Spark Advance going mad?

[G UK]

After completing the following work I am having a problem getting Blackie to run smoothly and was hoping to pick peoples brains:

New Turbo
Head Gasket
Valves re-seated
Belts, Pulleys & Water Pump

The problem is manifesting itself as an inability to use much above 0.5 bar boost or full throttle without getting severe coughing and misfiring and also after the car has been driven the idle sounds like the engine is constantly cutting out and when in reality the revs are reasonably stable.

In order to investigate the problem I have already reset the base timing and it is completely stable, checked leads and coils which appear to be OK and had the CAS cover off and given the CAS a quick check which seemed OK with the hall effect sensor seeming to be the newer version (I did give it a quick spray with WD just in case however). I have also consulted with both Neil and Paul with Neil suspecting the coil packs which Paul however tells me are pretty much new.

Now on to the attached scan, after checking all of the above and having the car running perfectly in the garage with a perfect idle I went out on a little run with the intent of inducing the problem. As such you will find the boosts between 400 and 420 seconds as the best examples of the under boost problem and as expected the idle at the end of the run (700sec +) is a good example of the idle problem.

The only thing I can really see wrong with the scan is that the spark advance is as expected at the start, variable under normal driving and then almost flatlines at around 13 - 14 deg right when I'm having the problem. On top of this when the idle is rough at the end of the run the spark advance is all over the place and bears little resemblance to the nice idle at the start.

If somebody can have a look and see if I am reading the data completely wrong and give me some ideas of where the problem may be it would be appreciated.

[saska]

Now on to the attached scan, after checking all of the above and having the car running perfectly in the garage with a perfect idle I went out on a little run with the intent of inducing the problem. As such you will find the boosts between 400 and 420 seconds as the best examples of the under boost problem and as expected the idle at the end of the run (700sec +) is a good example of the idle problem.

The only thing I can really see wrong with the scan is that the spark advance is as expected at the start, variable under normal driving and then almost flatlines at around 13 - 14 deg right when I'm having the problem. On top of this when the idle is rough at the end of the run the spark advance is all over the place and bears little resemblance to the nice idle at the start.

Graham, I don't see the spark advance flatlining as you do. What I do see is:
1. Prom ID: 399 (Dunno this one)
2. Faults: 100000
3. Under faults: PROM Error is On
4. Knock counter is 13. There is one knock event at 109secs and then 12 in rapid succession at 401secs right when your problem happens.
5. No knock retard was applied at either instance

You'll need someone more expert than me to tell you if any of that is significant but in the first instance I'd disconnect the battery for 30-60 minutes to let the ECU lose any learning it may have done.
I'd also consider backing off the timing a few degrees to see if the knock events went away -- but again see if the experts have comments.

[G UK]

Thanks for looking, the prom ID errors I assumed were something to do with the V5 switcher. The knock occurs when the coughing and misfiring is happening which makes me think it is more a false reading, I could be wrong though.

I have to correct myself however as when I said flat-line I meant in comparison to the rest of the run rather than actual flat-lining and it may be perfectly normal but I haven't got a good scan to compare to!!

[dapinky]

Graham,

Does this help at all - not only is it a baseline scan file, it's from your car (when Paul had it).

It was taken when everything was working, so may give you some sort of comparison (unfortunately I'm a bit of a muppet when it comes to interpreting these things, but it may help someone else to compare what was, to what is.

[G UK]

Thanks for that!! The drop in Spark Advance to a (nearly) flat 13 deg is perfectly normal under boost then but the mental post drive idle is hard to tell as this run hasn't really got one. Arghh, still nothing obviously wrong as far as I can see!!!!

Edit: I need a good look tomorrow mind as long as I remember to load both files on to the laptop and take it on fire duty with me!!

[rip]

The drop in Spark Advance to a (nearly) flat 13 deg is perfectly normal under boost

I read the reason for that fairly recently.
Under boost, the charge density is high so the fuel in the cylinder burns very quickly. More advance is therefore required off-boost when it takes longer for the flame to spread throughout the cylinder.

[GeoffSmith]

Spark advance looks fine to me. I question the PROM ID though (never seen that before), is this one of the standard ones from ESM?

The knock counter is just the result of the rough running - the knock detector is similar to a microphone so heavy spluttering such as you have experienced will be detected as well as knock (pre-ignition detonation). I am suspicious that something not monitored by the ECU is going AWOL - HT, plugs, fuel etc. by the fact that your idle before the run is fine (±12½RPM) whereas afterwards it is ±50RPM and more.

[Candles]

When I swapped my engine over, I forgot to re-attach the ECU's earth leads to the (UK) offside-rear of the intake plenum. They are an easy one to forget as once the engine is back in and the power steering reservoir installed, there's no room to see the earth point. This manifested itself in very erratic behaviour which sounds a bit like yours, especially the mad spark advance. I discovered it by chance, and whilst it was running touched the three wires onto the plenum: every time I made contact it ran completely smoothly. I felt like a bit of an idiot but was glad to have found the problem!

The ECU also has earths connected to the coil base-plate, which would have a similar effect if left disconnected I'd imagine.

[G UK]

Well I found the reason for the coughing under boost with a little help from Neil. Whilst we had the cars together we swapped the coil packs and suddenly his car had the same problem. As mine was also still doing it I rechecked the plugs and found that the ceramic insulation on the top part of the body had become cracked was falling apart causing arcing through on number 1 , the actual electrode and insulation on the base was oddly fine and showed no sign of poor running but the main ceramic body was badly damaged which makes me think the plugs may have been damaged when they were fitted and have since disintegrated.

On replacing the plugs the problem was solved and bizarrely even after swapping the coil packs back I've now got full boost so god knows what the problem was with them when Neil tried them in his car.

Despite having fixed the under boost problem I've still got a rough idle although I am yet to reset the timing after all the mucking anout with the coils and plugs.

[dapinky]

so god knows what the problem was with them when Neil tried them in his car.

Perhaps your coils are putting out a higher 'oomph' than Neils, and either seek out or capitalize on any weakness in the plug insulation - maybe Neil's plugs could do with a change as well?????

[GeoffSmith]

I am suspicious that something not monitored by the ECU is going AWOL - HT, plugs

Cough cough!

Perhaps the coil pack has been overheating due to the plug breaking down - this could explain your time related roughness and possibly the re-occurrence on Neil's car.

[G UK]

Still not been on the site much recently so a bit of an update.

All problems gone except for the rough idle, this is after changing the plugs and swapping the coils over to Neils car and back onto mine again. God knows what caused the plugs to break down or what caused the coils to play up on Neils car but it all appears to have turned out alright in the end.

[Paul.P]

Glad to hear you've just about got blacky sorted Graham, the coil pack was changed just after last group buy, wouldn't think i did more than couple thousand miles with it, & the plugs were the iridium ones that should of lasted approx 100,000mls i believe, they certainly hadn't done more than 10,000, so surprised it had started to break up, maybe just a one off duff one ?. Anyway, i'm glad to hear your finally winning & hope you have a trouble free year, must admit i do miss her !