Spark Advance freaks out under load

[pineapplecharm]

Like DavieBow, I used to get 41,42 and 54 engine codes every now and again. I've replaced the HES (twice, as I was worried I broke the first one) and haven't seen Code 42 since, and 54 appears to have snuck off too, but 41 still pops up from time to time. I replaced my coil packs last Summer and that seemed to get rid of it for a while but now it's back. It usually comes on when the engine is restarted warm and only very rarely does it come on while moving.

As it's so unreliable it's been hard to investigate and, as the engine ran normally, I was inclined to ignore it. This, it turns out, may have been a mistake. I now have a new fault, where I suddenly lose power when accelerating hard. It falters, makes a weird farting sound (although Mrs pineapplecharm has suggested this may not be the car) and feels as if the engine is not sparking, or is being starved of fuel. I connected ElanScan and pretty much everything I looked at was normal, with the following exception:

Spark-advance-blip.png

Notice how the Spark Advance tracks the throttle position pretty accurately, until I stomp on the power at which point it just slumps.

With more gradual pedal work things aren't quite so terrible, but it's still noticeable:

Spark-advance-gradual.png

The service manual says:

The Direct Ignition (DI) system does away with the conventional distributor and uses two separate ignition coils, a cam angel sensor, an ignition module and Electronic Spark Timing (EST) control circuitry incorporatedi nto the ECM.

At engine cranking speed the ignition module (part of the ignition coil pack) alone controls the spark advance, but at speeds above 800rmp, the ECM takes over ignition timing control based on inlet manifold air pressure, air temperature, coolant temperature, engine speed and detection of the onset of detonation.

This sounds horribly like it's the ECU throwing a fit about something. It could be one of the sensors but with no errors, and nothing in the scan to suggest any of the inputs mentioned is out of range, I'm struggling to blame anything but.

Comparing manifold pressure to spark advance across the whole run, for example, doesn't look too wild; in fact at higher pressure it seems to have less variance.

pressure-advance-whole-run.png

Same with throttle and engine speed - nothing stands out; in fact it looks tidier at the top end than lower down when the engine is running under light load.

I'm assuming a new ECU is remortgage time so any inspiration for what else to look at very much welcomed! The .ecu file is attached, in a zip archive. Power slumps were observed at 94sec, 110 sec (mild), 146 sec, 185-188 sec, 233-236 sec, 274 sec and 300 sec (mild)

[Giniw]

I'm not a specialist, but to my understanding it's quite normal that the ECU reduces the spark advance at wide open throttle and full boost.
It just tries to avoid knocking. If the petrol octane were higher it could keep a higher spark advance as the knock risk would be lower, and that would raise the power of the engine (some people who run on ethanol which has a higher octane value than petrol modify their ECU advance tables to get more power from their engine, but I have never heard of an ethanol converted Elan).
Anyway, I don't think you have a spark advance problem.

I now have a new fault, where I suddenly lose power when accelerating hard. It falters, makes a weird farting sound (although Mrs pineapplecharm has suggested this may not be the car) and feels as if the engine is not sparking, or is being starved of fuel.

I have never experienced it so I can't be sure, but that sounds a lot like an overboost resulting in a fuel cut.
I had a quick look at your ElanScan log and your max boost is around 0.78 bar, which is too high on a stock Elan (should be ~0.65 max).
I you have a stock exhaust, the usual problem is a sticky wastegate.

EDIT: your barometer reads 0.78 bar. A bit suspicious unless you are at around 2000m of altitude.

EDIT 2 : your knock counter seems really high to me. I have always had 0 on mine, except when stopping the engine where the counter sometimes go to 1 or 2 but it looks normal as the engine shakes a little when it stops. I have never seen any retardation on my car, yours does have 2.5° because of the detected knocks.
And by the way, you do have a fault code in your log (EST monitor error)

[Simon_P]

Have you checked the Base timing according to the instructions?

[pineapplecharm]

I had a quick look at your ElanScan log and your max boost is around 0.78 bar, which is too high on a stock Elan (should be ~0.65 max).
I you have a stock exhaust, the usual problem is a sticky wastegate.

This has come up several times in my online investigations but I never thought to check what the turbo pressure should be. Thank you! I will service the wastegate first opportunity I get.

EDIT: your barometer reads 0.78 bar. A bit suspicious unless you are at around 2000m of altitude.

Interesting; if the barometer is reporting wonky numbers that can't be helping the ECU manage the turbo pressure.

And by the way, you do have a fault code in your log (EST monitor error)

Yep, that's the intermittent Code 41 as I mentioned, which is what sent me on the spark advance line of thinking. Now wondering if the excessive pressure is causing it to try and over-compensate, which is what's throwing the error.

Awesome reply; thank you.

Have you checked the Base timing according to the instructions?

As it happens, yes - new timing belt not two months ago. Or do they 'bed in'? Might be worth borrowing the timing gun again just to check. Thanks!

[Simon_P]

Low pressure reading will affect all conditions.

Did the issue start after the cambelt change? My other suggestion was going to be loose cambelt or worn CAS drive as this would allow the variation at low load/revs which would reduce as the load/revs increase.

[pineapplecharm]

Did the issue start after the cambelt change?

This issue started well before then. Timing was my first thought and when I took the cover off the belt I noticed it was worn.

[GeoffSmith]

EDIT 2 : your knock counter seems really high to me. I have always had 0 on mine, except when stopping the engine where the counter sometimes go to 1 or 2 but it looks normal as the engine shakes a little when it stops. I have never seen any retardation on my car, yours does have 2.5° because of the detected knocks.
And by the way, you do have a fault code in your log (EST monitor error)

Code 42 EST (Electronic Spark Timing circuit) monitor error - Workshop Engine Management Manual section EMK4 Page 52 and 53 [from elsewhere on LEC].

This may be useful background.

[GeoffSmith]

EDIT: your barometer reads 0.78 bar. A bit suspicious unless you are at around 2000m of altitude.

The barometric pressure is the pressure reported by the pressure reported by the MAP sensor when the ignition is turned on. Low value explains the supposed high levels of boost and needs checking.

This and the EST problem could mean that there is an earthing problem.

[Danevansgolf]

take out your spark plugs when your engine is hot. Had similar problems like yourself and found my plugs were cracked and only visable on a hot engine. Hair line cracks in the ceramics will give you your symptons. Basically when you open the throttle hard the spark escapes giving you a feel of fuel cut.
Good luck. Keep us updated

[pineapplecharm]

Keep us updated

Yeah, sorry about that. It literally took a year for the stars to align sufficiently to get this sorted!

I you have a stock exhaust, the usual problem is a sticky wastegate.

In the course of investigating this, 2 of my 3 turbo shield bolts lost their heads (it's now held in place with a giant jubilee clip) but loosening the wastegate arm has fixed the problem completely.

Still getting code 41 but only intermittently and the car drives 100% fine. I'm just annoyed at myself for letting such a small issue become such a huge mystery.