Hopefully this is of interest

Back in August I posted the figures for my STi9 GL "straight out of the box". At that time I said that I would see what could be gained from the car purely from mapping leaving exhaust, induction etc std. I have to admit I made one change which was to fit a three port solenoid as a boost control assist.

Anyway I went to Power Engineering today to get a comparison. Apart from having to buy four new tyres due to getting a big nail right on the shoulder of the nearside rear I was not too unhappy at the figures. Big thanks to the guys for putting up with me ... car on rollers, discover nail, car off rollers, trip to Micheldever Tyres for four replacements, back to PE, car back on rollers straight away ... can't thank them enough and Merv/Steve (of Ecutek) who came down and ran the car.

Before 287 bhp and 276 ft lb torque
After 327 bhp and 323 ft lb torque

edited to add the run was done in 4th gear (of six)

http://members.aol.com/nathlang/BRDynoplot2.jpg

Car is totally std apart from mapping and the three port (oh a set of heat grade 8 plugs installed)

Fuel is my usual Optimax plus a bit of NF.

The top end is totaly restricted by the exhaust and std panel filter now, however much I tried bar wiring it shut the boost would not stay above 1.2 bar and the 327 was made at 1.19 bar boost, peak at 1.5 for max torque.

Have to say the car will run very high ignition timing, well over "normal" at 1.2 bar. Not saying how much though !!


So on to Phase 2, this will be in two stages but both will be with a full stainless 3 inch exhaust from turbo back. That will be in place by end Jan. 2004. Once verified then the system will become available. Phase 2a will be the exhaust plus a "good" panel filter in std air box, Phase 2b will be to use same exhaust but replace the std airbox/panel with an induction kit, this will require maf scaling as it will be a larger bore.

That will then move the restriction to inlet pipe and turbo (I hope).

Target for both these options .. well 350 would be good, I suspect thats the limit of the turbo. Boost will remain 1.5 bar peak ... i plan to hold 1.4 bar to the redline (if possible)although I suspect 1.3 is the reality.

Cuople of general observations, although the curves might look a bit "laggy" in fact, on the road, the car is not at all, the header and turbo combination is a real compliment to the six speed box, car "squirms" on accelleration, especially this weather.

From there Phase 3 will involve turbo/injector/fuel pump/fuel regulator upgrades. I plan to keep the top mount and go for 400-420 bhp on the std internals using the factory ecu remapped. Timescale for that end March 2004.

I won't be going internal on this engine ... 2.5 litre did you say ? (but with a decent redline)

Comment on the tyres, I was not sad to see the RE070's go even though they still had 1/3 of the tread left, replaced with F1's and, at last, I can throw it into a corner in the damp without the tail parting company. Oh and a thanks to Micheldever who, absolutely out of the blue, offered a full 4 wheel alignement to go with the tyres, that NEVER happens, always a 2 month wait so I made the most of it.

Finlly big thanks to Nathan for hosting the graph, hopefully have my own web space soon which I will then proceed to clutter with technical trivia !!

cheers

bob

[Edited by Bob Rawle - 12/19/2003 10:48:02 PM]

Bob

You distinguish yourself through a methodicial approach to development work, thanks for a cupola comments

17% extra torque is not to be sniffed at. Very impressive. Wonder what a decat and panel filter will give? Looking very interesting.

Very nice results and an objective write-up as always Bob :)

Given the mapping work so far do you think that the Spec C heads with larger ports & different cams are helping the engine achieve power more easily than the STi7/8 UK spec cars?

cheers,
callum

Difficult to quantify Callum, I think a considered view can be given as Phase 2 progresses, certainly it appears so however that is subjective.

Harvey I am quite looking forward to finding out.

bob

Thanks for keeping us updated with your progress.

Makes very interesting reading.

Looking forward to finding out the results of the next phase. :D

Nathan..

Bob, is the STi 9 using a VF35? Can I assume that the its a JDM STi 8 you have with the STi 9 engine in it?

I would be interested to see what you achieve with 1.5/1.6 bar like I'm running on mine.

Conrad

Bob, have you also got the boost / RPM graph?

[Edited by RSVR Racer - 12/20/2003 5:16:39 PM]

Its an STi9 ... there is only that for MY03 in Japan. It has the self same engine, headers, turbo (twin scroll) fuel system etc as the Spec C except the turbo is sleeve bearing not ball race, apart from that the same.

Boost v rpm graph ? Don't get that at PE it makes 1 bar at about 3400 on the road in 4th, 3200 in 5th and 3000 in 6th. I get 1.4 held in 2nd, third, fourth etc and 1.35 held in first (I love in gear wastegate and boost compensation) First peaks at 1.4, all others at 1.5.

On the run, peaked at 1.5, held 1.4 then tapered itself from about 6k down to 1.19/1.2 at 6800 then held that to 8000.

I wouldn't hold that boost to the redline on a std engine however I expect to achieve what I'm after with 1.35/1.4 bar above 7000 rpm.

bob

This is a JDM STi 9.

For some reason we called the Prodrive version which was the first official STi here a 7. Whereas there was a prior bug-eye version in Japan which was the 7 over there. The prodrive version was their 8 and the blob eye is their 9 but our 8. Nuff said ;)

The twin scroll units are not VF35 but VF36 and VF37 depending on whether you have a sleeve bearing or roller bearing turbo.

Nathan..

Damn beaten to it on speed of typing by Bob ;)

Nathan..

Bit more info, installed a Forge recirc in place of the std BOV. Modified to suit the boost I'm running/going to run. So far no issues with "maf jerk", crisper spool and better gear change transition. Marginal increase in peak boost (0.05 bar) but I will allow that for the moment. Factory bov cracked at 0.25 bar which is too low. Downside the car is very slightly more hesitant when very cold but that still vanishes after a couple of minutes driving as before. (this is because the ecu tries to run the fueling at stoich when the engine is stone cold.) I think that the fuel learning will compensate for that impact of the extra air so should revert to as before.

bob

Hi Bob,


I will soon be getting a TEK3 on my UK Turbo.
I run a full decat and panel filter; is it a good move to upgrade the turbo hoses and the std.BOV prior to the Remap?

ta.

J.

Definately change the turbo to intercooler hose, the plastic ones do split with increased boost, they are paper thin. Changing the bov b4 hand is advised if you plan to do that in any case as the boost response can be mapped accordingly.

For such little work done to the car to date, this shows how responsive the Impreza engines can be which is encouraging.

I also like the idea that you've started this thread. Hopefully a thread that will continue to run as you progress with the car, as it will inform others on how this particular model responds to component changes.

It's all very well buying tuning bits for the car but as we all know, they have to work in unison with each other to achieve the results. A turbo for example on one car may work better than on another, solely because of the combination of the other components within.

Good luck with the development Bob. The initial research and trial has obviously spurred you on!!

Alan :cool:

Thanks Alan, yes I will keep it going as the car progresses through its stages. All will go relatively "quiet" from end this week until probably end Jan as the exhaust is sorted, fitted and then remapped to suit. I also have a VF36 turbo to go on in place of the VF37, VF36 has the Titanium shaft and turbine wheel for better spool however the compressor side is the same afaik. Apart from doing something with the actuator it will remain a standard turbo to see what it can give.

Bob

Thanks Bob. In addition; a very interesting read!

OK, dropped the car off today for the exhaust development, now I have to wait patiently however its not going to be too long from the conversations held. Also took the STi5 to demonstrate the smooth path of the HKS exhausts from front to back for best gas flow.

Bob

Can't wait to see the difference the exhaust makes.

Nathan..

Likewise. Mine is far too quiet!

keep us posted on ur exhaust progress bob, i might have to have one for mine :)

Bob following your progress with interest as always. Are you recommending 8 grade plugs these days for fairly modded cars?

And Bob is it the 7235-8 or 7119-8 plugs you are using?? :)


Skassa

[Edited by Denmark - 1/3/2004 10:42:36 AM]

John ... yes is the short answer, Skassa I am using both, 7119's in the STi5 and the 7236=5's in the STi9.

cheers

bob

Bob,
Which 3 port soleniod are you using? Denso16102AA013 i guess.
still with the restrictor in place?

Thanks, Roy.

The standard Subaru solenoid, restrictors are individual to a car, turbo, boost run etc as is the pipework. That is if you want the best from the setup. Just played with an Evo this am, changed the restrictor dia by 0.1mm (reduced), tweaked the boost response settings and made a huge difference to pickup and held boost, as an example. Pity the owner had been misled into filling with BP Ultimate, even with two bottle of Millers it detted, his normal brew is Optimax plus one bottle Millers, huge difference so be careful out there.

A few observations I have made....

At about 30 degrees ambient temperature, it's difficult to get more than 245-250 g/sec through the engine, regardless of what is done with the exhaust.

In cooler conditions the airlfow will extend up to roughly 270 g/sec, which should make 350 BHP if everything else is in order.

The Prodrive catted downpipe does not affect the achieved airflow, I also could not detect any decrease in tolerable ignition advance even after repeatedly loading the car up, including a few visits to the speed limiter (at 255 kph, which is well off the end of the speedo, they do over-read quite a lot up there).

Using a standard downpipe with the cat in place does not appreciably reduce the achievable airlfow. It does slug the turbo spool, and it does have a marked impact on tolerable ignition advance after repeatedly loading the engine up. Result is that the engine is flowing the same amount of air, but making less power and torque with it.

I found it impossible to achieve 1.3 bar of boost at 6400 RPM or higher with a stock turbo regardless of what is done with the exhaust, the compressor chokes up. A good turbo will make 1.25 bar there, a bad one 1.2 bar. With an uprated turbo and headers it will make over 400 BHP at 6800 RPM with 1.45 bar.

The most efficient figures I've seen to date on an STi engine are 321 BHP at 1 bar with a full decat and decent panel filter. The most efficient figure I've ever seen from a Subaru engine with my own eyes was 507 BHP at 1 bar.

The boost threshold on a New Age STi with P15 equipped turbocharger is typically around 2600-2700 RPM, on a P18 equipped turbocharger is around 2900-3000 RPM, but it does depend on cam timing and exhaust choice. Add a few hundred RPM for a stock catted pipe. Prodrive sports pipe seems to yield the same boost threshold as a decatted pipe (it's like the cat isn't there).

Will be fascinating to see what the effects of the exhaust work will be on the twin scroll turbo :)

Cheers,

Pat.

Sorry for the off topic question Bob but Pat, have you mapped any Milltek equiped STi's and if so, how would you compare them to the Prodrive STi's with regards to spool and flow?

Matt :)

Getting well over your quoted g/sec figs with the std exhaust and induction Pat and matching your cool weather figs in the summer. You don't seem to be talking about the STi9. Which MY with which turbo's are you talking about here ? Weak actuator on the VF36 and 37's what with the twin wastegate design.

Don't plan to consider P15/18 back end, however there is one turbo that would actually fit the bill well, last visit to Brunters was a real suprise on an Sti8 with this installed. Motec management though to cope with cylinder by cylinder descrepancies.

The STi9 is a departure from anything else, the reason for this thread was that and nothing else, happy to discuss anything as always, but would prefer if THIS thread focused on the new generation not the old.

Bob

Matt, I've done C specs with the Miltek exhausts, mail me off line for comment if you want. Its not relevent here.

cheers

bob

Matt,

the only cars that I've done with Milltek on have been mongrel systems, a little bit from everywhere... not a whole system as such, so it would be difficult to compare to the Prodrive item.... to further complicate the comparison, I've remapped a few PPP cars and got a significant improvement, so what does one compare? The Milltek will come out top if you compare a remapped car with Milltek to a PPP, but if you remap the PPP too, it becomes a little more interesting.

If you need a catted pipe I'de suggest a Prodrive pipe, otherwise a good aftermarket free flowing pipe will do :)

Bob,

It would be interesting to have a look at the compressor wheel spec on the VF36/37, if it's getting 270 g/sec in hot weather then it's doing quite well compared to a VF35/30 (and obviously better than the ubiquitous TD05). As I stated, the exhaust does not appear to influence the airflow, only what you can do with that airflow, so I wouldn't expect your air throughput to increase significantly, but you should find some more timing if nothing else. Indeed if you're already getting 270 g/sec in hot weather you should make an easy 350-360 BHP when you get a decent exhaust on there :D

You're right, I wasn't talking about the STi 9, I should have made that clearer from the outset. For the record most of the figures are for New Age STi pre-version-9 of various origins:) This should make it interesting to compare the new generation STi models to the previous... already it seems pretty clear that the new turbo / header combination works somewhat better than the older cars... would be interesting to see what effect the exhaust has on the version 9, I would expect it to pull in around 2700-ish, but it would be cool if it came in earlier :) It will also be interesting to see how the thermal efficiency of the new engine compares to the version 7/8....

Cheers,

Pat.