anywhere near the cost of 1 subaru tranny.

and as far as tuning goes. you can't really slap a part on a subaru and expect much performance (or reliability for that matter) out of it. whereas with a honda (as is seen with 1239057918275 ricers out there you can slap I/H/E/etc on them and not necessarily need tuning ( or sacrifice much reliability). i am not at all by any means suggesting it is ok to not tune any vehicle.

and like i said that can be applied to just about any part on a subaru vs. honda/nissan

 

this thread has blown my mind:eek3:

 

I still thing you can get great power out of are cars for cheap!

My wagon used to be the buget build!

MODS
- APV1 for $300
- ERZ DP for $80
- Invidia Up-pipe for $100
- Borsal Twin Tip Cat-back for FREE!!!!!!
- K&N SR used for $120
- E-Bay header $150
- Group Pro Tune by Jorge for $300

So for $1,050 I got 252hp & 248tq and on 110 278hp & 3008tq!

And this would be even beter on a STi!

 

^^Not everyone can get those deals everyday.

So in the end you paid ~$16.15 per HP on that deal.

 

Haha never looked at it liked that!

Yea I got some good deals far as exahust but eveything else is pretty normal for used stuff.

 

so are you thinking that wrapping the stock headers are pretty much mandatory to benefit flow?

 

stock header is cast, and the crossover is pretty well insulated...no reason to further wrap the stock unit.

Its the thin walled aftermarket pieces that need the wrap/coating in order to make power right off. Otherwise, it takes a few dyno pulls to get it up to temp.

 

Like Will said the stock manifold is so thick that it retains all the heat....and thats pretty much all it has going for it.

I think the full-race will be nice because its fairly thick stainless but most aftermarket headers are thin wall which is great for weight savings but not so good for heat retention.

Matt

 

Speaking of thin....again...

I've done several perrin/ebay/megan/etc header installs at the same time that someone wanted an EGT installed.

Guess what I had to do to make the EGT stay in?

Weld it.

Forget drilling and tapping threads into it. The tap would strip the SINGLE thread it cut into it. :rofl:

They all really are that thin.

Even the TWE header I had on the RX was like that....weld it!

The stock pieces I could cut about 3~4 threads into it...which would hold it fine. I still welded mine on for safety, as it was in a not so easy to reach place w/the engine in the car. Others I installed in easy to reach spots I never welded and they never backed out.

 

good thread

but we are still lacking raw data!
dyno's please!

 

Dyno's don't mean anything. They are just a tuning tool.

 

Thats not entirely true. At the end of the day it still takes a given amount of HP to rotate a given weight drum a certain speed over a certain amount of time. It is a tuning tool but a lot of people view the drag strip the same way.



They do mean something.

 

I agree in part with both. 20hp difference between two different dynos on two different days could mean nothing. A valid comparison on a dyno would have to be back to back runs on the same machine, hence making it a good tuning tool.

Then again, even with the exact same setup, two cars could produce vastly different results on the same dyno. There are too many variables, the variances of which can only increase when compounded. Therefore, in my mind, a dyno reading could give an idea of how the whole car works as a unit, but not necessarily how one particular element of the car performs. For example, I have filled up at some gas stations that made the car start pulling timing, whereas other gas stations keep the car plenty happy.

All in all, dyno numbers, though quite meaningful, should be taken with a grain of salt. But thats just my opinion

 

Thats just what Mike is talking about and I totally agree with him, two cars with the same mods should make almost exactly the same power. There are variables like bearing clearances etc....but we are talking about 1 or 2 hp not 40.

If you take two hondas and bolt the same turbo kit on them they are going to make with in 5 hp of each other, what Mike is asking is why is there such a huge diffrence in Subaru numbers.

Batlground recently did 3 turbo RSXs with Greddy turbo kits, 1st one made 247, 2nd one made 245, 3rd one mad 250whp. Those are all within 5hp, now someone tell me why you can't do that with a Subaru.

 

DING DING DING WE HAVE A WINNAR!

Thats what I am alluding too. 2 cars with the same setups on the same dyno with the same tuner, should be almost EXACTLY THE SAME NUMBERS.

WHats good for the goose SHOULD be good for the GANDER. IE A Ultimate Racing Rotated 35R Kit on "JOEs" STI should make +/- 10whp of "JIMs" STI with the same kit, same tuner , same dyno.

There is not enough difference in engine variables to mean 50whp difference.

For instance, i have dynoed prob close to 400 hondas if not more. I know that a Drag kit on a GSR at 10psi will make 230-240whp.

I know that a 2.0L ROds and pistons motor with a 35R will make 600whp at 30psi.

I know that a hytech header adds 10whp to a GSR/ITR. I dont need to see dynos or test it, i just know, ive done enough of them.

I guess it all comes back to R&D and time on the market. THe other makes and models have just had more years of development and the subaru community in the US is vastly behind.

I just think people need to start experimenting more because from what i see alot of people THINK they know what works, when they have never tried anything else.

HOW DO YOU KNOW IF YOU HAVENT TRIED?


-mike

 

On the contrary Dynos are the BEST tool to use IMHO.

Now i see what you are saying, you have to be carefull what you read and who you believe.

If Perrin says their new turbo manifold makes 50whp over stock and they provide the dyno, it was prob rigged and exaggerated.

But when individual users do dynos and shops post UNBIASED tuning information, its the most valuable piece of information you can find.

 

You guys are all correct....I worded myself wrong.

You have no clue how many people call BS about my car because it 'ONLY' dynoed 350 awhp.

 

Oh yeah, I did forget the dyno thing.

As soon as I get done with this waiter LOF/ROT/AF...I'll chime in.

Done!


My theroy on Dynos:

You can make a dyno read whatever you want it to read. Even back to back runs may not be accurate...say you get a hot burst of air on the temp sensor on the dyno...that'll throw numbers off 50hp/tq in one run. Humidity, temp, dyno setup and fuel are usually always different across the course of one day. Suppose you start tuning a car at 4pm...but you do not finish until 11pm. The temperature could change from 75*F to 30*F during that time. Thats going to show a massive HP difference from when you started...but in reality, you have not gained that much.

Suppose you have everything climate controlled. The IAT is consistent, the room temperature is consistent....etc etc...then I would consider the numbers being 'accurate' to an extent.

Also....you get Peter at APS saying 'My SR11tybillion00000SRT4thousand0000 turbo makes 600 hp on an STi. Get one of those turbos in the US and put it on...no more than 400 at best. Why? Different dynos. Different day. Different temp/humidity/barometric pressure/hemisphere/etc. They all play a factor. That and I'm sure Peter has his dyno all jacked up on crack to make those numbers, cuz his shitty ass turbos sure dont make DICK for power in the USA thats for fucking sure.

At dyno is a tool. Its like my 26" Matco ratcheting breaker bar. I use it to complete a task. Dynos are used to complete a task. Some tools are more useful than others. The 24" breaker bar has less uses than the 26" ratcheting breaker bar. The Dynojet has less uses than a Mustang.

The point is, its a tool. People use it for tuning, for showing differences in power, for driveability issues and in some states its used for emissions testing.

If I had a dyno, I'd be on it a lot for driveability. The Mustang and other load bearing dynos are infinitely useful for diagnosing driveability problems. I've used the dynojet for it but its just not as good; there's no way to keep the car in the trouble spot, it just passes thru it as if you were on the street.

Also, one bad thing about dynos....intake testing. Usually on the dyno you have the hood up and a fan blowing cool air over the engine bay. So, the IAT is usually near ambient. You do not get real world performance on the dyno. In the real world, there is no ambient air constantly blowing into an open engine bay. Well, there is...but it goes thru the radiator first to get heated up to 160~200*F. So, you get nice hot air sucked into the intake, lowering HP/TQ output. Why does the SPT and other similar intakes show a gain on a dyno but losses on the street? They've got the hood up during dyno testing. Thats why.


Otherwise, all things being equal...the SUBARU's have had a long standing record with the EJ205/EJ257/EJ255 in the USA that rarely do you get two identical cars read within 5~10hp/tq. I've seen 2.5 RS's and other N/As read within 5hp all day long in N/A trim. But for some reason the WRX's and STi's/etc seems all read vastly different from car to car.

Why?

Dunno. We could speculate all year about this.

My theory is going to be initial break in. But, until someone buys about 5 STis and does different break ins on each one and does same day dyno testing in a controlled environment...we just wont know.

 

^Ding.

 

Back to back runs at TS, my car would make anywhere between 415whp and 433whp IIRC, just depending on how much time was left to cool off between these 'back-to-back' runs (between 2min and I think 10min. Scott, correct me if I'm wrong).

Maybe the added variability is because of the reflashed stock ECU. So do standalones produce more consistent results?

 

2 minitues a part isn't really back to back, at that point the headers have had time to cool etc, The other part is the stock ecu. Stand Alone Ecus do produce more consitent results....back to back pulls on Jeremys car result in results that are less than 2 hp a part. Thats the way it should be.

 

Milo I'm really interested to see what happens when you put the catless downpipe on the car. I really want to see that car make the numbers it should. There is no reason you shouldn't be all over 500whp.

 

I'm really curious to see how much the cat takes away. I want to dyno the car with the new exhaust alone, without tuning for it. I bought a whole new turboback so we'll see if it makes a difference. From what people have said, another potential power blocker is the internal wastegate. However, I don't really understand how this takes away any power. I understand that an external might make boost control simpler, but how much difference in backpressure could there be if I am running upwards of 80% duty cycle.

 

Your going from an internal wastegate port that is 25mm compared to an external that is a MINIMUM of 38mm (or 44mm) if you go with TIAL. It also varies depending on spring rate

You wont lose backpressure (thats what the spring is for)

DO you really think there is 50whp hiding in a different exhaust on your setup? (Im being serious not condescending) Im interested to see the results. can you tell me what exhaust you switched too?

Good thread, thumbup 4 me omghi2u

 

External gates do sometimes make more power than internal, the problem with internal is that it creates turbulence right behind the turbo which screws with the flow. With an external its all dumping away from the turbo. Its also not taking up volume inside of the exhaust pipe.

 

Personally, I doubt there is much, if any, taken away by the cat or the wastegate. However, unless I am missing something, aside from the turbo, these seem to be the primary differences between the standard rotated setups and mine. I won't be changing the wastegate. So we'll see if dropping the cat will make any difference. As for the catback, I am changing it to drop the resonator. Once again, I doubt the resonator makes any difference at all, but it is worth a shot so no questions are left hanging.

Old setup: Helix catted downpipe, Blitz Nur Spec-R catback
New setup: Invidia V2 downpipe, Invidia N1 Racing catback

I don't take your question as condescending at all. If anything at all, we are all trying to learn about these cars. Whether changing the exhaust gains power or not, we can be sure that it will gain us a better understanding.

 

THis is why i like this board so much.

Keep us posted. We have an STI we are doing a rotated kit on now that i should have numbers from in the next week. Ill be sure to post the dyno so we can compare another setup

 

Milo, another option for you if you're looking for big peak gains would be to forget about all of the 3" catbacks and step up to a 3.5" unit. I know at least APS makes one.

 

I am curious too. If the owner and you guys don't mind, please also post up details on the build and tune. Things that could help would be injectors, EM, exhaust setup, boost, AFR, type of fuel, etc.

 

This thread is great!

 

I cant say anything right now as we are trying to keep it under wraps, but its nothing spectacular.

WHat i can say is:
CObb AP
35R
Stock motor

 

I thought about that but the additional cost didn't justify it. A lot of cars with bigger turbos have made plenty of power with 3 inch exhausts.

 

Hmmmm, I predict 540whp on Race Gas...430whp on pump gas

 

LOL we will see. THats big numbers

 

Yeah but its been there done that for Dan....3 years ago....

 

540whp hm... interesting

 

The first STI that Dan did with a GT35R made 542whp on a bone stock motor. That was at 30lbs.

 

wow, care to list the mods if you have them? I am very interested to see what's been done. You can PM them if you don't want to post it here. Thanks.

and how long did the block last?

 

Its still running, its never had a problem.

I can't say anything about the one they are building now. But the other one I was talking about was pretty simple.

800cc modded stock injectors
Batlground Built Upipe and Downpipe,
Precision GT35R with a .68 backside
TGV deletes
Walboro
Hydra

Thats pretty much it.

 

I know that car....it did make that power.

Unless I didnt say it...I'll say it again: The Blitz Nur Spec-R is not 3" all the way thru. The muffler section is 2.5" or less ID. You can shine a light down in there and see the restriction.

 

Ill post the dyno tomorrow. but heres pics:





It ran like that for a over a year before he took it back to stock and sold it. Last i heard he has a mean Vette