BMW X5 managment light

[PAWEABMW]

Hi folks hope all is good i've been having intermittent problems since i installed an LPGTECH kit into my 3.0 X5 2002 the vehicle will run fine for weeks on LPG with no trouble altho every now and then the engine managment light comes on with the error code system bank 1 too lean,i filled the LPG tank last week on a run to the south it drove 160 miles which is what i get on a tank no problem on lpg,on my return journey i refilled and it started stuttering almost at once running like @!## almost backfiring with the managment light comming on.
I switched to petrol drove for a few miles turned on the gas and had no more problems ....strange cant get my head around it......

Any ideas guys on my last LPG installation on previous vehicles they either worked or they didn't this is annoying as it cannot be replicated on demand........as the X5 will sometimes run for week without issue doesnt seem to be a power thing either as there is no noticable difference in power on LPG.....

[PAWEABMW]

7 months on still having this nasty annoying problem...anyone shed any light.....i have copied a few screenshots from the software a few error codes that make no sense to me.......really annoying as you never know when its gonna happen as a result ive been running on petrol mostly as i nearly crashed when the engine cut out on a bend.....madness

Symptoms

Drives fine switches ove ok no power loss then hesitation,backfire,EML light....stall at idle
OBD code system bank 1 too lean
Lambda heater control failure

once reset it drives fine for a few miles then same again

after a few resets its fine sometimes for over a week

Doesnt happen at any particular rev range or load

might be just me but always seems to happen after i fill the lpg tank full when the level drops it seems fine

Looking at the software info i still have a 0.00 readout on the 2nd lambda sensor despite checking and re checking its connected ok seems to link in with the obd error code but wouldnt explain why it only happens intermittently.....

i've increased the mixture slider in the program to try and find a solution

totally out of ideas now any help would be great guys....

[Gilbertd]

Your code reader has answered the question for you. The heater on the lambda sensor on bank 1 is dead so it's giving no output. No output says it's running lean so the petrol ECU is almost certainly causing the mixture to go fully rich (which you would be able to tell by looking at the short term fuel trims using live data on your code reader) so one bank is drowning. Buy it a new lambda sensor.....

[PAWEABMW]

Thanks man i'm gonna get a look tommoro see whats causing this.....always great to hear from someone experienced thx again

[Gilbertd]

You're being told what's causing it, the lambda sensor on bank 1 is dead. Your code reader is telling you that, the LPG software is confirming it. If it was running lean, as the code reader says it is, the output from the lambda sensor would be around 0.1V, it would read something as long as that bank was still running. It isn't showing 0.1V, it's showing 0.0V all the time because it's dead. Your code reader is telling you it's dead because of a fault on the heater circuit. I would suspect it may be getting hot enough to give an output some of the time so the problem is there mostly when it cools down, when the engine is idling. Nothing to look at, a dead lambda sensor looks just the same as a working one. Replace it, it's sensor 1, left hand bank, upstream sensor by the looks of it.

[LairdScooby]

With that Lambda sensor down it's likely to put the OE injection system into "Limp Mode" as well so you won't even get reasonable economy/performace on petrol either.
Renew that sensor, get the OE engine management system reset to delete the fault code stored for the dead Lambda (probably an OBDII system at a guess, should be possible to buy a reset tool if so) and then recheck your gas system.Whether that stores fault codes or not i don't know as i don't know the system but it ain't gonna work right if the host system (ie the OE petrol) has problems and the gas system is showing you the fault - daresay the presence of the cars MIL is a good clue as well!

[LPGC]

I agree it is likely that the lambda heater circuit is causing the problems - but this isn't necessarily so.

Lambda probes need to be hot before they will work properly, hence the built in heater circuit. But they are installed in the exhaust so are heated by exhaust gas anyway. Some petrol ECUs ignore lambda probes when they detect a broken heater circuit but most don't and I don't think this is the case on this vehicle. In many cases (as they're heated by exhaust gas anyway) they will still output an appropriate mixture signal even with a broken heater, especially during a long run with the engine doing a bit of work like on the motorway.

The lambda probe does need changing first, and I hope that cures the problem, but it might not. Looking at your screen-shots, the installer has used the LPGTech system's Tech mode for calibration and the system may be calibrated too rich across most of the range and maybe lean at high loads anyway... In case of further problems after changing the probe, have a read here http://www.lpgforum.co.uk/viewtopic.php?f=28&t=13877

Simon

[hardcorepit80]

1.Your injectors are not big enough for this engine (or manifold nozzles in wrong place)
2.You have a gas leak somewhere !!!!!
3.Reducer connected (water) in wrong place or reducer faulty, not big enough.
4. Injector 2 and probably 6 faulty or nozzles in wrong place

[LPGC]

Hardcorepit, I can see the reasons why you made your points but might be a bit early to jump to all those conclusions...

1. 3.6ms Pinj to 4.6Ginj is not an excessive multiplier for OMVL SL (or indeed most) LPG injectors around at around 3.5ms mark - Why do you think his injector nozzles are too small? Even if that were the case, could be adjusted out by increasing pressure a little.. Also presumptious without knowing maximum Pinj.
2. Gas pressure too low/high 'errors' could easily be caused by blipping the throttle before switching a hot engine off... Low gas pressure anyway during blipping throttle, maybe holding on over-run, then gas cools down and pressure drops further (lower pressure on starting from cold). LPGTech also gives this error when starting if the vehicle previously ran out of gas! May have a leak, but too early to say...
3. We don't know if reducer and gas temp sensor spec are setup correctly.. Might agree with you, but maybe the reducer is plumbed into the heater circuit after the heater solenoid block.
4. The straight 6 engine only has 2 logical banks / 2 lambda probes. Cannot draw conclusions from the fact channels 2 & 6 are reading 10% longer duration on this screenshot - If there were a problem with fueling on 2&6, this would be reflected across the whole engine not just 2& 6, the petrol ECU isn't that advanced. This is just a screenshot... and firing order on the straight 6 is 153624 - note 6 and 2 are next to each other (so software might momentarily show these cylinders Pinj increase together due to throttle / ICV movement etc). Besides, LPGTech (and other ECU's) just sometimes do that kind of thing - false readings more an issue with the software than anything else...

Setup in the **** LPGTech mode making diagnostics involve more guestimation.

Simon

[hardcorepit80]

1. 3.6ms Pinj to 4.6Ginj is not an excessive multiplier for OMVL SL (or indeed most) LPG injectors around at around 3.5ms mark - Why do you think his injector nozzles are too small? Even if that were the case, could be adjusted out by increasing pressure a little.. Also presumptious without knowing maximum Pinj.

All injectors fully open error.
I wouldn't even think of fitting these to that car...

2. Gas pressure too low/high 'errors' could easily be caused by blipping the throttle before switching a hot engine off... Low gas pressure anyway during blipping throttle, maybe holding on over-run, then gas cools down and pressure drops further (lower pressure on starting from cold).

Blipping throttle have nothing to do with this
THERE'S A LEAK!!! Error occurred 251 times!!!!!!
High pressure error comes up over 3 bars I think
In my opinion reducer is faulty.....

LPGTech also gives this error when starting if the vehicle previously ran out of gas! May have a leak, but too early to say...

Never ever happens to me and probably anyone....

3. We don't know if reducer and gas temp sensor spec are setup correctly..

Even if you set it up incorrectly shouldn't read under min.

Might agree with you, but maybe the reducer is plumbed into the heater circuit after the heater solenoid block.

That's my suspicion.

4. The straight 6 engine only has 2 logical banks / 2 lambda probes. Cannot draw conclusions from the fact channels 2 & 6 are reading 10% longer duration on this screenshot - If there were a problem with fueling on 2&6, this would be reflected across the whole engine not just 2& 6, the petrol ECU isn't that advanced. This is just a screenshot... and firing order on the straight 6 is 153624 - note 6 and 2 are next to each other (so software might momentarily show these cylinders Pinj increase together due to throttle / ICV movement etc). Besides, LPGTech (and other ECU's) just sometimes do that kind of thing - false readings more an issue with the software than anything else...

You have a tendency to over thinking everything, what firing order have to do with this Are you trying to say, that with all injectors working correctly,(and right fitting) you can have inj. time differences when running on gas
Easy to check, just switch them off in the program and see what happens....

Setup in the **** LPGTech mode making diagnostics involve more guestimation

In what way

[LPGC]

Hardcorepit, I have a tendency for being correct mate!

Look at that 3D map - all those peaks and troughs... Would you expect such peaks and troughs necessary or proper for correct calibration? LPGTEch mode setup caused that mess and the installer didn't know any better.

OMVL SL's are capable of giving excellent results on the 3L X5 with correct setup. Wouldn't be my first choice but have converted many BMW 6's with OMVL injectors in the (distant) past. What would you fit?

You MAY be right on most of your points, way too early to say, but even if you are right - None of what you said would correct the OP's immediate problem... Your points only serve to confuse the OP. And let's face it - If the OP came to me with that vehicle, he would leave with it running perfectly without changing injectors etc. If he came to you, would you even manage to fix it (hmmm)? Wouold you charge him for fitting injectors too?

You don't seem to like anything negative said about LPGTech - Why not? It is only one system, and surely you are familiar with a great many systems?

Too much faith in auto calibration and error interpreting/reporting tech - LPGTech autocal works about as well as any other... that is badly.

So on your points again...

1. Chances are it is running too rich at high loads. But, like you say, the reducer might not be up to the job. If the reducer isn't up to the job for high loads, that's another issue on top of all the OP's reported problems. You list an injector that shouldn't be pulsing for 4.5ms if petrol injectors were doing 3.5 and I'll show you a system that is setup with too much pressure.... If you list LPGTech Dragons I will .. Only as linear as OMVL's. There are only a few exceptions here in terms of injectors.

2. Curing any leak wouldn't cure any of the OP's reported problems.

3. We should therefore advise the OP to try driving the car with the heater on to see if same problems still occur. However! LPGTech's Tech mode will have made a complete mess of calibration if the heater water suspicions are correct.

4. See 3 and also above bits on not being able to see multiplier or have usual thorough installer control over calibration. In LPGTech mode you are handing system calibration over to LPGTech software on a long term basis. Much like Stag's ISA2 / KME's 'poor fuel quality adaption'. None of them work properly in most cases, none of them anything like a decent substitute for proper calibration by a knowledgeable person, all of them aimed at 'done in a day installers' whom are far from perfectionists and amateurs who really don't know how to set nozzle size, pressure or calibrate systems properly.

I'm not even overly criticising LPGTech ECU's here - I've fitted a fair few myself and had good long term results. But I would never set one up in the gimmicky Tech Mode. Lost count of the times where I have shown and explained to customers, whom have come in for diagnostics/repairs on their LPGTech systems, where and how Tech mode all went wrong and led to their problems. Look at that 'corrections' chart again, the one labelled OBD.. The chart shows how multiplier has been changed according to fuel trim data. So what do you think will have happened if you/we are correct about reducer water flow being a problem and the the reducer was spluttering out liquid gas into the vapour line.. You reckon that could have effected fuel trims and that in turn affected Tech calibration - so when the system does run again with correct reducer temp the calibration might now be totally incorrect?

Simon

[hardcorepit80]

I have a tendency for being correct mate!

That's what you think and I bet all your customers must love you for all that technical talk you're probably flooding them with

Look at that 3D map - all those peaks and troughs... Would you expect such peaks and troughs necessary or proper for correct calibration? LPGTEch mode setup caused that mess and the installer didn't know any better.

That 3D map is actually PETROL map.... so how tech mode could caused that mess

None of what you said would correct the OP's immediate problem... Your points only serve to confuse the OP.

As far as I can see, installation is either fitted incorrectly or have faulty components and you trying to tell OP to ignore all errors and blame ECU for poor running...

You don't seem to like anything negative said about LPGTech

Just don't like when people say bollox about something they don't like/know.

You list an injector that shouldn't be pulsing for 4.5ms if petrol injectors were doing 3.5

Where on those screenshots injector 2&6 pulsing 4.5ms and if that's normal to you, well.....

Curing any leak wouldn't cure any of the OP's reported problems.

I can't believe that came from professional installer

In LPGTech mode you are handing system calibration over to LPGTech software on a long term basis.

ONLY if you're ADAPTATION is ON and it this case is on , which it shouldn't... because installation is a mess.

PAWEABMW: sort your components first, then set it up as it should be.

[Gilbertd]

While you two are bickering about it, I've been looking at the screenshots. Now I will admit that I have very little experience with multipoint systems and none at all with the LPGTech system but just look at the figures.

Why is the petrol injection pulse length on cylinders 2 and 6 consistently higher than on the other 4?

Why are the petrol injection pulse lengths going down as the revs go up (No 1 at 700 rpm, 3.2mS, at 1700 rpm, 2.9 mS and at 3520, 3.07 mS)? Surely there should be more variation than that and I would have expected them to go up with the revs (more revs = more air so should require more fuel to achieve stoichiometry)?

It was my understanding that the multiplication factor should be between 1.2 and 1.5 with the closer to the lower end the better, these all seem to be around 1.4 so that might mean it needs a bit more gas pressure or bigger nozzles.

I'll agree that the reducer temperature does seem very low but were those screenshots taken just after it had switched so is still warming up?

But the one thing that still stands out as far as I am concerned is that lambda 1 is giving no output. It's doing an impersonation of Monty Pythons parrot, it's dead, defunct, deceased and needs to be changed. It's probably that more than anything else that is bringing the MIL on.

[hardcorepit80]

ut the one thing that still stands out as far as I am concerned is that lambda 1 is giving no output. It's doing an impersonation of Monty Pythons parrot, it's dead, defunct, deceased and needs to be changed. It's probably that more than anything else that is bringing the MIL on.

Agree to that, that car doesn't run right on petrol.

[LPGC]

Harcdorepit,

I explain things in simple terms to customers and yes they are always pleased when I invariably sort their problems out - problems created by installers that don't know what they are dong. Seems it may have read the petrol map incorrectly.. There wouldn't really be such peaks and troughs on a petrol map. I will repeat here - there are only a few ways to correctly map an LPG system and 'automatic mode' doesn't come into any of them. No automatic mode is advanced enough to know the engine oil temp, whether gas temp is actual or a false reading due to picking up under-hood temps, etc. That LPGTech Tech mode bases mapping on that 'collected' petrol map should tell you a few things.

No good just reading stuff off a screen and repeating the same message, you've got to read between the lines. Very likely everyone/anyone reading this thread will notice the error messages, I certainly did... but like I said, they might not amount to much. It might be telling that on one of the few occasions you come across with 'authoratative' comments is when you get chance to repeat error messages, much of the rest of the time it's you asking the questions.

Bolx about stiff I don't know? I have fixed dozens of LPGTech systems installed elsewhere (including for some people that found me on this forum), have installed quite a few myself (no fixing needed). So who are you... amateur or pro.. how many have you installed? (Wonder if I've fixed any of yours?).

Curing any leak wouldn't cure any of the OP's reported problems.

I can't believe that came from professional installer

Like I said Pits, curing any leak won't sort his problems.

In LPGTech mode you are handing system calibration over to LPGTech software on a long term basis.

ONLY if you're ADAPTATION is ON and it this case is on , which it shouldn't... because installation is a mess.

Might agree adaption shouldn't be on... However, the whole point of Tech mode is adaption of some sorts at least... If not long term, at least it is there to help amateurs calibrate the system in the first place (map adaption).. And like I said, it makes a mess of things.. It makes a mess of things long term, it makes a mess of things short term. Long term adaption particularly makes a mess of things if anything goes wrong with the engine... If adaption is left on, then if there is a problem with the engine the adaption continues to. err, try to adapt.. throwing the mapping off. So then, when the owner sorts the problem with the engine, it then won't run correctly on LPG because the mapping has changed....

PAWEABMW: sort your components first, then set it up as it should be.

Best thing you've said. Not much need to say it though, already said by Gilbert and myself

So who are you - a so far un-named pro (favourite system LPGTech eh - would probably have fixed some of your stuff in that case) or someone who has done a few DIY installs, maybe completing one of those 'day courses' on one particular system (that are so far below my level of competency) and now supposedly knows it all (about that one system)?

It was my understanding that the multiplication factor should be between 1.2 and 1.5 with the closer to the lower end the better, these all seem to be around 1.4 so that might mean it needs a bit more gas pressure or bigger nozzlese than anything else that is bringing the MIL on.

Does depend on a lot of things though mate. Say you had a full group engine that idles at sub 2ms and some poor (say Rail) injectors were fitted (would probably give good results if calibrated properly on an old group engine), then a multipler of 1.2 at idle wouldn't get the injectors above the minimum (say 4ms) pulse duration and engine probably wouldn't even idle (I can picture the surging now!)... At highest loads, in most cases should aim for a multiplier of around 1 to avoid overstepping injection duration window (if reducer pressure falls off at high load can aim for a sub 1 multiplier). Multiplier is programmable across the load range, changeable against Pinj on most systems. Most injectors need a falling ramp from 0 to around 2ms, a steep rise to a peak at around 4ms (the 1.2 to 1.4 you mentioned), a less steep fall to around 6 and them a gradual flattening out to around 16ms (between around 12 and 16ms should be almost flat but falling very gradually) assuming 16ms peak Pinj... So you generally end up with kind of like a heartbeat monitor type map shape (we've all seen em at least on the telly!). For the full group engine that idles at 2ms and peaks at 8ms the map multiplier shape can be very different, likely much steeper, for Valvetronics can also be very different (less change across pinj range). Where all 'automatic' mapping systems go wrong in the first place is in getting the shape of this map wrong - and it would be so simple to improve upon, (can even make an Excel spreadsheet get it just about correct) if only you know the true spec of injectors selected.

I bet I could make an Excel spreadsheet for a given injector, given idle Pinj, given idle Ginj), given idle multiplier, given maximum (if any) boost pressure, given idle Map, given peak Pinj, would tell you the multiplier at each and every point between over-run and peak load for the map, much more accurately than any of these gimmicks.

Simon

[hardcorepit80]

For someone who watching this forum it's well known, that discussing with you is like talking to a brick wall, no matter what anyone says you've got always 100 things against, thinking that you're one and only true pro in England who fitted 1000's of any known lpg sytems....

[Gilbertd]

Simon, rather than arguing, pontificating and trying to score points against someone else who, if you had ever bothered to look at his posts, also knows his stuff, why not answer the OPs questions?

There's things there that don't look right. They wouldn't look right if the car didn't even have an LPG system fitted and we were trying to diagnose a problem on petrol. One dead lambda sensor and injection times the same at 3,000 rpm and at idle ain't right. That's what needs to be concentrated on not who knows more about what.

[LPGC]

ignore this post... some sort of error occurred.

[LPGC]

For someone who watching this forum it's well known, that discussing with you is like talking to a brick wall, no matter what anyone says you've got always 100 things against, thinking that you're one and only true pro in England who fitted 1000's of any known lpg sytems....

I know what I know and will say what I know, which is vast compared to a lot of people's knowledge. This subject is totally my bread and butter, has been for many year, and I certainly know it very very well. I will stick my neck out when others seem to be incorrect, maybe inexperienced and/or talking bolx. I have fitted thousands of LPG systems of many different makes / types. These days, I am not quite so busy converting vehicles but am as busy as ever due to the repairs I do. In recent times I have sorted a dis-proportionate number of LPGTech systems that were 'setup' in Tech mode. Even on this forum there are several users who have brought their vehicles with LPGTech system to me for putting right - I have put all such installs right.

Simon, rather than arguing, pontificating and trying to score points against someone else who, if you had ever bothered to look at his posts, also knows his stuff, why not answer the OPs questions?

There's things there that don't look right. They wouldn't look right if the car didn't even have an LPG system fitted and we were trying to diagnose a problem on petrol. One dead lambda sensor and injection times the same at 3,000 rpm and at idle ain't right. That's what needs to be concentrated on not who knows more about what.

I did look at his posts mate, having forgotten who he was completely, had forgotten that we had argued in the past. But I didn't get the same impression as you regards 'he knows his stuff'. The impression I got originally, and still have, is that he has read the LPGTech manual completely from back to front several times, has great understanding of that, but takes everything in that manual as gospel with little practical experience. Undoubtedly a clever bloke, but inexperienced with LPG and placing too much faith in autotune systems. He and I got off on the wrong foot... might be my fault, but I'm not going to let it go without saying anything if someone mentions that this or that autotune system works great when at around that time I am re-calibrating the mentioned system on a regular basis due to said autotune system getting it very wrong and causing problems. In this case it will now have caused a problem!

In my first post I concurred that the lambda sensor should be changed as it has a heater circuit problem, but I also explained that this might not completely fix the OP's issues... Not least, because the LPGTech system Tech mode adaption system has been left on and under these conditions it will make a mess of things. It has been suggested that adaption shouldn't be left on this long and I agree whole-heartedly. So then, how long should the adaption be left active? My suggestion would be at least until the multiplier looked correct across the full load range, pity you cannot see the multiplier in Tech mode, but then people who use self tune / adaption type modes probably wouldn't know what that shape should be anyway... Much better to tune the system manually, where you can see the multiplier, where you know you got the shape of the map correct, where you know that if there are any engine issues the LPG mapping will not be affected so the engine will still run properly on LPG after the issue is corrected.

Been too busy to do much on the forum lately, got all my LPG work to do plus swapping the autobox on the GV... Following the incident pulling the caravan (mentioned previously), gearbox now free-wheels when trying to select 3rd or 4th. I am swapping to the gearbox from the GV I am scrapping (since I robbed it's ECU, radiator, and will be robbing it's rack shortly, might as well scrap it after that..). The autobox I am swapping in is one which I put together, which took far too much of my time... new bits in the box, diff from another gearbox. Pita of a job to change the box, though. Hence I have not been a**** to write this reply until now, but far from because my points were incorrect or not valid, more because I have had better things to do. In scruffy's now at half past midnight.

Simon