2.5 Conversion problems - need help please!
 

Hi all, I had the 2.5l conversion done on my 1995 STI RA, with uprated clutch, but std ecu ( I didnt take it above 0.2bar boost and it didn't det) first of all, every so often it wouldnt start, and did when IT wanted to, but now, problem is that the car would be a pig to start when warm, but idled fine, and then it didnt start when it was cold unless with a bump (only small, but started every time and never with the key) I got CEL faults on the boost solenoid and crank sensor so I replaced them, still the same! now it wont even idle either, and runs like a bag of ****, any thoughts?

maybe 2.5 starter, new batt, MAF, crank and cam sensors and coil packs, i'm not bothered if I cant put my foot down before I get the ECU, just want it to Idle, as it used to on the std ECU, but wont now.

Has the fuel pressure or injectors been altered / changed?

if not then it will be lean by about 8-10%

Simon

No it's std at the moment, i know it'll be lean, but it used to at least idle.

I would say you have either burnt the plugs, or burnt holes in the pistons, or melted an exhaust valve..

I'd check the plugs and do a compression test and not be so silly to run it on the original ecu like that..

For about £100 or less you can get an adjustable fuel reg and at least get the fuel close, or but some 440 injectors which would also on the standard ecu get it close

Plugs I would have assumed to be ok, HKS Iridium, but if running lean, it could be a possibility, I also changed my turbo because it was passing oil, may this have been a contributing factor?

Will do a CT and find out the results and post them, should they be around 170?, lets just hope it's not a valve or piston! would a greddy e-manage be a better solution to the injectors? or should i just wait for a power-fc/gems when the money isnt so tight?

you are probably going to want / need bigger injectors later.. if youy get an e-manage you will have to bin it / sell it when you get the ecu so may as well go for something that will benefit you later.

Simon, is it worth a further think a bit more about how a MAF based ECU works? Not being rude, bear with me...

I don't understand how increasing the fuel flow with injectors or regulator is necessary, having only run 0.2 bar boost. Warm start was probably an issue because of the cranking pulse widths used being a little short for the bigger engine (they are likely generous enough at cold start), but once it reads the MAF sensor it should settle down and still manage near enough the correct AFRs for the airflow it is pulling, assuming the inlet tract and MAF sensor mounting tube are as before. 440s will overfuel and sound like a bad idea to me.

I wonder if there could be an earth fault with so many sensors/actuators being suspect? Coolant temperature sensor worth a thought as well.

My JECS 2.0 map was very suitable for my 2.5 as it happens, with an increase in idle speed it just worked and did about 400 BHP out the box, that is why I know this myth about running 2.5 engines on 2.0 ECUs is just that... Manifold pressure for manifold pressure it was more retarded, just as you want for a bigger bore motor. Shouldn't be done blind, but I would happily run a 2.5 on a 2.0 map armed with only a knocklink and AFR meter at low boost.

E-manage with the ignition harness would do fine, it is a hugely underestimated bit of kit, but a PowerFC is only a little more once you have the extras and is plug in.

John,

Not rude at all.. you can tell I started my engine mapping on a non-MAF baised ecu :rolleyes:

Will the MAF really compensate that much? I suppose it will be consuming more air and the MAF will read the extra and and add more fuel so of course it makes sense.

Okay lets hope it is something very simple then

Simon

Yes it will compensate massively, it just reads say 23% extra airflow, so approx 23% higher on the load axis (as MAF/RPM=load is by far the major part of the load calculation) and will use the hopefully accurate airflow reading to fuel the desired AFR in that map cell. Since the whole cruise/idle area has values very near to stoich for desired AFR, it should keep running that no problem. It will come out of stoich a bit sooner in terms of manifold pressure than the 2.0 because the airflow is higher, but there is a huge leeway on acceptable AFR on light-moderate loads, can run fine at 15:1 as it would at 11:1, lot more than 23% in there. In addition to that, closed loop on JECS (at least AE800-802) has a +-25% adjustment authority. I think it would run a 3.0 off boost actually ;)

MAF based ECUs do need a different approach, they are so incredibly neat if the MAF sensor reading is smooth and accurate/reliable.

No idea what is wrong with this car without seeing it, and then I probably wouldn't work it out as it sounds like a bas**** ;) but if it idled and drove OK before, then I doubt it is ECU related.

not at 0.2bar but I assume there would not be enough room at the top of the tables with 23% extra on the MAF reading?

Could well be MAF fault on this car it has obviously been working a little harder..

Brake cleaner might be worth the effort.

Agree lots of room on cruise as mine will run 15.5 on the Link, or 16 or leaner if I were to let it on the Apexi.. which I didn't, after the discussion we had John the other day 14.7 seems a better bet, especially if it is not monitored, ie. someone elses car.

Simon

like I said, never went above 0.2, so in theory it should have been fine, given that it has compensated for the extra airflow into the engine by more fuel on idle, but that isn't the main problem, and yes john, it's a real pig and it's pi***n me right off.

Simon, if you hit the last load zone on the table, it just continues with that timing and AFR. Usually that is quite sensible, but for big boost on big turbos sometimes the timing is advanced, esp on late JDM GC8s.

That AFR will continue to be correct as long as the MAF sensor stays within the limits of the airflow table, after that it will clip to the maximum airflow, also it could be clipped if the "load" limit is exceeded, so you could have a high MAF voltage at medium RPM from lots of torque. In these circumstances it will go leaner as airflow increases, but the limits are relatively high.

John, do you know what the limits are for the MAF, so say I upgraded the ECU and kept the MAF, what max boost limit would I be looking at on the 2.5 before i had to upgrade?(if poss and to what) or do away the MAF and get a GEMS etc, or is it just the limits of the map?

Adam

I would suggest an Apexi Power FC. That has five programmable MAF curves with 32 points each. Z32 MAF is a common option, as it putting original MAFs in larger tubes. Keep a MAF sensor if you possibly can, they are more accurate than speed density and vastly superior for transients, better environmental compensations as you don't need to worry about EGT and EGBP in terms of your fuelling. Downsides are flow reversions with certain dump valves/intake combinations, especially with long FMIC pipework and higher boost levels. Only real need though to get rid of a MAF is with a 4" intake on a rotated mount turbo on a short ram with a ported shroud.

Re the limit on the MAF on your model, I am not sure, you'd need to measure the MAF voltage at various power levels with a given intake, or perhaps some people with 92-96 cars will be able to guide you.

I get everything you said, but wtf is a rotated mount turbo on a short ram with a ported shroud? u lost me there

Rotated mount is where the turbo inlet faces the inner wing at a 45 degree angle to the usual position. Short ram is a short intake without lots of bends. Ported shroud is a compressor cover which has slots cut in it to allow air to circulate.

The result of all this = big mutha turbo that produces a lot of turbulence = difficult to use a MAF.

Thanks John.....now if anyone knows why it wont start.......

Very techy talk going on here, but i think you're going in too deep for Scoob+Bike=Fun's problem.

If it ran of a fashion before albeit showing difficulty starting (and probably *fairly* poor actual running), would it not be a good idea to go back to basics and check all the small hoses linking such things as the 2(3?) port solenoid, map sensor etc etc.
Basically all the hoses which ultimately attach to the inlet manifold..

my my97 sti ecu ran mine very well for 8000km's.

Are you shaw the cam timing is setup right as they can do some strange thing if not and run like a big bag of poo.
I know a guy in the usa how did the same project and had these types of problems untill he took the cam belt off and put it back on.

Just an idear.

will be checking all the basics, and now obviously the cam timing, but as the problem got worse overtime, i'm thinking it could be a sensor fault, so new MAF will be ordered unless anyone near can help me by letting me try theirs? it's the black one with the grey plug, MY95. Thanks to all, will keep you all posted

Near to where?

oops sorry, manchester. I've just taken 2 plugs out, and had to stop because of rain, but the 2 plugs out are NGK IRIWAY 8's i'm sure it should have 7's, is this right? 1 is sooted, 1 isnt that bad. lets see what the compression test brings...fingers crossed! if quite a lot of oil is passing through the turbo seals, this would cause it not to start wouldnt it? there is excessive residue in the pipe connecting the maf to the turbo :(

I'm in the South East.. doh..

The early MAF are fairly strong so hopefully it is something else.

oil residue and plugs different doesn't sound good.

Simon

:cry: :mad: well, after doing a C/T, results are as follows, not sure if my cyl numbers are right, 1 is front left, 2 back left 3 front right and 4 rear right as you look under the bonnet. Not a happy bunny!

1. 108
2. 88
3. 88
4. 62

now, i'm no genius, but this isnt right, not checked the cam timing yet, so that could be a possibility???? could it? I hope so! so everyone...am I looking at another engine rebuild (just had this one done 1700miles ago)

please help!

p.s. didnt disconnect the injectors so could fuel in the C/tester do this, also results on closed throttle

Adam

oh feck!

Sorry dude!

Fuel in the tester shouldnt be a problem,but it could be worth doing the test with the throttle fully open,as you will get a much better reading.It seems strange that there is such a variance between the readings(this could be down to what i said about not opening the throttle).I personally think that youve got to be VERY unlucky to have reduced compression on 3 cyls through holed pistons/burnt valves running at the boost levels you are-its hard to diagnose without looking at it but at worst i would say the head gaskets are shot.
Its definately worth checking ALL the hoses for splits/attached-even check that the blanking nipple is over the spigot on the middle of the inlet manifold(if it has one or hasnt allready got a boost gauge hose attached to it)-even a very slight leak of unmetered air getting in(especially after the throttle body-ie manifold)can cause these probs as youve described.

Keep us posted,id be interested to hear the outcome:)

Gary.

Did you do the test from 1 through to 4 in your above post.

did you have a battery charger attached whilst testing each cylinder.
Did you have all the plugs out whilst testing each cylinder.

Due to the ever decreasing compression values you got, I'd say it was the battery dying off and the enginbe turning slower each time you did a test.

As above, also jam the throttle fully open to make it as easy as possible for the engine to turn over...

Andy

Thanks Guys, Been really busy, but will re-do the test with a batt charger attached, just in case. Could it be that the head bolts will need re-torquing up? just a thought, I had heard that they need doing after running in, is this a myth? and if so, can it be done whilst the engine is in the car? I'll re-do the tests, but tried number 4 with an open throttle, and the same result,so thats not good.:mad:

if no joy, then it's coming out on thurs-fri and getting stripped down, made a lovely space for it in the dining room:norty:

Adam

Ok, tested with fully charged battery and fully open throttle and crank sensor fully d/c and in reverse order, turned engine over 20 times on each cyl

1. 112
2. 88
3. 95
4. 70

I have checked all hoses, and they seem fine, might just start stripping all the ancileries down and start from scratch, but it doesnt sound good with the results and will probobly just take the engine out unless anyone has any ideas for me to ponder over?

Adam.

cam timing out?