Glowplug

iTrader: (8)

How are you measuring the AFR on the test run, wide or narrowband AFR meter?

And is the boost gauge reading any different from the origonal 1.12 bar?

harvey

iTrader: (48)

If you want to read about mapping or other tuning related matters you could look for books by Authors such as Dave Walker, A Graham Bell or Julian Edgar. I have not read any of these myself so I have no idea as to the content or worth to a Subaru owner and they may be general without direct application to Subarus. Dave Walker used to write for one of the tuning magazines many years ago when I was tuning British bike engines.

I think that successful mappers doing that for a living, over many years will tell you there is rarely a day or a week goes by without learning something new or gaining a greater understanding. Technology has moved forward at a great pace as far as cars are concerned over the last ten years.

Don't embark on mapping your own car lightly. It only takes a small mistake or lack of understanding to cost an engine.

john5f

iTrader: (3)

Cheers Harvey

First things first. I would need to UNDERSTAND EXACTLY what is happening before even considering doing anything myself.

I have been building and repairing computer systems, writing software and supporting operating systems for over 30 years after completing basic training, so have experienced technological advancements first hand. The basics have remained pretty much the same but systems have become far more sophisticated. I can imagine that car technology has moved on in a similar way.

Many thanks for your advice and pointing me to the reference material.

I'll be in touch

John

john5f

iTrader: (3)

Well, I have just re-connected the J1 resistor and been for a drive to try it.

The performance is noticably less, as I would expect, but the problem still seems to be there. At just over 5000 revs it feels pretty much the same as it did before, i.e. acceleration goes flat. I obviously backed off as soon as I felt that happen.

So, with that eliminated, back to the drawing board.

Harvey, you have email.

Glowplug

iTrader: (8)

Wouldn't put too much on that test mate, as you said earlier, your ECU is from a wagon should be either a:
1C, 3B, V9, V5 , V9, 4B, which=
10.55PSI (0.7BAR)
DROPPING TO 6.84PSI (0.5BAR)
***STARTS TO DROP @ 5000RPM***
FUEL CUT 13.79PSI (0.95BAR)
And will be mapped for the smaller TD04.

The AFR map will be totally different as your AFR monitor should of shown you

john5f

iTrader: (3)

Thanks for your comments

CPU is a 3B and I hear what you're saying but I accelerated gradually and wasn't boosting above about 0.5 bar. I still felt it hold back like it did before at just over 5,000 rpm. I definitely didn't hit fuel cut.
I realise that it's mapped for the TD04 so didn't expect it to run spot on but it was the best I could do with what I have access to.

john5f

iTrader: (3)

I'm still intrigued about why it seems OK when not up to temperature. I realise that the mixture is enrichened but, looking at the graph, by that much?

dabow

iTrader: (19)

just a quick one mate, what plugs do you have and what are the spacing, and have you cleaned the maf? also the coils?
one could be braking down

also worth blowing through all the boost hoses.
could be just a simple thing thats happening.

harvey

iTrader: (48)

I have responded to your communication but you said you had replaced the ECU. Crucially that is not what you have done.
Replace the ECU and check out your AFRs on a wideband Lambda sensor and I suspect you will have no dip in AFR at 5200.
I am happy to send you a second hand fuel pump and FPR but I think you need to verify your ECU is not the cause of the problem because that is the way it looks like to me at the present time.

dabow

iTrader: (19)

i'll to the proper guys in the know
as aboth, seems to point to electronic, and not macanical

john5f

iTrader: (3)

Perhaps I should have posted this before:

Plugs all looked OK but changed them anyway for new NGK Iridium BKR7EIX gapped at .65 and double checked for tightness.
MAF cleaned then alternative MAF installed.
Different coil packs fitted also each one removed, replaced with spare and rotated. No cracks evident in any of them.
Boost solenoid cleaned (using diagnostic mode).
Battery earth connection doubled to different location.
In-tank pipe from fuel pump checked for blockages.
Voltage at fuel pump connector checked at 12.02v with engine off and ignition on, also connector block under rear seat at same voltage.
Checked vacuum/boost hoses for leaks and none evident. Also no noise to indicate split hoses. I haven't had the hoses off and blown through them, but boost reaches the levels expected.

Can't think of anything else off the top of my head.

john5f

iTrader: (3)

Stupid thing double posted

dabow

iTrader: (19)

every thing id have checked then.

as harvey said, could be the ecu.
you'll get to the bottom of it
on a personal note, the last time i used irridion plugs
my boost wasnt right, used ngk platinum 7b's ever since with no issue

may be just a one off set i had

john5f

iTrader: (3)

I didn't have an alternative ECU so disabled the chip by re-connecting the J1 resistor. I understood that this would revert the ECU back to standard. Also, I only drove the car to see if it felt any different. It felt virtually the same as before although the symptom appeared to be at revs a little lower than the 5,200 previously.

It sounds as though I need a completely different ECU AND have the AFRs checked with a wideband sensor rather than just the drivability.

I have just had feedback from the chip supplier who said that misfires read as lean and noticed that there is a very small drop in RPM just before the lean spike (the little twirly parts of the graph occurring before the lean spike). So it might be something else that is causing misfires rather than a fuelling issue.
He also said that, although he has not seen an ignitor unit fail yet, cars aren't getting any younger so that's a suspect. This would tie up with it working from cold better than the fuel pump issue does and would explain the power being lower than expected.

I will have to try to find someone who could lend me an ECU to verify whether that is the cause of the problem.

Harvey, I would still like to fit the Walbro etc. Also can you also manage an ignitor just in case please? I will email my address etc.
Many thanks

harvey

iTrader: (48)

Per my other communication, I think you will be doing yourself a favour by changing the ECU to that from any other JDM, not UK MY93-96 preferably fitted with a TD05. I expect you will find you don't have the issue.
The Walbro is here and I will have to find an OE FPR and I will communicate later today. No igniter at present and until the ECU board is replaced not particularly relevant.

harvey

iTrader: (48)

Pump and FPR with you before lunch Thursday but I still reckon it is ECU.

Fuzz

I'll stick my neck out as well and say that's a very rapid change for a FPR / pump.

Didn't misfire at 5200 did it??
That would show a belter of a lean condition like that?
Admidtedly though it doesn't recover to the rich condition.. Hmm

Map on chip or MAF sensor having a funny five minutes..

john5f

iTrader: (3)

I agree that it is a very rapid change. Whatever it is, as soon as it hits 5,200 it breaks down, hits a blockage or something.

Yes, it seems to have misfired at 5200 as the revs dropped fractionally - check the squiggle on the chart immediately before the spike.
Apparently, according to what I'm told, a misfire causes a lean condition.

MAF changed and identical symptoms.

Reverted the original ECU to standard by disabling the chip and, although that supports a TD04 and I have a TD05, the result was pretty much the same.

Next step is trying a completely different ECU from a jdm saloon, hopefully today, Thursday.

I wish it was just a funny five minutes but it is consistent and happens all the time at 5,200. Everything seems great below that.

Tidgy

stop playing around with it and get it mapped properly. if theres a fault somewhere that can be sorted at the same time, then you don't have to worry about it.

Glowplug

iTrader: (8)

OK, re-read the thread.

Am i correct in thinking (and i hope not) that you are infact 'guessing' the AFR? (after the RR session) If so

You cannot guess it, it needs to be metered/read/monitored with a wideband AFR reader or software.

Yes you can feel a 'mis' or a drop or serge in power but no way can you 'know' what the AFR is.

You need it on the RR or on the road with a mapper sat next to you.

harvey

iTrader: (48)

Update?

john5f

iTrader: (3)

Been really busy with airport runs the last few days so not had time work on Scoob. I'll post updates as soon as I have them.

Thanks

scoobiewrx555

You can pick up a perfectly good used ECU from eBay for not a lot. If it turns out not to be your ECU that's the problem you can always sell it and get more or less most if not all of your money back or even a bit more if you're lucky.

As for mapping at your power levels ESL is by far the most appropriate and cost effective solution as it retains the factory ECU and therefore factory knock control unlike some aftermarket ECU's that although have knock detection and can display detected knock via a check engine light or multiple LEDS in a box or on your dash, they cannot pull timing upon detecting knock or as a final fail safe add fuel before your engine goes pop. Unfortunately by the time you've seen the knock light it's too late and the damage is done. You cannot react quicker than the ECU so better to let the ECU deal with it.

Having your car properly mapped is much cheaper than either buying a 2nd hand engine of unknown history or having your engine rebuilt.

People still don't seem to be getting the message that chip solutions don't work long term or short term, and eventually are a sure way of doing serious damage to your engine and your wallet. They may be cheap initially but that's exactly what they are....'Cheap as Chips!!' Go and search SN to see just how many people are falling foul of 'chipping solutions'.

One size does not fit all and never will!!

harvey

iTrader: (48)

Your ECU number is on the ECU board, something like S19 or some similar marking. Located in passenger footwell. Just pull back carpet and then remove cover to expose ECU board. If you need one we will have a past customer who has moved on to a Simtek. Let me know the number and I will put you in touch with the guy with the appropriate ECU. At a guess £30 posted.

john5f

iTrader: (3)

At last I had time today and tried a totally different ECU on the car and had it on a RR.
We tried before and after runs and, while the charts are not identical, on both runs the afr curve is pretty even up to 5,200 rpm then the lean spike occurs. We have now therefore established that the ECU is NOT the cause of the problem.

A SSM was hooked up and showed no faults whatsoever so the problem is caused by something that the ECU is not monitoring.

It is now pretty certain that the engine is misfiring and that is what is causing the afr to lean read.
The problem appears to be narrowed down to ignition components and/or wiring. I need to check the wiring for bad connectors, pinched wires and earthing and re-checking coil packs & plugs.

I have also now fitted a Walbro fuel pump, different fuel regulator and ignitor so I have known components in the system.

I will post further update of results.

Thanks again for all the knowledgeable advice.

Glowplug

iTrader: (8)

Well done

Did you fit the pump, reg and ignitor before the RR run??

joz8968

iTrader: (13)

I bet it's a rogue coilpack/s breaking down...

john5f

iTrader: (3)

Harvey,

Yes, thanks, I fitted them before the run so we had a known starting point.

The tuner also gave me good information about mapping and how the ECU works and the way upgrade chips and ESL daughterboards integrate with it.


Joz

It would be good if it were that simple but it is still a possibility I guess, although the first thing I tried was to get a different set of coilpacks. I also tried rotating 5 coilspacks to see if it made any difference but I will be trying that again to make doubly sure I didn't miss one. One of the originals broke up and 2 had hairline cracks around the edges. The one with no signs of faults I used in the rotation.

Thanks for your input

harvey

iTrader: (48)

If the misfire is ignition related the AFRs will be rich.
If the misfire is fuel related the AFRs will be weak.

john5f

iTrader: (3)

The tuner said that the AFRs overall are on the rich side and that the misfire is giving a lean read causing the spike on the graph.
Still checking electrics for pinched wires, poor connections etc.

Thanks