Page 1 of 1

Supercharged Rangerover fixed by LPGC

Posted: Sun Jun 17, 2018 4:57 pm
Fixed a Supercharged Rangerover the other day.
Before I saw the car the owner told me it had an occasional misfire when running on either fuel, tended to run better on LPG than on petrol, usually had 'system too lean bank 1' and 'multiple cylinder misfire' OBD errors.

When it arrived it was only running on 7 cylinders on petrol and wouldn't run on LPG at all, the owner said this situation had developed only very recently. The system fitted was Romano but not the usual AEB2568 ECU'd Romano system, this was the Romano Antonio ECU'd system.

The first thing I did was look into why it would only run on 7 cylinders on petrol. I eventually found the problem, broken original vehicle wiring to cylinder 1 petrol injector. This took a while to find, after I'd found it I ran new wires from the petrol injector to the LPG ECU petrol injector break wiring and petrol injector positive wiring. This cured the misfire on petrol.

Then I looked into why it wouldn't run on LPG, found the problem to be a broken tank solenoid wire above the petrol tank where the installer had cut into the solenoid wiring to connect the petrol fuel return solenoid. Fixed that problem and then it would run on LPG again, but not properly.

The installer had made the mistake of using a common vac pipe for both the vacuum operated lube system and LPG system (reducer vac connection and pressure sensor connection) and the mistake of connecting this common vac pipe to the vacuum side of the supercharger rather than to the plenum side. I could already see in software that the LPG map was wrong, wrong in a way (rising map numbers toward higher engine load conditions) which pointed to the vac pipe being connected to the wrong side of the supercharger... this in itself would mean the LPG system would have no chance of fuelling the engine properly under boost conditions because A. with the reducer set at 1.5 bar above reference pressure there would only be 1 bar of relative pressure at best under boost conditions instead of the necessary 1.5 bar (1.5 bar necessary with spec of injectors fitted). B. It would be impossible to map correctly with actual reducer relative pressure following pre-supercharger pressure. I fitted the vacuum connection to the engine in the correct place and rerouted the lube pipe to avoid lube running through the LPG system pressure sensor. Being supercharged it should really have an electronic forced lube system fitted anyway, but the type of lube system fitted was beyond the scope of what I was tasked to do so I only advised the owner about the lube situation.

I cleared all OBD codes and went for a drive on petrol to allow fuel trims to be re-learned. Because of the misfire on petrol and previous incorrect LPG setup/calibration this took some time and involved constantly clearing system lean/rich codes for both banks of cylinders, this is because the supercharged rangerovers have underlying fuel trims that take some time to be learned and won't be learned while-ever there are any OBD codes present (even pending), so getting the car's fuel trims something like normal takes some time as the petrol ECU has a great many fuel trim ranges. Even after this I noticed that bank 2 trims were a bit high positive at idle.. a possible MAF or bank2 lambda sensor problem but not too bad that it would necessarily be a problem.

Tried calibrating the LPG system with existing components (Romano Antonio ECU, Romano HD reducer at 1.5 bar, Romano's singular injectors) now properly plumbed and wired in (by me) but it still wasn't having it. For starters I noticed that during changeover (one cylinder at a time) the engine faltered a couple of times. I checked petrol injector breaks matched LPG injector outputs for each cylinder and found that on bank 2 the original installer had made the mistake of getting petrol injector wiring for cylinders 2 (front passenger side cylinder) and 6 (third cylinder back passenger side) but had still connected LPG injectors as though they hadn't made the petrol injector break wiring mistake. Easily corrected this by swapping LPG injector plugs for cylinders 2 and 6. Instead of bank 2 seeming a bit lean on petrol bank 1 now seemed a bit lean when running on LPG but the Antonio ECU seems a bit basic and doesn't allow setting a bank trim... I might have been able to use a bank trim setting to get around the problem of fuelling being rich/lean on opposite banks to petrol trims seeming rich/lean. But bank trim issues at low engine loads/idle can also point to other problems such as dodgy injectors or injectors that are struggling to meter fuel properly at short pulse duration and the supercharged Rangerover engine has a wide dynamic operating range (because being supercharged it idles with pinj of under 2.5ms but at high load pinj can increase above 16ms). Summing up this paragraph so far, to ensure good results this engine really needs injectors that can cope with short pulse durations perhaps better than the injectors that were fitted even if the injectors fitted were in perfect working order but regardless of perfect order or not it remained there was a bank trim issue that I could not dial out using a none-existent LPG ECU bank trim facility. Either the injectors had to be swapped or the ECU had to be swapped, but preferably both.

I gathered up my notes / findings and contacted the owner to ask how he'd like me to proceed. He'd originally contacted me by email and had said he'd like me to remove the full system and fit whatever I advised in order to get the vehicle running properly on both fuels. In my original reply I'd advised I'd probably get good results with his existing components if I re-fitted them properly but I was expecting the ECU to be an AEB2568 and now with everything I'd seen it was seeming we might have to go with his original idea of changing components anyway, ECU and injectors. The owner now told me to do whatever I thought so I left the original Romano HD reducer in place but removed the Romano injectors to fit MJ injectors and removed the Anotnio ECU to fit a KME Nevo ECU. Now fuel trims are just the same on LPG as they are on petrol, although to avoid sailing close to the wind regards positive fuel trims under idle conditions on bank 2 I set the Nevo ECU to load 0.2ms of injector pulse time onto fuelling for bank 2. The vehicle now runs smoothly with great throttle response at any engine load on either fuel. The original installer made a better effort than I've seen on some LPG converted supercharged RangeRovers, close but were never going to win any cigars! They'd made a few LPG schoolboy errors during the install but had already set off on a bad footing with choice of components... although their choice of components was at least more suitable than on many supercharged Rangerovers I sort out. No idea how the broken wire to petrol injector 1 had happened or when it had happened, the owner said he'd had a garage try to find the cause of the misfire on petrol on cylinder 1 and he and his mate had spent all day trying to find the problem. With most other LPG system ECUs fitted the petrol ECU would have reported an 'open injector circuit cylinder 1' OBD code due to the broken petrol injector wire but the Antonio ECU applies petrol injector emulation even when running on petrol, hence the petrol ECU did not report the open circuit to petrol injector 1... this will have made identifying the problem more difficult for most garages or owners (an 'open injector circuit' error describes such problem pretty well!).

Re: Supercharged Rangerover fixed by LPGC

Posted: Mon Jun 18, 2018 7:47 am
by lawler999
Bloody hell Simon you are good :D
Please , will you have a look at my mazda cx7 :wink:

Re: Supercharged Rangerover fixed by LPGC

Posted: Tue Jun 19, 2018 11:06 am
lawler999 wrote:Bloody hell Simon you are good :D
Please , will you have a look at my mazda cx7 :wink:
I'll have a look but it's a matter of when. I'm usually converting a couple of vehicles to LPG per week and often don't know until (say) Thursday if I'll get any time on (say) Friday so have been telling people who own vehicles I didn't convert that I can only 'pencil in' bookings for repairs (servicing vehicles I originally converted also takes priority over repairing vehicles I didn't convert). People know my reputation for always correctly identifying and being able to fix LPG problems and are willing to 'play it by ear' with me rather than visit somewhere else only for some other guy to go through the motions, not correctly identify problems, not know how to fix problems but charge more than me anyway.

As you know, your Mazda CX7 is direct injection and is fitted with an AEB based LPG vapour injection system. I've converted CX7's to LPG myself using AEB based vapour injection systems but your brand of AEB DI system isn't the usual Emer or OMVL and suppliers of your brand system are less forthcoming with updated firmware / interfaces / software etc than (say) TinleyTech who deal with the Emer and OMVL AEB DI systems. With any type of DI vapour injection system (AEB or otherwise) the installer doesn't have as much control over fuelling as with most port injection systems because the interpretation of petrol DI pulses (what we might usually term pinj but on DI systems cannot directly measure pinj as pinj is not a single pulse so interpretation has to add up multiple DI pulses to arrive at a single figure for pinj minus any petrol that is still allowed to flow through petrol DI injectors to keep them cool) is handled by firmware and we can't change how the firmware does this interpretation or how firmware allows a proportion of petrol still to flow through DI injectors. I don't know if the firmware of your unusual brand (for a DI system) does the interpretation differently to the brands I've fitted on CX7's, and/or if that is part of the problem (would expect the main part of interpretation to be the same but don't know if aspects are changed to suit your brand LPG injectors)... And/or if you have some of the more usual type problems that are generally associated with the vapour injection aspect such as pinj's that are too short for the spec of your LPG injectors and/or mapping and/or nozzle placement/plumbing. Only when we get to the general vapour injection aspects on that list do we get to points we could easily/directly address but I do expect some of those aspects to be a large portion of the problems with your setup. A port injection system fitted on a DI engine will always have a few compromises / fudges. I know you've visited other installers with the CX7 who haven't managed to sort it... I expect to do a lot better, not because I have any way of making the compromises / fudges better (because they're inherent with fitting a port vapour injection system onto a liquid direct injection engine) or because I can do anything about firmware (I can't) but because I fully understand the vapour injection aspects.

Re: Supercharged Rangerover fixed by LPGC

Posted: Wed Jun 20, 2018 1:36 pm
by 4eyes
Mmmmm. Driving around in a supercharged range rover waiting for fuel trims to settle sounds like a real bind.....(wink). Another impressive and informative write up, cheers!

Re: Supercharged Rangerover fixed by LPGC

Posted: Wed Jun 20, 2018 5:14 pm
4eyes wrote:Mmmmm. Driving around in a supercharged range rover waiting for fuel trims to settle sounds like a real bind.....(wink). Another impressive and informative write up, cheers!
Thanks Dave.
Supercharged Rangerovers are relatively slow compared to many vehicles I convert to LPG and have to calibrate... But the novelty of driving fast cars for diagnostic / calibration purposes wore thin on me a long time ago! These days I'd rather have to drive a 1.6 hatchback flat out than drive something with best part of a thousand horsepower (which wouldn't be a supercharged Rangerover) flat out.
Should point out that 'flat out' doesn't necessarily mean maximum speed, it means maximum engine power, can get this situation during acceleration without having to drive too fast. Having said that I've driven plenty vehicles that want to spin the driven wheels if the throttle is slowly fed in from an already high speed, pressing the accelerator gently toward the floor from a lower speed can increase the likelihood of wheelspin. Not so bad in a 4wd supercharged Rangerover because that doesn't happen.