Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Marks DTM Calib]

I wonder if the LPG setup on a lighter throttle does not like the retard applied by the ECU.

As many know, when the autobox goes for a change, it tells the ECU which applies a retard value to the ignition timing to back the power off and give a smoother shift.....


I suspect that the LPG fuel or even the ecu is not so happy with this

[Kevin Wood]

I wonder if the LPG setup on a lighter throttle does not like the retard applied by the ECU.

Could be. I've wondered about this as well but as things we have changed have made some difference it had been left on the back burner.

Not sure the LPG ECU would see the retard as it doesn't see the spark signal - unless the injector signals and/or the tacho output from the main ECU are also retarded - possible, I guess.

Could be, as you say, the fuel itself can't take this much retard (or can't at the mixture we happen to be burning at WOT).

Other options on my "Back Burner", to throw into the pot:

1) Retarding the ignition pushes the timing of the spark closer to TDC, which is where the peak cylinder pressure occurs. We know LPG requires a stronger spark so maybe it is beyond the capabilities of the ignition system to ignite it at the increased pressure around TDC as opposed to typical full-throttle advance (20 odd degrees BTDC?).

Would have thought the main ECU would detect a misfire then, though. :-/ It would also be variable between cars IMHO, as ignition systems do vary in effectiveness. Might be worth trying a set of more closely gapped plugs though?

2) The main ECU does something odd to the injector signals during an upshift like missing an injection cycle or hold the injectors open for a full cycle so the LPG ECU misses a trigger. This could throw the LPG ECU for a couple more cycles, depending on how it triggers its' own internal scheduling based on the inputs from the petrol ECU.

I do see the petrol injector duration spike to quite high values in the LPG live data - an indication that it has held the injectors open or perhaps that the LPG ECU has interpreted something that way?

3) Did wonder if DBW cars are closing the throttle during upshifts, which momentarily throws the vacuum referenced LPG vapour supply. Discounted as problem also present on non-DBW cars, so vacuum and vapour supply, should be stable throughout the upshift.

One problem is that the live data from both Tech 2 and the LPG ECU isn't really collected fast enough to see what's going on. Maybe I need to get a scope on some of the signals and see if I can spot anything. :-/

I also have a wideband lambda sensor that could be jury-rigged onto an Omega with a bit of messing about. Not sure whether it'd be fast enough to tell us anything interesting, though. One thing we could do is to ensure that the fuel mixture is optimum at WOT. Not something we can really see without it.

Kevin

[feeutfo]

... its now the opposite way round i suppose you might say, on 1.2 bar vap pressure gear changes where fine driving normally at sensable throttle settings, only when you booted it did gear changes lag, and lag quite heavily in comparrison to the current 1.1 bar, which now is a joy at full throttle, but shows a much smaller amount of lag certainly when lifting from full thottle, although this can be driven round if you lift gently, and i suspect a tiny amount during normal throttle gear changes, hence my switching back to petrol to check changes are still smooth.

Generally though the whole problem is improving with less vapourisor pressure, hence Lazydockers sugestion of going further to 1.0 bar perhaps.
hth
 :-/ :-/ :-/

edit, cross posted with kevs previos reply.

[Kevin Wood]

By changing the pressure and remapping we might well be changing the mixture at WOT - hence the thought about getting it right with a wideband lambda sensor.

Bogging while coming off the gas is a sure sign that the mixture is (momentarily) not right, however, the petrol ECU would be making corrections at this point to compensate for wall-wetting of the intake manifold, which doesn't happen on LPG.

We also have to bear in mind the slowness of the LPG injectors, whether that's compensated for by the LPG ECU and whether there's enough time in the engine cycle at high revs for them to complete an open-inject-close cycle.

Kevin

[feeutfo]

cant help but wonder if it needs an lpg specific map, is it ever going to respond correctly trying to match a petrol map when running on gas? is this the key differance between our stag set up and Entwoods install?

[Kevin Wood]

cant help but wonder if it needs an lpg specific map, is it ever going to respond correctly trying to match a petrol map when running on gas? is this the key differance between our stag set up and Entwoods install?

AFAIK Entwood's setup is very similar. LPG ECU (albeit a different one) piggy-backing off the standard petrol ECU.

Kevin

[Marks DTM Calib]

Not sure the LPG ECU would see the retard as it doesn't see the spark signal - unless the injector signals and/or the tacho output from the main ECU are also retarded - possible, I guess.

It does not affect any of the standard signals, the LPG setup would not see the retard at all hence why it could be a possible.

1) Retarding the ignition pushes the timing of the spark closer to TDC, which is where the peak cylinder pressure occurs. We know LPG requires a stronger spark so maybe it is beyond the capabilities of the ignition system to ignite it at the increased pressure around TDC as opposed to typical full-throttle advance (20 odd degrees BTDC?).

Yes, and worth a go with a slightly smaller plug gap to see if it helps.

Would have thought the main ECU would detect a misfire then, though. :-/ It would also be variable between cars IMHO, as ignition systems do vary in effectiveness. Might be worth trying a set of more closely gapped plugs though?

From memory (I would need to check) but, the ECU needs to see 5-10 miss fires in a certain period to raise a fault code so a short one off hiccup on one cylidner wont result in anything being flagged

The main ECU does something odd to the injector signals during an upshift like missing an injection cycle or hold the injectors open for a full cycle so the LPG ECU misses a trigger. This could throw the LPG ECU for a couple more cycles, depending on how it triggers its' own internal scheduling based on the inputs from the petrol ECU.

You would expect the engine ecu to pulse the injectors after the gearchange and the retard has been removed in order to get a fast power recovery.

I do see the petrol injector duration spike to quite high values in the LPG live data - an indication that it has held the injectors open or perhaps that the LPG ECU has interpreted something that way?

To me, this backs up the above theory

Did wonder if DBW cars are closing the throttle during upshifts, which momentarily throws the vacuum referenced LPG vapour supply. Discounted as problem also present on non-DBW cars, so vacuum and vapour supply, should be stable throughout the upshift.

The throttle valve is just to slow to operate in order to support this. On the original C25XE fitted in Cavs and Cals they had a second throttle valve which had a very fast motor fitted to it to provide a TC function.....needless to say it was quite slwo to work!

[Kevin Wood]

Yes, and worth a go with a slightly smaller plug gap to see if it helps.

That's reasonably easily done.

From memory (I would need to check) but, the ECU needs to see 5-10 miss fires in a certain period to raise a fault code so a short one off hiccup on one cylidner wont result in anything being flagged

I think you're probably right. Also, IME, once it's stored a misfire code it's also cut off that cylinder to save the cat so it would be a somewhat more noticeable problem!

You would expect the engine ecu to pulse the injectors after the gearchange and the retard has been removed in order to get a fast power recovery.

It might well throw in some enrichment and the injectors are already close to fully open. I have spotted a setting in the latest stag ECUs to set a level of fuelling for an "injectors wide open" scenario said to be required on some Mazda engines. Hmm.

The throttle valve is just to slow to operate in order to support this. On the original C25XE fitted in Cavs and Cals they had a second throttle valve which had a very fast motor fitted to it to provide a TC function.....needless to say it was quite slwo to work!

Yep, I think this has been eliminated anyway. Vacuum certainly doesn't blip on a DBW car and, as said, the problem exists on both variants.

The bog down feels more fuel related to me, based on a seat of the pants that's used to mapping Megasquirts. Not a hard power cut like a misfire would give.

Kevin

[UrbanFox]

Am I right in thinking that this only affects automatics?

[Lazydocker]

Am I right in thinking that this only affects automatics?

I would think so... It will get resolved. I'm determined to get to the bottom of it, as is Teilo.

[Lazydocker]

TBH it's so intermittent I'm having issues pinning it down.

Still think it's the injectors though :-/ Although mine seems to be fine with a full tank and get worse or fine when nearly empty and plays up when full (depending when it was calibrated) :-/

What if you calibrate at half full?


Shouldn't the reducer maintain a constant(ish) pressure (within its capabilities) despite tank pressure?

Thing is, the Vap pressure isn't dropping noticeably when the problem occurs :-/

Hmmm, just popping my head in to say I have an interest in this. My 2.6 has a similar issue when on WOT going past 5k rpm.

Interesting... What kit have you got fitted?

[tunnie]

TBH it's so intermittent I'm having issues pinning it down.

Still think it's the injectors though :-/ Although mine seems to be fine with a full tank and get worse or fine when nearly empty and plays up when full (depending when it was calibrated) :-/

What if you calibrate at half full?


Shouldn't the reducer maintain a constant(ish) pressure (within its capabilities) despite tank pressure?

Thing is, the Vap pressure isn't dropping noticeably when the problem occurs :-/

Hmmm, just popping my head in to say I have an interest in this. My 2.6 has a similar issue when on WOT going past 5k rpm.

Interesting... What kit have you got fitted?

Its adding to the DBW pool as well, occasional on mine (granted less likely due to not as high flow needed)

Are any cable throttled engines affected?  :-/

[feeutfo]

Am I right in thinking that this only affects automatics?

I would think so... It will get resolved. I'm determined to get to the bottom of it, as is Teilo.

afaik yes, only autos.

[feeutfo]

TBH it's so intermittent I'm having issues pinning it down.

Still think it's the injectors though :-/ Although mine seems to be fine with a full tank and get worse or fine when nearly empty and plays up when full (depending when it was calibrated) :-/

What if you calibrate at half full?


Shouldn't the reducer maintain a constant(ish) pressure (within its capabilities) despite tank pressure?

Thing is, the Vap pressure isn't dropping noticeably when the problem occurs :-/

Hmmm, just popping my head in to say I have an interest in this. My 2.6 has a similar issue when on WOT going past 5k rpm.

Interesting... What kit have you got fitted?

Its adding to the DBW pool as well, occasional on mine (granted less likely due to not as high flow needed)

Are any cable throttled engines affected?  :-/

Yes, Lazydee's. And soon, TB's i guess.

[Lazydocker]

on the way home a rare chance for full throttle run into top gear from a stand still revaled faultless gear changes
no problem at all, couldn't fault it. But lift and it changes early and lags a fraction still, as it would.

I have found this too... If you get it into 6k+RPM it seems to change perfectly, just if it changes anywhere between about 5200 and 6k it has the lag...

Really odd :-/ :-/