Fuel trims all over the place!

[LPGC]

Might be wondering what was up with petrol trims at 15%.. Bits like petrol system MAF, MAP, lambdas can cause similar. As can a blocked cat or incorrect engine timing.

On both fuels are we talking fast trims / slow trims / fast trims pulling slow trims?

Had the laptop connected up yet to check things like temp and pressures readings on the LPG system?

Simon

[57jam89]

Well I watched live data today on my way to work, 1st 30 mins on petrol. The trims are at idle around - 2% to 1.8%, light/normal acceleration 13-17% and cruising around 11% but increasing to 20% under load.
Then watched lpg trims - idle around -6%, light/normal acceleration 7- 11% and cruising around 7% but sometimes jumping to 25%.

These are short shift trims and both banks read about same.
I did notice then on petrol the trims are slowly changing and only by a few % each time but on lpg they are changing every quickly and jumping big % (like 10, 20, -5, 11% etc).
Spark advance is changing from 7- 25° (is this normal?) And maf etc I think are normal.
Cats in both banks have been gutted.
RP didn't come up.

[LPGC]

Just re-scanned through the thread... Jag S type with fuel return, it could be that - Next step maybe monitor fuel pressure on OBD.

Fast trims effect slow trims over time.. We pretty much add slow and fast trims together to obtain actual trim. E.g. If slow trim is at -20% and fast trim is at +20% then actual would be near zero and we might expect a near zero slow trim to be learned over time. When slow trim is learned we expect fast trim to fluctuate just a little (not 20%!).. In this example we could assume fuelling wasn't far off what we'd expect (at this point) and could also assume that fuelling must have been rich for some time until recently.

Worth noting that trim errors can occur after a vehicle has been repaired, because before the repair trims might have been incorrectly learned over time... Say there's a long standing issue such as MAF sensor reading (incorrectly) higher airflow than actual - under such condition the slow trim might gradually make it's way to -20% without generating an OBD error. But if you then replace the MAF sensor with a new one so reads correctly, the slow trim will still be at -20% but the fast time would now be +20%.

Some petrol ECU's interpret high (or highly negative) fast trims as an error condition so will not change the slow trim (remember slow trim is supposed to be effected by fast trim), this condition can result in generation of an OBD trim error even if the engine sees the correct mixture. / Some ECU's cap the fast trim, preventing the engine from seeing the correct mixture, so again resulting in an OBD trim error for only a slightly different (but potentially bit more serious) reason.

Most times fuel trims are mentioned we are talking about slow fuel trims, fast trims usually pull slow trims and since actual trim is slow trim + fast trim, as slow trims gain value so fast trims lose value, so eventually the fast trim value should fluctuate only slightly and average around zero. The more accurate the fuelling, usually the smaller those final fast trim fluctuations will be, so sometimes you'll read installers talking about little fast trim fluctuations or comparing fast trim fluctuations on LPG to fast trim fluctuations on petrol... But it's no good talking about fast trim fluctuations if slow trims aren't also correct (usually not too far from zero and not much difference between petrol and LPG)... Handy to know both slow and fast trims!

Spark advance will quickly change by a large value depending on engine load, rpm, etc... normal.
Why reckon your MAF is normal - What are the readings showing?

Simon

[57jam89]

Ok thanks for info.
I think the maf is working ok as it's increasing with throttle.

Checked fuel pressure and it's the same on lpg or petrol at around the 375 kpa.

Also looked at long term trims today on petrol and then short trims are around 15% long term are 10%.

[LPGC]

MAF might be dirty or on it's way out... When this is the case the MAF might still show increasing flow as actual flow increases but just not show the correct (high enough) level of flow. If MAF shows 10% less airflow than actual we might expect trims to be +10% etc...

Unlikely you have a vac leak as that would show as high+ trims at idle, becoming more correct under load. An air leak between MAF and throttle body, however, could also give symptoms you're having - unlikely as you'd notice one of the large diameter intake air pipes not connected properly.

Simon

[57jam89]

Well I have fault code PO106 Manifold absolute pressure/barometric pressure circuit range/performance come up today.
Could this code caused due to the fuel trims being out?

[Tubbs]

Be worth running it on petrol and seeing if it logs the same fault. I know it hurts ! But sometimes needs must !

[57jam89]

Hi all,
Last night I swapped the MAF and cleared all fault codes. I then drove to work today and back home, on the way back I took some details/pictures of live data.
I have now been driving on petrol for the last few days and still on petrol -

At idle after 30 mins driving-



Cruising at 50 mph-



Stopped engine for 5 mins and the restarted, at ilde-


Then cruising-


Then idling before switching off-



I then rechecked codes and have PO106 and PO108 pending.

Notice my fuel trims have now gone back to normal but not sure the MAP readings!
What should the MAP reading be?

Thanks.

[LPGC]

Well I have fault code PO106 Manifold absolute pressure/barometric pressure circuit range/performance come up today.
Could this code caused due to the fuel trims being out?

Won't be directly due to trims being out. It is just about possible that there's a connection with what you said - If ECU has noted MAF readings were out it could use MAP as primary means of fuelling (limp home mode) and then notice a problem with the MAP sensor - But more likely it would notice a dodgy MAP sensor in any case.

From memory I believe you MAP sensor is at the back end of manifold, opposite end to throttle body.

Notice my fuel trims have now gone back to normal but not sure the MAP readings!
What should the MAP reading be?

100kp is atmospheric pressure so at idle should be around 30kp and flat out near 100kp.. That's if your scanner is reading data correctly etc(!), most likely is. So maybe your map sensor was dodgy not the MAF sensor, maybe both (unlikely) or maybe petrol ECU or at least the reference voltage or an earth or dodgy..

I'd be checking reference voltage at the MAP sensor, believe it should be near 5 volts. Would also swap the old MAF sensor back in and see if trims go wrong again.

Simon

[57jam89]

Thanks Simon, yes map is at rear of manifold and I have a spare one.
You said map readings should be at idle around 30kp and flat out near 100kp, so whys my scanner reading 178,200,125,86 and 71?

Thanks.

[LPGC]

so whys my scanner reading 178,200,125,86 and 71?

May be due to scanner displaying incorrect data (interpreting a different stream in the data flow as map or just scaling), some strange aspect of the ECU (not showing correct data unless map sensor is primary input i.e. MAF noted as dodgy), dodgy map sensor, dodgy wiring, dodgy ECU (or just one aspect such as reference voltage)... Would think favourite is dodgy map sensor or wiring. But I believe there is a vac connection or two near the map sensor, if such were disconnected it wouldn't account for your scanner showing 100+ readings but could maybe account for your fuel trim and map sensor codes.

Further to previous post, you should be able to check not just reference voltage but also earth and output voltage at the map sensor... Without driving the car (you stood beside it with a multimeter), fast idle should give one extreme of voltage (most vacuum), engine off ignition on should give the other extreme (atmospheric pressure). Both extremes will only be between the reference voltage and earth so only extreme in context! Also check consistency - idle with warmed up engine should always give similar results etc.

Simon

[57jam89]

Well I do some checks today,
MAP sensor - with ignition on engine off I have ground at pin 4 which is fed from ecu and 5 volts at pin 2 again from ecu. Did not check with engine running as it's to hard to connect too!
MAF sensor - pin 1 is fed + from a relay which is fine with ignition on, also have ground from ecu to pin 2, at pin 3 (signal wire) there's 12 volts which going by jaguar's workshop manual there should be 0-5 volts. If I pull off the plug from the MAF there is still 12 volts at pin 3 in the plug!

Also checked air duct from MAF to manifold and both vac connections at rear of manifold all fine!

I had swapped the MAP sensor but run out of time to look at live data.

[57jam89]

I have now done 2x1 hours driving on petrol and it's running spot on, map, maf, fuel trims fine and no fault codes.

So it was a faulty map sensor, like you said Simon.

Thanks for your help.

[LPGC]

Thanks for your help.

No probs, what's the scanner showing for map now?
Simon

[57jam89]

[quote="57jam89"Thanks for your help.

No probs, what's the scanner showing for map now?
Simon[/quote]

Hi, most of the time it's around 40/50 mark and around 90 then flat out.

Thanks.

[57jam89]

Hi all,
Over the last 10 days I have noticed my jag has a misfire under light load (then going back on throttle after lifting off at cruising speed). I checked the map and found lpg does not match petrol one, I have adjusted the map 3 times over the last 10 days but just can't get them to match.

Here's the map today -



After fitting the new Europegas Oscar ecu and magic injectors last April and getting tubbs to check/set it up, it's been running fine.

Any ideas?
Thanks.

[LPGC]

Any ideas?

Plenty ideas, I also have plenty ideas on how to ride a bike... a bit easier to teach someone how to map an LPG system remotely than to teach them how to ride a bike remotely but you'll have got my point.

There are problems with the matching the lines calibration method - Points plotted on those lines equate to petrol/LPG injector pulse duration against manifold pressure and are plotted regardless of rpm. However, the amount of air (and hence fuel, and hence injector duration) the engine cylinders get also depends on engine rpm, as the cam profile and valving arrangement of the engine also effect how much air etc the engine cylinders will trap at a given manifold pressure and rpm. If you gather petrol points at a specific rpm you must also gather LPG points at the same rpm to compare like with like. In addition there is another problem with matching the lines method on your model of vehicle, given that it's an S Type Jag and needed a petrol return to be fitted - The petrol line is plotted with the fuel return disabled and the petrol ECU seeing normal fuel pressure, where the LPG line is plotted while the petrol injectors are not operating but the fuel return is. If there is a difference in petrol pressure (and there will be) at any load point for the engine between running on petrol and running on LPG, then this will mean you actually should see a difference between plotted points running on petrol and running on LPG. As said before, you cannot use the compare the plotted points method to calibrate your LPG system, you must calibrate the system using the fuel trim method. It also helps vastly to be aware of injector performance and to be able to anticipate correct shape of the map.

Your multiplier map contains multiple peaks and troughs which really shouldn't be there (discounting the peak before 2ms)... In addition, towards the higher load area (say 12ms plus) at high enough rpm, the petrol system will probably go open loop and therefore plot exactly the same point on the gas line as on the petrol line regardless of where you set multiplier and regardless of the engine receiving a lean or rich mixture.. It seems your system hasn't plotted points for the high load area anyway, so how did you set mixture at high loads? Your screen shot shows the engine running at low load where the multiplier is near 1, yet gas injector duration shown would reflect a multiplier of nearly 1.3.. This might prompt me to ask - What injectors are selected in settings screen? What reference pressure is entered in settings screen?

How are you finding tech support from your suppliers? Have they explained any of the above?

Might have been better all round if you'd bought a kit from such as myself or Dai, where real and knowledgeable tech support is included. Hope to see you to sort your latest problem.

Simon

[57jam89]

How are you finding tech support from your suppliers? Have they explained any of the above?

Simon[/quote]

Hi Simon, I like the bit above and you know the answer!

And yes the only way to sort this out is a visit to you, but being 190 miles from you it's the old problem of finding time.
I will phone you and arrange then I know I have the time (but it's not going to be in the next 2 months).

Thanks.

[LPGC]

How are you finding tech support from your suppliers? Have they explained any of the above?

Simon

Hi Simon, I like the bit above and you know the answer!

And yes the only way to sort this out is a visit to you, but being 190 miles from you it's the old problem of finding time.
I will phone you and arrange then I know I have the time (but it's not going to be in the next 2 months).

Thanks.

Only 190 miles away! That's not far compared to the distance some customers travel to get here for repair of any type of LPG system problem or conversion of certain (complicated to convert properly) vehicles... Examples of customers coming here from similar and further distances on this forum - usually after they've paid installers closer to their home for ineffective repairs. When they come here it gets correctly diagnosed and cost effective repairs are discussed/chosen/done every time.

I know the answer but a lot of people considering a DIY LPG conversion don't...

Much as I enjoy talking LPG, other tech subjects and helping people out, I partly come on forum as a way to advertise and get skills noticed. Someone on a very similar thread (to this thread) had it pointed out to them (in advance of them buying a DIY kit) that they could buy a DIY kit from myself (LPGC) or Dai (Classicswede) which would include very thorough tech support (over and above what could be expected of a supplier, such as in depth discussion at any time in or out of office hours with someone who really knows what they're talking about, gets better results than the supplier anyway, and is prepared to explain things clearly and in detail). Additional benefits would include free check of their workmanship, free calibration and free cert from 'a competent person' if they'd care to visit.. But the person decided to save just a few quid by buying direct from a supplier, then found the supplier tech support not all they'd expected, then came on this forum seeking tech support where myself and Dai are the only likely experts to reply... Which seemed a bit cheeky! Still, as it happens I did give the person advice on forum, though being blatantly cut out of the deal I very nearly didn't... People should expect a poor response if they were to e.g. buy plaster from a builders merchants and then ask the builders merchants next door how to do plastering (because the first merchants prices were a bit cheaper but they don't teach people how to plaster very well - not that anyone at the first merchants knows how to plaster properly themselves anyway!).. I advised because it served me to show skills while at the same time showed that the person would need to pay some installer for calibration if not buying a kit from myself or Dai, but I might just as easily have decided I would sell more DIY kits if I hadn't helped. Now it's my intention to sell some more DIY kits, so in future regardless of if a DIYer is aware that I or Dai sell DIY kits, if someone buys a DIY kit from elsewhere they will be left to it as far as I'm concerned.

The way it used to work was like this... Suppliers would not supply to amateurs, a middle man was involved who might be a pro installer or might actually work at the suppliers but moonlight giving tech support outside work hours. Suppliers wouldn't mind moonlighting, they get to sell another system (through the middleman to the amateur) but didn't get staff tied up giving tech support to amateurs. Only after an amateur had completed several installs and become proficient, so no longer really an amateur, would suppliers sell to them direct. One of the cheapest suppliers, also guilty of the worst tech support and service looks set to be a supplier no more, I've heard of talks between other suppliers of raising prices to the public but giving pros discounts so in general prices to pros will remain about the same - Suppliers probably still don't want to be tied up giving tech support to amateurs on a kit which (these days) they might have made £10 profit on. I would invite any supplier who says different to have an on forum comparison on prices and levels of service between what I or Dai offer compared to themselves.. The way it used to work looks maybe set to return to be the way it works in future.. This could be better in many ways, particularly in terms of results the DIYer achieves, including safety, and could help pros (especially those that don't sell DIY kits with a good level of tech support included - most) remain in business.

Most of the above could warrant it's own thread, I may well start that thread.

Simon

[57jam89]

Hi Simon,
It's good to hear some people travel more than 190 miles for you to sort their problems out, I don't have a problem with the mileage it's finding the 3 and half hours each way and the time you need which I'm finding hard!

And yes I fall into the category of than person who buys parts at the lowest possible price without tec support and then come on here looking for help. Maybe people like me need to removed from this fourm!

I do realise I don't know what I'm doing in setting up my S type correctly that's why I will now pay you to sort it (if your still interested in doing it).

Thanks.