LPG Discussion Forum
http://lpgforum.co.uk/

Do modern car ECUs advance timing when running on Lpg?
http://lpgforum.co.uk/viewtopic.php?f=33&t=14242
Page 1 of 2

Author:  4eyes [ Thu Mar 30, 2017 2:03 pm ]
Post subject:  Do modern car ECUs advance timing when running on Lpg?

I recently purchased a obd dongle in order to read the CVT fluid temperature on my 2011 Subaru Outback. This dongle has also allowed me to look observe other things like timing, knock correction etc. I run on lpg most of the time, but recently ran out of it and ran on petrol for a while. I noticed the engine was " pinging" slightly at about 1200rpm, especially at low speed on an incline. It does not do this on lpg. Is this caused by the petrol ecu adapting to the higher octane lpg, then struggling with 95 octane petrol.? I have observed the ECU trying and failing to advance the ignition advance multiplier ( i think its called) while on petrol to 1.0 and it didn't do it in a couple of days and about 100 miles of driving. When i switched to lpg, it got to 1.0 ( where it apparently should be?) within 10 minutes! Any input appreciated!

Author:  4eyes [ Wed Jun 28, 2017 3:42 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Any thoughts on this? Had to run on petrol again today for a while, and car runs a little bit @!## while on petrol, slight hesitation while accelerating, but idles spot on. Runs perfectly on lpg, smooth and quiet. Seemed to improve a little the more miles I did on petrol. Trying to convince myself that it's the ECU adapting to lower octane petrol. Would really appreciate comments to hopefully put my mind to rest...

Author:  Gilbertd [ Wed Jun 28, 2017 6:06 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Some will but it depends on what Octane fuel the car was originally designed to run on. I've got a 1990 Maserati Biturbo Spyder which is intended to run on 98 Octane. If I have had to put 95 Octane in it at any time, it runs badly and is noticeably down on power. I assume this is because the timing is being backed off as a result of the output from the knock sensors. It's doubtful that an ECU would be designed to give more advance if the fuel has a higher Octane, they will just run at whatever was programmed as the optimum for the fuel they recommend, but many will retard the timing if given something inferior. Might be worth sticking some 98 or 99 Octane in it and seeing what the ignition advance multiplier does then.

Author:  Brian_H [ Wed Jun 28, 2017 7:49 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

4eyes wrote:
Any thoughts on this? Had to run on petrol again today for a while, and car runs a little bit @!## while on petrol, slight hesitation while accelerating, but idles spot on. Runs perfectly on lpg, smooth and quiet. Seemed to improve a little the more miles I did on petrol. Trying to convince myself that it's the ECU adapting to lower octane petrol. Would really appreciate comments to hopefully put my mind to rest...


How old is the petrol in the tank?

Author:  4eyes [ Thu Jun 29, 2017 6:24 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Thanks Brian and Gilbert. The fuel is fresh. I am thinking it's worth running on petrol for a little while to see if it corrects itself. 95 octane ( normal unleaded) is what's in the tank, but what's stumping me is that the car instantly stops it's slight pinging the second I switch back to lpg. It's like the car has got used to running on higher octane (105 ish?) LPG and don't like petrol ! I thought that a problem on petrol would be amplified on lpg?

Author:  Gilbertd [ Thu Jun 29, 2017 7:55 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

This is just one of the reasons why LPG is better for the engine than petrol. The Octane rating of Propane is 113 and autogas sold in the UK is usually 100% Propane. That sold in most European countries is often a blend of Propane and Butane which has a lower Octane rating (around 95 if I remember correctly) but even at a 50/50 blend that would still give you around 104. Is 95 the recommended fuel for your car or is it expecting to be run on something better?

Author:  4eyes [ Thu Jun 29, 2017 8:09 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

95 is what's recommended, but owners manual says ok on 90. I think 95 is normal unleaded?

Author:  Gilbertd [ Thu Jun 29, 2017 4:06 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Yes, normal unleaded is supposed to be 95 Octane so if that is what is recommended it should be OK. Unless it's been breathed on a little of course, unless you've owned it from new you don't know if a previous owner gave it a remap or other mod. Let the petrol tank get really low and bung in some higher Octane stuff and see what happens then.

Author:  LPGC [ Thu Jun 29, 2017 8:53 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

I would expect an engine that adapts ignition advance to suit the fuel, as many modern engines do, to adapt ignition timing much quicker than what could be imagined from reading the OP's situation. Less of a 'long term trim' like system for ignition advance, more of only a short term trim. If you had super unleaded higher octane petrol in the tank and nearly ran it dry, then filled with normal unleaded you wouldn't expect it to pink. The difference between super unleaded and normal unleaded is less than the difference between LPG and super unleaded, but still I've never known an engine knock on petrol after running on LPG including all the Subaru models.

Could it be a case of broken / unattached knock sensor or similar?

Pity I'm on suspension on Subaru Legacy forum for telling the truth about the LPG conversions on 3 SpecB Legcacy turbos all of which didn't run properly when converted to LPG by BRC installers, 2 of which still don't, and my explaining in detail on what would be the nature of the problem. This happened even though the guy (who eventually took exception to my explanations and summing-ups) had previously (and I quote) written ''Thanks profusely to Simon for all the help he has given with my problem'. Problems which took 6 months to fix and are still not fixed on the other 2 Legacy's. Seems to me it turns out that the installer for the guy who complained about my explanations was his mate because he did say he was his 'Go to guy for all vehicle problems' . he didn't like my conclusions about the system fitted which reflected negatively on the installer and the brand of system fitted (which is probably the only system he fits). This installer couldn't sort the problems himself, he had to draw on the help of BRC's tech support guy who's firm went bust recently (as I would expect for ah installer of this type). The nature of the problem agreed entirely with what I told the guy it would be on the forum moths before he had me suspended. The guy's last post before the thread was deleted claimed that actually none of the problems were the installer's fault because the installer had fitted and mapped the system properly, but the problem turned out to be a mere O ring on the boosted air side leaking. Months previous I had explained how the Seq32 system didn't read manifold pressure directly but learned it during calibration, so if at any later time there was a discrepancy between learned manifold pressure and actual engine load the LPG system would compensate fuelling incorrectly for manifold pressure (in effect not read reducer pressure correctly) and therefore fuel incorrectly and bring about all the problems the guy was experiencing. I explained all this in my last post on the thread which was the last post on the thread (because I imagine the guy had no comeback whatsoever against my very through explanation). So with nothing else to add, he threatened that the forum would be sued by himself and maybe BRC and the forum owner (who seems a decent bloke) backed down. I wish they had threatened me with legal action instead. I would have gone to court and trounced the guy, the installer (who would probably do better just to fit clutches instead of fitting LPG systems and having to rely on tech support from others who are far more competent at fitting LPG systems than him) and BRC with technical knowledge and over-whelming evidence that would prove what I had said was correct and perfectly explained the reason for the problems. The guy had previously asked me how much to sort the problem and I'd given him a no mission accomplished no fee quote which further explained what his problem would be and gave helpful insights. His installer and the installer's 'network' in the meantime had left him with problems for 6 months, which could all have been prevented from the outset. If I had converted the vehicle it would have been an average install to me and would have run perfectly from the day he collected it from me, perhaps not least because I wouldn't fit a system that brings about problems of it's own if there is something as minor wrong as a leaky little O ring.

Simon

Author:  4eyes [ Fri Jun 30, 2017 7:09 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Thanks guys. I did try super unleaded but no change. Car is 2.5 non turbo, and not had any remap ( previously owned by undertaker, seriously)! Maybe a sensor problem but no codes as yet. Car runs perfectly on Lpg though. I thought that an issue on petrol would be worse on lpg?

Author:  LPGC [ Fri Jun 30, 2017 10:11 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

If a knock sensor was disconnected you'd probably get a code but if it just isn't working or isn't bolted to the engine then maybe no code.
If ignition is too far advanced you're much likely to get a knock on petrol than on LPG.

Maybe what you're hearing isn't pre-ignition, petrol injectors make a different noise to LPG injectors etc.

Simon

Author:  Gilbertd [ Fri Jun 30, 2017 11:24 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

It will be worse on LPG but only if it is a fuelling issue. This sounds more like an ignition or ignition timing issue which will be less noticeable on LPG due to the higher Octane rating of the fuel in use.

Author:  4eyes [ Fri Jun 30, 2017 12:08 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Thanks again guys for your time. I agree it seems like timing issue that's only audible on petrol. Just checked knock sensor visually as they can crack apparently, and looked ok. Will maybe replace it soon.

Author:  4eyes [ Sun Jul 02, 2017 11:15 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

If the knock sensor is kaput but isn't giving a code, then if I disconnect it and car performs as normal, that means sensor is broken? If car is worse with sensor disconnected that means sensor is ok?

Author:  Gilbertd [ Mon Jul 03, 2017 10:15 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Got to be worth a try. Normally the knock sensor is a microphone tuned to look for specific frequency ranges so the ECU can recognise pre-ignition. If you disconnect it you may get a fault code, in fact, I would expect you to, but if you don't then that could suggest it's broke.

Author:  LPGC [ Tue Jul 04, 2017 10:44 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Another point related to knock sensors being microphones... If there are certain types of rattles on the engine the ECU can interpret them as engine knock and retard timing, but this would give opposite symptoms.

Might expect unbolting the sensor from the engine to cause more advanced timing, which would make for worse knock.

Don't know what unplugging the knock sensor would do do timing - would it (a) use a default ignition map not applying any info learned by use of the knock sensor, or would it (b) continue to use last learned values. If (a) then unplugging it might prove a dodgy knock sensor, or that it wasn't bolted to the engine properly, or dodgy petrol ECU.

Has it been mentioned yet that a lean mixture on petrol can cause knock... Lean mixture on petrol could be due to LPG system calibrated too rich.

Simon

Author:  4eyes [ Wed Jul 05, 2017 6:13 am ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

LPGC wrote:

Has it been mentioned yet that a lean mixture on petrol can cause knock... Lean mixture on petrol could be due to LPG system calibrated too rich.

Simon

Thanks Simon. I have read on some American Subaru forums that my engine is tuned a little lean in places, Alot of people seem to have this issue with no fix apparent. Would you be able to expand on the point above about the possible rich set up of the lpg ecu effecting the petrol ecu? Thanks.

Author:  4eyes [ Wed Jul 05, 2017 12:00 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Well I loosened the knock sensor on the engine block and the pinging was worse, so am I right in thinking the sensor itself is working?. Still didn't get a code, but I did not try disconnecting the electrical plug as it was difficult to get hands on it without moving other stuff.

Author:  LPGC [ Wed Jul 05, 2017 7:41 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

All other things being the same (stuff like engine oil / coolant temp, air temp, correct fuelling) and without wishful thinking as in 'finding exactly what you're looking for syndrome', I would think that'd show the knock sensor is working at least to some extent. And I would think it'd either work or not work... but could be worth trying one from the scrappies.

Simon

Author:  4eyes [ Wed Jul 05, 2017 9:10 pm ]
Post subject:  Re: Do modern car ECUs advance timing when running on Lpg?

Thanks again Simon. I'm happy now that the knock sensor is ok. After further searching and reading it seems this problem is quite prevalent in modern subarus , especially in the states, all in the name of keeping emissions lower. Ffs. Dealer service seem to think it's normal apparently. Ah well, glad I had it converted to gas now! Runs perfectly on gas.

Page 1 of 2 All times are UTC
Powered by phpBB® Forum Software © phpBB Group
https://www.phpbb.com/