Looks interesting, subscribed :thumb:

The car has had it's first Dyno session, I'm sure Graham will be on here soon to give a full summary of the findings. The spec at the moment is completely standard apart from the Forged short engine...........the results were very good for a standard set up..........numbers to follow.

Ok so as Andy said I spent all day yesterday with the test car on the dyno. I have to give great credit to Andy and his engine building skills, it has done only 200 miles from the rebuild before strapping it on the dyno and doing the best part of 90 runs!!

I started of by reading and resetting the ECU to log the first fuelled run before it learnt a/f and ignition correction and generally analysing and testing the results of the ecu learning ability to my surprise.

It's important to note the following:
Forged bottom end only, but otherwise completely standard exhaust, turbo, airbox, filter etc etc.
Running on Vpower and had been since rebuild.
Logged intake temperatures were very similar to road temperatures, will do a road dyno run over the weekend to validate results.
Best dyno run was with the up-pipe hot, i.e one run then immediately another.

This was the standard map run on Vpower:
Lambda and Power

http://www.enginemapper.co.uk/testmule/std.jpg

So then I set about putting a map on it to get the most out of the setup. It's worth remembering that it still has all three catalytic converters in the exhaust setup.

Power and Torque:
http://www.enginemapper.co.uk/testmu...pped6plugs.jpg

Andy bought some 7 grade plugs from Ebay to have a go with them. To be honest I was very dubious about them as you might expect but the results were pleasently surprising. No other changes done from the run above to this one:
http://www.enginemapper.co.uk/testmu...pped7plugs.jpg

Not bad for a standard Bugeye Impreza! I had brought a pre-mix of Meth/Vpower with me to the dyno to try out on the car, but due to the Injectors running at 95% on both standard and remapped, (yes 268bhp was achieved with no extra fuel!), I had to modify the map to run low boost for the results to be meaningful in terms of percent of power increase:
http://www.enginemapper.co.uk/testmule/Meth%20test.jpg

The before,(red), is running with the injectors scaled to run 13% more fuel to make the AFR correct and low boost duty. The after, (blue), was running with extra ignition timing and netted an overall 10bhp gain and the same in torque.
I don't have a Vpower version to compare to see whether netted any gains without changing the map, but I will put that on the list for next time! Hopefully when we have some injectors big enough to cope with the Meth mix higher demand on fuel AFR.

So to sum up:
Standard, 220.4bhp and 205lb/ft.
Vpower map on NGK 6 plugs, 261.2bhp and 278lb/ft.
Vpower map on NGK 7 plugs, 268.4bhp and 282lb/ft.
20% Meth/Vpower netted 5% power increase when tuned.

Graham

impressive gains from just a plug change...:)

Very impressive to say its standard makes me wonder why ive bought fuel pump, 3 port,sti tmic,up pipe, sports cat and 2nd decat pipe. ready for remap if you can get those gains with it just been remapped with standard parts. I`m just guessing with all the parts ive bought there it will make it safer and easier to get those gains ??

It will make your turbo spool quicker, produce your target boost easier at lower load, make sure your engine doesn't die from fuel pump wearing out and generally make it easier to get the power as well as more efficient.

That said it's all about the mapper and the map, that's ultimately what gets you there! (blowing my own trumpet :lol1:;)).

Hope this helps!

Graham

Well done Graham, I took the car for a spin yesterday and the map has made a huge difference to the car, ok it's only 268bhp and in scooby terms is very mild but it goes to show what can be acheived with a good mapper. Next step will be to change the turbo for a td05 (front entry conversion) and we will see what if anything that acheives with no other mods?
If anyone has any bits and bobs they would like to donate for testing let us know.

interesting thread, looking forward to some results on 'common/popular' mods, many people have preconceived ideas and expectations on HP figures gained from specific mods, I’ve seen quite a few disappointed owners of highly modified cars who’s figures don’t appear equate to the mods on the car.

I’m building a ‘99my UK turbo primarily as a track car but hopefully (if I can finance it ) I aim to race it in the Nipponchallenge / Northern Saloon and GT series, my aim is a modest 300bhp in line with the Nippon GT300 class, currently I have the following VF 28, yellow (440) injectors, 3 port boost solenoid, Walbro in tank to swirl pot feeding a bosch motorsport external fuel pump, equal length headers, aftermarket up pipe, full decat 3inch S/S system (all exhaust is of unknown make /origin), at present intercooler is a blob eye WRX top mount although I’m considering a FMIC, air filter is a HKS panel in OE airbox, (although this will change to a CAI if I go with a FMIC), oil catch tank to atmos this as been a budget build using a common JDM base (turbo , injectors) then adding bits begged, blagged and borrow’d, I currently need a uprated FPR as I don’t think the OE item will be up to the job of controlling the Bosch pump and I also need to make a decision on a FMIC and induction set-up.

Then hopefully I’ll be ready to have it mapped, it’s already been mapped (ECUtek map no2??) so the licence is paid but it will obviously require a complete re-work to get it spot on.

Any comments on the above and my expectations most welcome, I can find no mention of ECUtek mapping on your website is this something you can do or do you prefer /recomend an alternative.

My initial comment would be the turbo, we recently built and mapped a 2000 P1 running a VF28, it peaked at 286bhp with 312 ftlbs, otherwise it was pretty std apart from a big exhaust. The in tank pump is all you need at this level, I would be looking at a TD05/Vf34/35 as the next investment along with a re-map. Graham will be able to comment on the ECUtek qustion............Andy

thanks for the input, I may just look round for a more suitable turbo, although around 290bhp and 310 ish ftlb would not be far off my intended class limits, I'm more interested in torque and drivability than headline power figure.

re the fuel pump it's not really a performance thing, i fitted it along with the swirl pot to avoid fuel starvation on the track, this becomes an issue when the tank level drops below around 1/2, also it allows the correct amount of fuel to be filled rather than over filling this can amount to a weight saving of around 20-25kgs for a typical 15min race.

The VF28 is more than capable for around 300bhp, I think the VF28 on the customer's car was a little tired as I had expected more than 286bhp.

FMIC is preferred really as bhp increase is almost guaranteed with no extra stress although the maf will need to be rescaled correctly to work with an induction kit and you will need to make the induction kit well insulated from under bonnet heat and ideally a decent cold air feed.

Ecutek I don't use, just something I haven't got in my artillery of tuning tools, mainly on the basis that these ECUs have a habit of giving up during the reflashing process so it's a bit of a risk. Personally I'd recommend upgrading to a different ECU to run mafless to save complication and much finer live control. Simtek is the preferred weapon of choice.

Hope this helps,

Graham

A friend of mine has just replaced his VF28 for my VF35.....It was dyno'd at 332bhp with the 28.....He is even happier now after Simon just remapped it lastweek with the 35 where it made 343bhp...

Great thread tho fellas..:thumb:

thanks for the input, I think the VF28 will be staying if it can make 300bhp then i'm happy, there are plenty of other things i need to spend money on, re the ECUtek at the moment retaining ECUtek appears to be my best option as I already have the license, reflashing it to suit my spec is the most economic option at present, but a Simtek or similar is on the cards in due course. I’ll be watching with interest your progress with the test ‘mule’ project. I’m assuming here that the results will be broadly applicable to my 99my car and indeed earlier cars, I certainly hope so as it would be nice to get a true idea of what various specs / mods are worth in power and torque, lots of figures get posted (like the quote below) which mean very little on there own.
I think to be relevant the full spec needs to be known, and indeed the nature of dyno testing makes comparative results very suspect unless they were achieved on the same dyno , on the same day with the same air temps , humidity etc simply changing the position of the air temp sensor on a static dyno can dramatically change final figure (this trick is very useful in showing an apparent hp gain when fitting air filters / exhaust etc which actually make little or no difference).

Re your test mule, as your following popular tuning paths / mods what will you do prior to fitting a replacement ECU if you don’t use ECUtek (ECUtek been quite popular at least on late classics) most owners don’t opt for a aftermarket ECU in the initial tuning structure, what options are available that retain the OE ECU?

The car made 332 with the VF28, 440cc injectors, Fmic, IK mapped with a Link ECU.

I’m sure it did and I’m not disputing the figures ,no all I’m saying is there are so many variables involved in dyno testing it’s very difficult (impossible maybe) for the results to be definitive, the same car may only show 300bhp on a different dyno on a different day, I realize people need to quantify their cars but really if a car is setup correctly the BHP figure is secondary, I know it nice to have a great big BHP to tell people about but dyno print outs don’t make any difference to performance on the road or track it is what it is.

What I do find interesting about your original post is that the VF35 only made 11hp more than the VF28, I’d have expected a bigger difference between the two, but hopefully this thread in the fullness of time will reveal with reasonable accuracy the actual differences between not just turbos but all manner of aftermarket tuning goodies.

the vf35 is so similiar to the 28 that why would you expect a big difference in bhp?? from memory its only really the exhauist housing of the 35 (smaller to aid spool/responce) that differs vastly from the 28 - other than the 28 having a superior ball bearing cartridge

alyn

Firstly let me say I’m relatively new to Imprezas so I’m not familiar with the all the ins and out of various turbos, but I’ve been advised on more than one occasion that the VF35 is far superior to the VF28 and a worthwhile upgrade, so my expectation of power output was based on this advice, judging by your comments this would not seem to be the case, and exactly the reason we need this type of ‘test’ thread and attempt to get a true idea of what the changes we make actually do in terms of power and torque.

Ok just a mini update on this, the test car has a VF35 on it now, I'm just wiring up an adapter for a different ECU to try also, so will be tuning the original ECU and VF35 soon along with the aftermarket ecu.

The 410cc injectors that are in it are being changed for some of my modified 750cc injectors for testing. Flow tests and spray patterns have been excellent so far, just need to check they work as they should as I've had big problems with the modified 440cc yellows in older cars. I'm looking into a solution for these!

Graham

Ok here's the update I've been waiting for and hopefully you too!

So lets recap, we got to the stage that we were at last time, (268bhp), with everything standard on the car except for some Aga 7 rated plugs and running on V power only since the bottom end was forged.

Our ideal target for this car was to get to an initial stage of 300bhp and similar torque. So the next stage plan was launched.

A VF35 went on for starters, but due to the original injectors reaching 98% on the previous map I had to do something with them in order to be able to keep up with the VF turbo.

Some of my modified blue injectors were called for. I modded a set and sent them away for flow matching and cleaning, returned with excellent results, 750cc and all within 1.8% of each other.

So off to the dyno I went, with a Cosworth filter stashed in the boot ready for an air filter test also.

I must thank and commend the guys from Silverstone Autosport for letting me commandeer their dyno for 3 hours. Absolutely cracking bunch of guys and only down the road from me!

At Silverstone:

http://www.enginemapper.co.uk/testmule/silverstone.jpg

Anyway on with the power graphs....so after some playing around I managed, (albeit right on the limit), to extract:

http://www.enginemapper.co.uk/testmule/DDdyno3.jpg

As I wrote on the dyno chart, this is just VF35, my own 750cc injectors and PFR7Bs. You'll remember that I had some cheap Ebay 7 grade spark plugs in. I mapped the car first on these and managed 294.6bhp, thinking that they may be the limitation to achieving 300bhp swapped them out for NGK PFR7Bs instead and am surprised to add that they made no noticeable difference to the cheap nasty plugs!!

It was evident that having all the cats still in place was not only making the turbo a little laggy, but also limiting the amount of boost available to play with, only being able to boost 1.3bar mid-range tailing off to 1.2bar top end. With decats in place I expect to be able to hit 1.4-1.5bar mid-range easily to hit the same torque as bhp. Torque currently is slightly less than the TD04 at 277lb/ft.

So next up was the air filter test. Being that I was at Silverstone, (an appointed Subaru service centre), they had a Pro-R air filter on the shelf to try and I'd brought along a Cosworth filter that Andy has stashed in the car. Both brand new of course.

The Cosworth was first, no increase in peak power, but a noticeable improvement in breathing at the top end:

http://www.enginemapper.co.uk/testmule/DDdyno4.jpg

The Pro-R air filter went on immediately after with almost identical results. In-fact it was about 2bhp more than the Cosworth at it's biggest improvement. So to be fair, the Pro-R filter lives up to bigger named filters and of course is warranty endorsed by Subaru!

http://www.enginemapper.co.uk/testmule/DDdyno5.jpg

I'm pleased with the results from today, though a little disappointed we didn't see over 300bhp. The Cossie filter has stayed in the air-box along with the NGK 7s, VF35 and EngineMapper 750cc injectors still on the car ready for some more miles. Next up will most likely be exhaust related, cat back, decats, headers hopefully to see us past 320-340bhp.

After that, well I saw a shiny 20G sat in Andy's workshop last time I was there with the Test Mule's name on it! ;)

Feedback and debate welcomed!

Graham

Nice work there! Do you mod the standard WRX injectors then?

nice work bud

That's quite eye opening - seeing that increase on an air filter only change.

You were stressing the importance of air supply when my car was on Silverstone's rollers Tuesday - mine being compromised with induction intake under bonnet so taking in warmed air

btw - you managed to map the bug OK with those wobbly wheels? :lol1: :)

Nice work Graham, and a big thanks to Silverstone for the Dyno time. It will be interesting to see what happens next when we start with the exhaust mods, I have new headers and up-pipes here to test as well as a new HP fuel pump to keep things safe. It will be interesting to see what the airfilter does once we start to exploit the turbo's true potential, i.e. when we get over 300bhp will the standard v's Cosworth filter be more evident? Watch this space..............

Great thread guys - subscribed :)

Been a very interesting read so far guys, well done for your efforts!!

Can't wait for the next update :norty:

Stalking me again I see Jay :lol1:

awesome thread im subscribed

:lol1:

Your comment wasn't on there when I opened the page at lunchtime.... I'm just a slow reader.

Don't forget my money on Sunday ;)

I do indeed Myles!

Bonesetter: indeed, every little helps!

Thanks for all the comments, I'm going to try and get a back to back test vs the OEM ecu and a Simtek sorted in the next week as the OEM ecu has always had a little hesitance off boost which seems to be hard to iron out. So will see if running mafless sorts this and whether we can get any more power out of the beast.

I'm determined to see past 300bhp on this setup also, so will add a meth mix in next which I predict will get us to 310-320bhp.

Graham

Graham - maybe an idea that you would like to myth bust is about petrol going stale and loosing performance if left in the tank for too long?


Maybe a dyno run with a fresh tank of juice then keeping a a few litres of the same batch aside for a month, then dropping it into the tank to test if the car reacts badly to it?