Did a write up on it on LPGforum here's the link

That makes interesting reading.

I was never brave enough to convert my Supercharged L322 to LPG through horror stories of the valve seats burning. Probably just as well, it was never my daily car and only did 10k in the 3 years I had it :)

David.

Fixed the LPG system on 3 Rangerovers in the last couple of days, plus on a Jeep Cherokee.

Including:

An L322 with BMW 4.4, recently had engine fitted by a garage in Barnsley (and I visited the garage to fix it). Stag300 with Zavoli reducer and V30 injectors. Mechanic who fitted the engine had connected the LPG system's vacuum connection to the front diff breather pipe (hehe), mixed up wiring for the reducer solenoid and reducer temp sensor, then when I got it so it would run on LPG it was running way too lean so I needed to calibrate it. It was obvious to me that the vehicle had previously had a BRC system fitted but the only parts of the BRC system still remaining were the multivalve and wiring to the tank (still had superseal plugs on, which along with the Zavoli's temp sensor plug also being Supaseal explained the wiring mix-up).

Sport 4.2 Supercharged with BRC system, reducer solenoid clicked but tank solenoid didn't. Tank solenoid reading 10Ohms and wiring 12v to it saw it open so it wasn't the coil at fault. 16 Ohms across the wiring to the tank solenoid even with the reducer solenoid disconnected so this was the fuel return solenoid not the reducer solenoid and the break would be between the front end and rear end (fuel return solenoid fitted at the rear), found a break in the wiring under the drivers door.

P38 Misfiring on LPG and incorrect mixture on petrol, needed a lambda sensor but also new LPG injectors, he's going away to fit a lambda probe and will return for the injectors.

Jeep Wrangler 4L with Aldesa system (shite and rare system), supposedly came in for a check-over for a new buyer but was delivered on a trailer with seized osf caliper. Tank rather rusty so they daren't put any gas in it before I'd checked it. No number plates. Broken reducer temp sensor (fixed) and seized tank solenoid (fixed). It would then run on the vapours of gas left in the tank so I disconnected the tank's fill pipe and stuck around 5L of gas in the tank from a tank I have in the yard... Tank held pressure OK and I reckoned although the tank was rusty it wasn't dropping to bits and hadn't suffered severely thinned walls. Fuel trims OKish so I told them it was fixed but would need the rust cleaning off the tank and the tank given a thorough underseal.

Strange how writing P38 followed by a capital M I nearly misread my own writing, expected to read P38 Mustang hehe.

Sounds like you're being kept busy, Simon!

Mine is running great on LPG at the moment but more often then not I still get the EML light on for high fuel pressure. Have had a good look around it but can't find any signs of an emulator or fuel return. Other then changing the reducers I've done no work on the LPG so I'm wondering if the previous owner just put up with the EML light being on all the time.

It's annoying me though and I'd like it putting right though.

One thing I have noticed is very occasionally it makes a loud noise under load, almost like a fog horn kinda noise but fainter. If you let the car go into overrun it stops, and if you switch to petrol it stops too.

David.

Simon will probably confirm but some LPG reducers do that if certain conditions are met. It's the diaphragm hitting resonance.

Makes sense as it's from the left hand side of the car where the reducers are mounted. Does it even happen on brand new ones?

As yours is doing it with brand new reducers, apparently it does.

It's sometimes possible to dial out resonance by adjusting reducer pressure setting slightly

Simon, my jeep is having a new HG fitted in Elland right now... I might have to see if I can swing by on the return for a checkup/service. The lights are doing that disco thing that either means a bad sender unit or it's due a service. I'm not sure which it is on the current system!

Ok Miles, let me know when you're ready and we'll sort something out :-)

Disco thing meaning it will run on LPG but the level sensor lights are flashing?

yep, it runs fine but sometimes it flashes the lights :)

Only just seen your last post Miles. It'll flash lights on the LPG switch when it's switched back to petrol because it's worked out gas injectors would need to pulse for longer than the available window between 4 stroke cycles for calculated pulse length... Usually due to dodgy calibration, too small nozzles, too unlinear injectors or too low pressure. In your case, since I fitted it and injectors will be linear enough, nozzle size will be correct and calibration will be correct it will be due to too low pressure... so the ECU is compensating for low pressure by increasing gas injector pulse duration. so calculated pulse length has become longer than window.

Let me know when you're ready and we'll sort something out if you like? Got loads on lately but we'll come up with something.

Been fixing an L322 today, expected an easy fix but it's been a pita lol. I'll post about the L322 on a more relevant thread. Edit.... Hang on, this is the more relevant thread, I'll post below lol.

Fixed a few LPG installs on various vehicles today, most of the owners have been badgering me to bring their vehicles in for about a month but I've had loads of conversions on.

One of them was an L322 with complaints of it making a whining noise during driving, when the driver puts foot down the MIL light flashes and the engine shudders/vibrates. The owner didn't know what system was fitted, had the system converted by an installer in the NorthWest 4 years ago and it had always had problems running properly on gas but lately they had got worse.

I expected an easy fix, whining probably the reducer resonating, MIL flashing and engine vibration obviously a misfire, probably both related and the reducer kicking out incorrect pressure.

When it came in the install looked like it the installer had started out with good intentions... neatly wrapped wires over the engine and injectors on an OK bracket even though their choice of injectors meant the engine cover couldn't be refitted (and a bad choice of injectors anyway being Barracuda) then got pushed for time and decided to bodge other parts of the install... dodgy looking wiring near the +ve jump start pole that prevented anyone removing the heater air filter bulkhead panel for access to the rear end of the engine.

Although I'd seen the same type of changeover switch many times in the past I didn't remember the make of ECU/system that used that switch so had to open the engine bay electronics box to see the ECU, turned out it was a 'Blue ECU', seldom seen in the UK though I had worked on systems based on that ECU in the past and already had the software for it, one of the better hardly known ECU's. The Barracuda's look the part, they're all shiny and bigger than BRC or Keihin, they can be high flow, but they're slow in response and very over-rated.

Connected a generic scan tool, Ltft's at around +10% but stuck there due to OBD error codes being present despite Stft's being maxed out and the engine struggling to run gasping for breath with high manifold pressure due to lean mixture on both petrol and LPG, but worse on LPG.

Found out the installer or some other LPG fitter had wired injectors for cylinders on bank 2 (cyls 5678) in the order 8765, so during changeover and in certain other conditions there was hesitation and misfires due to the ECU channel routing problem, this bit an easy fix.

Did a bit of basic recalibration in the yard just occasionally using a bit of engine load wasted by heating up the torque convertor (holding on the brakes). The installer had calibrated using the wrong process, had dialled in all mapping corrections in RPM correction instead of using the map and had used too many offsets in the initial settings (such as 1ms added ginj for undefined 'idle' conditions). I reset all the anti-intuitive rpm corrections they'd dialled in and started my multiplier map calibration without the unspecified offsets deleted. I wanted to increase Ltfts's in the hope it would run better on petrol at low loads and then I could match gas trims to petrol trims.... and for the most part managed to do that, though I still expect it either has an inlet manifold vacuum leak or a dodgy MAF. I haven't smoke tested it yet but the owner's OK for me to have the car a bit longer now and I've had a cancellation for Monday so I might smoke test it then, though I do expect a dodgy MAF. Special tool needed to remove the MAF, like a Torx bit but with only 4 sides like a posidrive/Philips screwdriver, but with a centre pole so the tool needs a hole in the end of it's centre.

Managed to steer Ltfs up to the max at around 20% at light loads getting lower at high loads, Mil light would always be on with such fuel Ltft's but at least it might run OK(ish) with correct mixture, so took it out for a drive... Oh dear, the gearbox refused to upshift for TPS or vacuum and would only upshift for rpm. And the noise the owner talked about that I expected to be reducer resonance turned out to be not from the reducer but from the gearbox...

Back to base, connected up wider ranging OBD gear, the gearbox ECU had a stored code for lack of canbus coms. Cleared the error code, back on the road again.... this time it went up through the gears a lot more like it should and the noise had gone and the OBD code from the gearbox ECU didn't re-occur.

Felt a bit warm and the climate control was set at 22C so I turned it down (a test I would have run anyway...), found that running on gas without the cabin being heated the reducer temp started dropping sharply. Back at base I found the original installer had plumbed the reducer incorrectly, without under-bonnet heater water flow solenoids being open the reducer would see only a small amount of hot water but one of the heater matrix's would see a small supply of hot water even when it shouldn't. Removed the middle bulkhead / air filter panel to see piping at the rear of the engine. The pipe I needed to T to had degraded and swollen... should I try to plumb to it or not? I did, then soon regretted it, the heater coolant pipe had become very thin and swollen, try to plumb a T into it and any clip on the pipe just tore through the pipe. Took me ages to replace most of the length of that pipe (including all of the area that had swollen / thin-walled).

At least now it drives well with maxed out positive low load Ltft's, LPG wiring routing and mapping sorted, and will run OK on LPG without the heater blowing hot air. Just need to look into possible vacuum leak or MAF problem, could even be due in part to evap purge because it's got a dodgy petrol filler cap fitted that stinks of petrol. Also need to look into the fast idle (1000rpm) with correct mixture, these should do around 630rpm idle and it does with lean mixture so maybe the IAV has incorrectly learned with incorrect mixture (which I can sort).

If the worst came to worst... couldn't find a vac or evap leak and a replacement Maf didn't correct it, I am confident I could approach from the opposite direction and adjust the Maf signal voltage using a pull up resistor connected to a voltage divider circuit to tailer the low load Maf reading to something that will give good results... Just need to remember whether the Bosch MAF on L322 BMW's has PWM or voltage output...or does anyone reading know? At 1000rpm fast idle with fully warm engine with no load the MAF reading in live data is around 5g/s, I reckon it should be around 6g/s... Anyone have a BMW engine'd L322 to compare readings?

Lpgc wrote:

Anyone have a BMW engine'd L322 to compare readings?

No but I've got RAVE that covers it:

Mass Air Flow/Air Intake Temperature (MAF/IAT) Sensor
The MAF/IAT sensor is located in the air intake ducting, between the air cleaner and the throttle body. The sensor outputs intake air flow and temperature signals to the ECM to enable calculation of the mass of the air entering the engine.
In addition to the air flow and temperature outputs, a regulated 5 V feed and an earth are connected between the sensor and the ECM, and the sensor receives a battery power feed from the main relay.
Air flow:
The air flow signal is produced from a hot film element in the sensor. The film is connected between the 5 V feed and the air flow output to the ECM. The film is also heated by the battery power feed and cooled by the air flow into the engine. The greater the air flow, the greater the cooling effect and the lower the electrical resistance across the sensor. So the air flow output voltage varies with changes in air flow and, from voltage/air flow maps stored in memory, the ECM determines the mass of air entering the engine.
Air intake temperature:
The air intake temperature signal is produced by a NTC thermistor connected between the 5V feed and earth to complete a voltage divider circuit. The ECM monitors the voltage drop across the thermistor and, from voltage/temperature maps stored in memory, determines the temperature of the intake air.

Can't find any data of what the output from the MAF should read though. The same section on the P38 gives figures for idle and at 3,000 rpm but the L322 manual doesn't seem to show that.

Gilbertd wrote:

Lpgc wrote:

Anyone have a BMW engine'd L322 to compare readings?

No but I've got RAVE that covers it:

Thanks Gilbert, that does imply it's just an analogue output as would normally expect. I'm sure I've got notes somewhere on normal Maf g/s readings at idle on this model but can't find them... With a vac or evap purge leak not all intake air would be coming through the Maf, which could explain low Maf readings and high fuel trims. Or if the Maf were dodgy would also explain low Maf readings.

I know there are a lot of dodgy aftermarket Mafs around for BMW engines in general, and on the BMW's a Maf that looks identical can be a different spec... So if I find the problem is the Maf do I advise buying a replacement from a dealer (Landrover / BMW) or a much cheaper one from Ebay etc, or fit signal voltage pullup resistors...

Interesting post on the Supercharged, Simon. I had my Supercharged smoke tested recently due to a very small lumpiness at idle when cold. There was nothing found on it. The JLR specialist who smoke tested it said that vacuum leaks on the 4.2 SC engine are very rare.

David.

dhallworth wrote:

Interesting post on the Supercharged, Simon. I had my Supercharged smoke tested recently due to a very small lumpiness at idle when cold. There was nothing found on it. The JLR specialist who smoke tested it said that vacuum leaks on the 4.2 SC engine are very rare.

David.

The Jag/Ford engines have fussy management systems that will usually pick up on even a minor vac leak, they can even tell if the evap purge flows the correct amount of air.

Still have the L322 from the other day here, I haven't found any vacuum leaks. Have tried pull up resistors from both 5v and 12v to the signal wire, the best compromise I can achieve using resistors is quite negative trims at warm idle no load jumping straight to positive trims at idle in D. First I tried the pull up from 5V and got the results just mentioned, then I thought a pull up from 12v would make for more trim consistency between idle no load and idle under load but it doesn't improve consistency much. I've removed the Maf and tried to clean it but still lean mixture highly positive trims, I've ordered a new Bosch Maf from Euro car parts. £161+Vat. I did find that with the Maf signal (g/s reading) corrected at idle the idle rpm speed corrected itself.. seems to be an aspect of the ECU to open the throttle a bit regardless of correct idle rpm speed if the Maf reading is too low.

I've had 2 conversions cancelled this week, one by a guy who I've since read in my notes has cancelled on 2 previous occasions, one by a guy who bought an imported car 2 months ago and was told by DVSA it'd take 6 weeks for him to have his V5... He was booked in for today, told me his docs still hadn't arrived so he'd have to cancel on Friday evening - could have told me your docs hadn't arrived a but sooner mate! But I've pulled another conversion forward and now have a customer coming from Scotland tomorrow, LPG conversion and towbar to fit.