Singlepoint LPG technical question

[GavinC]

'95 960 with Tartarini Tek '97 singlepoint system.

I presume, because the system has a stepper motor... it is technically 'closed loop.'

I PRESUME the Idle Air Control Valve simply draws petrol or LPG as switched. So if idle fluctuates on petrol, but is silky perfect on LPG - the IACV cannot be the problem. (I presume...)

Would the Throttle Position Sensor be part of this operating system for both petrol AND LPG?

On the other hand, from experience over years, the MAF does not appear to operate for the LPG side of the fuelling system.

What sensors would be common to both fuels?

I'm asking the question because the car idles very smoothly and runs perfectly on LPG... but far from it on PETROL just now...

Or could there be anything in the LPG system (eg. an electrical 'short' or a failing emulator) that is somehow impacting on the PETROL side? Could that be at all possible??

Because it might appear to be primarily a 'petrol' running problem I also have a recent post in the 960 section - 'Idling Problem.'

Thanks in advance for any pointers.

[Clifford Pope]

"Would the Throttle Position Sensor be part of this operating system for both petrol AND LPG?"

I suspect not. I've just had exactly these symptoms on a 240, and the reason for the sudden hunting at idle on petrol was simply that the throttle link rod had twisted so the the TPS was clicking in as the engine rocked at idle. It was rock steady on gas, like yours, so I concluded that the gas idling must be independent of the TPS.

IAV takes pure air when running on petrol, but gas/air mixture when on LPG.
I've noticed that gas being "drier" the valve tends to stick a bit if the car is run exclusively on LPG without regular breaks for petrol operation, which seems to keep it free.

[richjc4]

is the lpg and petrol running of the same lambda sensor?

[GavinC]

There's just one lambda sensor on this '95 960 and I believe both petrol and LPG run off it. The sensor was new - and a new cat - 18 months ago.

Thanks to you and Clifford for your interest.

[classicswede]

What you need to do is put some fresh liquid gold in the tank. Run it on pertol for say 20 miles and then see if it idles ok. If the car will then idle fine on petrol the issue is caused by the ecu picking up bad habbits from running on LPG. The only way to solve it without updating the system would be to fit an ecu reset to clear the memory every time you switch off.

If after a run on petrol it still idles poorly then you will need to remove the idle control valve and give it a good clean. The idle control valve runs in a slightly different area when running on gas and can get a build up of gunk stopping it from moving t o where it needs to be for petrol running.

If non of that works try running some injector cleaner through to clear out any crap in the petrol injectors.

[Clifford Pope]

Originally Posted by classicswede

The only way to solve it without updating the system would be to fit an ecu reset to clear the memory every time you switch off.

Can you explain how, or point to instructions somewhere?

Can this be done while running - eg if switching from gas to petrol on the move?

[GavinC]

Fixed! Thanks everyone. Fresh BP Ultimate and ran for around 30 miles, no change. Checked operation of Idle Air Control Valve, checked voltage to and ohms resistance at Throttle Position Sensor, checked Throttle Body and Butterfly for clean and non-sticking, cleared OBD codes. CONCLUSION: Fixed by disturbance of electrical connector at TPS??? Or fresh petrol (classicswede's suggestion)... and it may simply have taken a little longer to 'clear' it. Clifford still has a query arising, however. Classicswede?

[Clifford Pope]

Glad to hear the problem is fixed.
But then, I thought mine was too! I did all the things you did, and also found the link rod seemed a tiny bit out of adjustment. But at the weekend after a 200 mile run on gas (30 mpg at 51 p per litre!) it started hunting at idle when I switched back to petrol.
So has the Idle Air Valve really gummed up in that short time?
Hence my interest in the idea that the real cause is the ECU learing bad habits, and that automatically resetting it might be the solution.

If I knew which wire fed the ECU power I suppose the simplest method would just be to have a supplementary switch and turn it off for a few minutes after stopping the engine. (Like disconnecting the battery, but without losing the radio code and all the presets)

[classicswede]

The closed loop LPG systems normaly have a a means of memory reset using the yellow wires. There is an AEB memory reset relay (try tinley tech)

The other option is to try running with the lambda signal wire off. When teh labda is disconnected the ecu will read 0.5 volt and run on base map. If your car will run fine like that fit a relay to the lambda signal wire powered by the Blue gas live supply to disconnect the lambda while running on gas. That way the ecu will not make any adjustments while on gas.

[Clifford Pope]

That sounds a neat solution - thanks very much.
Where do I pick up the Blue live supply wire - at the changeover switch?

Just a question though - one of the codes that pops up says no signal from the injectors (pretty obviously because they have been turned off). Others have similarly been looking for functioning of the TPS and AMM. Would disconnecting the lambda signal wire avoid all spurious error messages, or only those relating to mixture?