Lean combustion?

[Chas54]

I thought I'd check out my throttle plate passing theory by backing off the little upside down stop screw. So I removed the big rubber inlet duct and started the engine. As slackened off the lock nut on the screw I locked at the elan scan screen balanced on the windscreen and realised there was no TPS voltage showing. I was expecting to see my 0.4 volts and I would be able to see if I undid the stop screw. The engine then stopped.
Pulled the TPS terminal rubber boot off and saw the black cable was adrift.
Resoldered it back on but prior to that it must have been hanging by a thread.

Then with my 0.4 volts showing I undid the screw which now read 0.38 closed.
I'll have to test today and re-adjust the IAC bypass to calibrate for the less air passing the throttle.
Then readjust the TPS to read 0.4 volts at this new closed position .
Wether this miniscule .0.02 volts extra closing will mean anything or not, I'm not sure.

Again thanks for all advice & comments.


Chas54

[Tuga2112]

dont get your hopes up on that.

my TPS was off by more than that for God-knows-how-long and i never had any noticeable problems.

[Chas54]

Went for my standard 10 odd miles run to test today.
Warmed on driveway one stop at road works on the way out and one on the way back, stopped in between in a lay by to look at elanscan . Running well and every time i snatched a look , the O2 voltage looked promising.
The !AC by pass is now quite a bit further open which probably account for my lower IAC step counts.

No ECU light until just about half a mile before arriving home, although i felt no hesitation in the engine. Around 2173 secs engine run time. On the drive way at home i could see the O2 voltage very low indicating rich again. A blip of the throttle and it was working again.
At the time it occurred the car speed was a steadyish 50mph but the O2 voltage was dropping and the manifold vacuum was improving. The improving vacuum should cause a drop in fuel pressure , but the O2 voltage shows the mixture was getting richer ie more fuel . Also the injector bandwidth was decreasing again i would have thought this would cause it to get leaner . I dont understand this . But the O2 integrator went up to 210 .. I read somewhere this has some bearing on the Block Learn Memory but its function is really unknown to me .

It feels a lot better but it still isn't perfect and I'm not sure what has made it feel better. I still have the external leads directly to the O2 sensor .

One thing i am currently looking at is extra shrouding for the injector signal cable from the CAS. It runs very close to the HV ignition leads and is basically a perfect victim for induced voltage from the flux pattern . I am going to try some copper cable shrouding with an earth at one end.

And yes, i'm clutching at all sorts of straws.

All comments greatly appreciated

Chas54

after TPS resoldered road run.zip

[Tuga2112]

since your clutching at straws. and since im happy to offer random suggestions with varying levels of usefulness.

i am wondering if you had missed sparks. as unburned fuel would make its way back to the intake via the EGR causing the next cycle to record more CO emissions despite the fact the injectors bandwidh decreasing.

i have onse witnessed a car running on 3 cylinders and i gotta be honest it was less obvious than i expected.. and if it only happens sometimes instead of constantly. it may be a sign.

so. the suggestion id make is to double check the plugs to the coil pack control module. see if everything there is well attached and theres no corrosion. the fact that right under the coolant reservoir has its risks.

[Chas54]

Tuga

I see what your getting at and as i had the ignition plate off the other day with the coils, i might give it another check. Its a bit awkward getting hold of the leads down there.
Its not the EGR as i've blanked it off at the flange. But i may try my strobe on each HT lead as it would give me an indication if im losing a spark. Recentley I do sometimes notice a miss in the firing when the engines idling .

thanks

Chas54

[Chas54]

Its a bit damp today so ive started with a look on the internet using the AI function . Whilst taking almost everything the Artificial Imbecile comes up with as dross , I started thinking about this description of the GM block Learn Memory'. :-

Block Learn Memory This description for closed loop models only
Open loop models don’t actively use BLM //integrator values during operation but use the manually adjusted CO pot,

For handling long term fuel trim corrections

ECU use a centreline value of 128 . At 128 the ECU is running on its pre programmed fuel maps without needing fuel corrections.

Fuel Adjustment
If value rises above 128 ECU detects lean combustion and adds extra fuel.

If value drops bellow 128 ECU detects a rich combustion and subtracts fuel.

Matrix cells

BLM acts as a block of cells or a grid divided by RPM & MAP.

Driving at differing loads the ECU switches cells and adapts to component wear, mild vacuum leaks and fuel quality changes.

Integrator v BLM

Integrator represents short term fuel trim ( adjusting up to 20 times per sec)
If the integrator stays away from 128 for too long , those changes are migrated to the long term BLM

BLM reset

If you’ve fixed some faulty sensor or vac leak the ECU will try to compensate based on the stored behaviour tables stored .

Disconnect battery .
Clears fault codes and clears fuel correction memory
Drive car at varying speeds and conditions to allow ECU to re learn the ideal fuel trim conditions and store to BLM



So Does this mean if I drive about a lot more, the memory will gain more experience and correct itself ?

Chas54

[Tuga2112]

i cannot answer that question with any factual data but i can tell you that i disconnect the battery fairly regularly and either mt car is "on spot" or it adapts very quickly because i dont notice any difference between its responsiveness before and after disconnecting the battery

i know for a fact its not the flip side (taking a very long time to pick up the new "defaults" because i have had disconnected the battery and reconnected multiple times over the years with sometimes hours between events and sometimes months (and just today, more than a year)

that "memory" of the ECU in my particular experience seems to be a mute point. and i never had a conversation with anyone who had experienced something that was strong enough evidence to suggest that "learning" made a difference.

[Chas54]

Checking that each plug is sparking with the strobe gave me a sense of deja vue. Ive been down this route before .
With having the wasted spark system , Gunsons say my strobe is incompatible . I see a rapid spark on two cylinders but the other two are slower and occasionally miss altogether. . But not from the same coils. i can check timing OK but not the existence of a fault. .

Ive tried swapping plugs about ie good plug in missing spark cylinder without effect, but swapping the leads shows it is them at fault. No 1&3 . I have one old lead ive put on No.1 and that cylinder is now showing sparking on the strobe but need one more for No.3 which is intermittent with my only other old lead . It does show sparking well with the other good old lead now on No.1.

It may not be my only problem but it would possibly account for my rich O2 sensor.

The set of leads i had on with the two were relatively new and not budget items.

Chas54

[Tuga2112]

i suggest removing the coil packs and closely inspectinvg them for cracks.

i dont know at the present. but a few years ago they were easily and affordably avaialble from rockauto

if this all comes down to the control unit behind them, i have 2 in the shelf (1 spare) new old stock one from brit-car-nut that was mistakely sent instead of a pod lift module, that i can sell. but from observation, just like the ECUs those dont seem to ever fail. and confirming that is the actual problem would be rather difficult (unless you forgot to mention the RPM gauge is reading wrong).

[Chas54]

Tuga

I had the coul packs out last week to inspect the ignition plate and earthing connections. I took the time to have a good look for any signs of arcing , and they are good.
Swapping the HT leads about showed me two leads were causing a missing spark intermittently.
It's a bit of a stretch as the leads are all tailored in length to fit. I had one lead slip out the coil a little whilst running and could here it arcing straight way
.. really noisy . The new leads are on the way. but I will be proving them on arrival.

The rev counter looks pretty accurate to me when comparing the elanscan RPM figure.

I am hoping your suggestion about missing a spark intermittently may be causing my low O2 sensor voltage .

Thanks for steering anyway
Chas54

[Tuga2112]

well. your grasping at straws.. im trying to find any straws worth throwing your way considering after all these messages and investigation from the other guys got nowhere. im perfectly happy to offer the most exoteric options.

did you buy the leads from mr retro leads by any chance ?

i got mine from him. and have been very pleased with them so far.

[Chas54]

I have ordered the ignition leads from Mr retro.
Just waiting to get them before diving in again .

Chas54

[Chas54]

New HT leads finally arrived today.

Fitted but still have an intermittent loss of sparks but only on no.1 and no.3.

Idling in the garage , runs nice and smooth then every now and again it looses the spark to 1 & 3. I can see this when checking with a timing strobe and just pointing it at the engine. Checking each lead in turn and waiting for the dip in revs. They are of course on different coils although physically close on the coil mountings, they are the middle two closest together.

I was a bit disappointed, but as it seems to be worse now i'm convinced its something i have just disturbed that has made it worse.

I have just fitted a copper mesh shrouding cable sleeve around the lead from the CAS to the coil packs , but haven’t connected it to any earth yet and insulated the original sheathing. I did have the CAS unplugged so will have to look at the plug and socket . I disturbed the bunch of cables to/from the coil pack plugs.

I noted an engine run point time when this event occurred at around 546 seconds . 1&3 lost their sparks and the engine speed dropped . The manifold vacuum deteriorated at the same time. That could be cause or effect , not sure. I think i can prove which by artificially pulling a steady MAP vacuum with a test pump i have . . Leave it set at a steady 0.36bar and see if the lost sparking still occcurs.
At around 818s i switched to service mod for an IAC bypass adjustment.

Also I wonder if anyone tried this before but with the "Waste spark" ignition system, I can't see why i can't swap no1 coil lead position for no.4 and then no. 2 for no.3 ?? . If i actually have a coil problem it should then cause 2 and 4 to loose their sparks intermittently.

When considering the waste spark system, I can see the frantic flashing on my strobe for both sparks at no.2 & no.4 but on 1 & 3 i only see half the flash rate , and then the occasional complete loss.

Any thoughts and comments welcomed

Chas 54

new ht leads on 1 and 3 idling.zip

[dapinky]

Whilst I have no new insight, the idea of changing the plug leads (on the same coil pack) will cause no problems (assuming that the length/run route allows it).... also mixing in one or two of your old leads may or may not help to confirm the issue..... this is really stumping most of us, there really can't be much else to look at.

I know that you are not really local, but if you get to the stage that you can't go any further, then you are welcome to drive here and we can take as long as it takes to swap as many parts from your car to mine (or probably Angies, actually ) to see if we can eliminate the fault on yours and transfer it to mine in as logical an order as possible.

[Chas54]

That's a really kind offer Dave and I'll definitely think about it .
I'm not out ideas yet, I'm going to try swapping no 1 position on the coil and use it on no.4 plug. And the same with no 2 & no 3. That should theoretically transfer the missing spark and aid a better diagnosis.

I'm also also doing my forced value on the MAP to see if that stops the misfire.. it would prove which comes first the lost spark or the dip in vacuum.

I've done a compression test to satisfy myself it's not a sticky valve.

I'm bracing myself for taking the plenum off again. The worst jobs cleaning the 2ndary throttle body flanges.

Thanks again for all the advice and comments.

Charlie

[KenS]

Probably would be better as a discussion in a pub, but if you are open to more thoughts here goes - seems odd that 2 coils or 2 ignition leads could have faults. Since its cylinders 1 and 3, could it be a shorting/intermitant injector that upsets the supply voltage or raise the ECU ground voltage sufficiently to impact the ignition unit? The latest plot does show that something has got worse or a new problem. A failing injector could be linked to the original problem with lean mixture and now it has got worse?
If the problem is ignition based, I guess swapping over the two ignition coils on the backplate may lead somewhere. If the problem moves to 2 and 4, it implies the 2 coils at fault, if it stays on 1 and 3 could be spark plug gaps opened up on 1 and 3 or something bad in the coil baseplate or maybe very weak mixture in cylinders 1 and 3 affecting the spark resistance?
If a faulty injector is the root cause, pulling each injector plug in turn may give a further clue?

Just for visibility - top plot is your latest plot with the poor ignition pulses and revs being modulated, lower plot from 22-9-2026 without that issue (but still lean running and closed loop issues).

3-10-26.jpg