This is been running on petrol so it shouldnt be a gas/injector issue.
I'm in Fife, Scotland.
Indeed, but it might have explained the trims being way out on LPG.
Fife's a bit far to lend a hand I'm afraid... wrong end of the Island :)
I'd do a compression test if you haven't already. Two adjacent pots with oily plugs could be a sign of a blow between the two. Oil could be valve stem seal issue but could also point to gummed oil control rings, especially if it hasn't had regular oil changes in the past. BUT, if it was burning oil I'd expect to see the HC figure go through the roof as that shows unburnt Hydrocarbons and there would be loads.
I'd agree with Sloth on the LPG injectors being crossed, as that would cause one bank to go one way and the other the other way. It's not particularly relevant at the moment but could explain the rough running on gas. Where are the injector spuds fitted, ideally they should be fitted next to the petrol injectors, then there's no chance of them being wrong, but if they are fitted higher up, just beneath the join with the plenum chamber, then the ports do cross over. It might also be an idea to clean the plugs and then try running it on gas to see if they foul up the same as on petrol.
aye fair enough, i'll have a close look at the LPG layout. Will take a pic of the injector points. Its been on there for over 10 years though. There was a note at the last LPG service saying one bank of injectors wasnt working properly and they'd need replaced "when the vehicle became undrivable" ffs.
I've been reading a bit about the TVR guys replacing their Lucas ERR722 injectors with some Bosch Gen2 200cc injectors from a Mustang, which are supposedly much better than the lucas ones. As it happens, i have 12 210cc Bosch Gen2 injectors in my shed from various Audi 1.8T engines i've stripped for parts. Think i'll give them a clean and swap them in, rule out dodgy injectors.
I will try to sort out a compression test. I think i have a tester somewhere!
So found my compression tester and some free time, went out to test it and found the battery completely flat. Doors wouldnt even unlock... Great.
Stuck it on the charger and instead tackled some other bits. Stripped the intake off and removed the injectors. Swapped in my spare Audi injectors and refitted the fuel rail. Also removed and inspected the ignition coil, and discovered that it looked somewhat melted. Almost like it had been on fire! rather odd! It seems to be working fine, but something i probably want to change at some point. Reinstalled the middle inlet manifold and figured the battery might have enough juice in it now as doing the injectors took feckin ages.
So got the compression tester on the go, i started at cylinder 8, as that was the oiliest and got these results:
8: 11 bar
6: 8.5 bar
4: 13 bar
2: 13 bar
1: 13 bar
3: 12 bar
5: 11.5 bar
7: 11.5 bar
The battery was getting a bit weak towards the end, so i suspect the last three cylinders would probably have managed closer to 13 had the battery been fully charged. However clearly something odd going on with 8 and 6, those two cylinders are the one that are fouling the plugs up with oilyming too.
Reassembled the inlet manifold, and gave it an oil change then managed to get the two lambdas swapped out for new ones. Not fired it up yet though.
Did you get a chance to chuck a shot of oil into the low pots to see if the pressure increased?
Nah the battery was really low by the end, and i didnt want to abuse it more than i already had. My chargers a bit puny and it had only been charging for a couple of hours from dead flat, so not really a huge amount in it. Leaving it on overnight to fully charge.
Will need to try it again with a shot of oil in the cylinder once the battery has some charge in it! Is it worthwhile doing the test on a hot engine?
Hot is best (if you can get it warmed up) as pistons and rings are expanded.
Makes it more fun juggling hot plugs etc as well :-)
yeh it'll warm up just fine, i just didnt finish putting it together. Need to fit the MAF/intake pipework and some other minor bits.
I'm going to try again to get an MOT on it now i've changed all those bits, as even if i'm going to have to pull the engine to bits, its FAR easier trying to diagnose things when you can actually drive the bloody thing on the road!
If you do a wet test and the compression on those two pots comes up, then you probably are looking at gummed rings, if it doesn't then you are looking at valves or my suspicion of a failing head gasket between the two pots. In either case, it sounds like the heads are going to need to come off at some point. But if you manage to convince your MoT tester to do his job properly and get it tested running on gas, it should go through with the readings you posted previously. At least then you can run it around and get a better idea of what the problem is. If the rings are gummed it might be that a bit of running with some synthetic oil in it could persuade them to un-gum themselves.
yeh, i guess i'm going to have one last go with this place and see if i can get it thru the test. Looking at whats been happening, the High CO figures from above were due to the MAF being disconnected. The last test got aborted due to missfiring, which i presume was due to the fouled plugs. I'm hoping now, with fresh plugs, and hopefully non-leaky injectors, it'll pass.
If it was consistently failing with the same results, i'd be a bit annoyed, but every time its gone in, something different has happened.
If it fails again, i guess i'll need to find somewhere else that will let me test it on LPG. Which reminds me i need to buy some LPG hose, as i found a few melted bits that needs sorting.
If he refuses to test on lpg, I'd report to the ministry, they have to test as you present, not what they feel like doing..
I know, hes being a bit of a cock, but hes also the most-decent MOT place around here. I've had a good few run-ins with some other places over much bigger issues (like failing the car for things that arent actually faulty...) and this guy seems fair and reliable. Last thing i want to do is cause issues for him.
And to be honest, its failing the test because its broken. Out of the four or five tests i've had it on, its only actually successfully completed one emissions test, and that was the test with the MAF unplugged that returned the 1.3% CO readings. The rest of the tests were stopped due to the car either dying, or chucking out so much unburnt fuel that the machine aborted the test.
In that state, LPG wouldnt have helped. I'm also unconvinced the LPG is actually working correctly, and would like it running properly on petrol before i start looking into the LPG issues.
Well its still broken. This time becuase i've cocked it up.
The injectors i swapped in i thaught were 210cc, which i figured was close enough to the required 200cc. Turns out, after looking up the actual specifications, they are 245cc. Which seems to be too large an error for the ECU to correct.
So, I found these:
https://www.fiveomotorsport.com/lucas-range-rover-gems-hotwire
Crossreffed the bosch part number here in the UK, and managed to find a set of 8, new, for less than the cost of getting the broken lucas ones cleaned.
More waiting.
Ok, so it wasnt my fault after all.
Fitted the new, correct, smaller injectors, and guess what? Still wont fucking run.
I've literally NO idea what is wrong with it.
Start the engine from cold, it runs like a total sack of crap, give it lots of throttle and keep it running at 2-3krpm (lots of manipulating the throttle) until the lambdas come online (you can see it switch on nanocom) and you see the short term trims climb from 0 to 38.75, and as it does so the engine begins to run better. Unfortunately it then jams itself at 38.75 and just sits there.
If i get the revs up to 3000rpm, then you will see the sensors begin to switch, with the short term trims cycling in the high 20's, drop the revs down to 2000rpm and it goes streight back to 38.75 and the more you lower the revs the worse the engine runs. Its next to impossible to keep it running at 1000rpm.
I really need to understand what these trims mean. In my mind, a positive trim is the ECU ADDING fuel, suggesting the engine is running very lean, and is having wind the fuelling right up to try and compensate. That simply doesnt make sense, unless this long term trim of -160 really is subtracting huge amounts of fuel. If thats the case i simply cant see how i'm supposed to fix this? I cant get the LTFT's to come back up to 0 unless the engine will run and drive. I cant drive it if it wont idle, especially as it has an automatic trans...
From memory it's a GEMS? Have you tried cleaning the IACV stepper motor? If it's gummed up it could be causing problems, likewise the TPS..
It us also worth trying to reset the adaptive values and see what that does to the running of it.. I think on GEMS they are reset on next power up after being told to do it. I know on Bosch it clears adaptive values straight away, even with the engine running...
TBH i think resetting the adaptives is what caused this whole issue.
It ran ok, but was fouling 3 of the plugs on one bank. I suspected leaking injectors.
I replaced the injectors with some Gen2 Bosch parts i had already, fitted new lambdas another set of new plugs and reset the adaptions, and after that it wouldnt run worth a shit.
I thaught it was because i'd fitted injectors that were too large, bought the correct ones, and its exactly the same.
Clearly, adaptions should be clearable without issues like this, but i guess they were maybe masking some sort of underlying issue.
I know you have tried cleaning the MAF, but was that one of the other things that you have actually tried swapping?
There isn't really that much else that could be causing it - unless as GillbertD mentions about a breach in the HG between 2 cylinders... If it runs like crap in closed loop, then the only parts in the closed loop system are really the MAF, IACV, and Lambdas. Ok, there are a bunch of other sensors too - and the injectors..
Ooohh, out of interest... what about the IAT temp sensor... I know that can be a bit contentious I think to get a reading from it in some diagnostics (or maybe it's just the Bosch ones as it's part of the MAF on the later ones) but if it's failed and it thinks the air is a different temperature to what it actually is, then that could possible cause overfuelling and bad running... Also could be worth looking at the TPS live values aswell if you haven't already..
Now it doesn't bother me too much as running on petrol is only as a real last resort, but Nano reports my STFT at 38.75 and it stays at that even after I've reset the adaptive values (although Marty's comment that they don't reset until next power up may explain that). It runs well enough to get me to the next LPG filling station but the lambda sensors stay at 5.07V suggesting a very lean mixture and the smell from the exhaust, and the way it runs, suggests very rich. Like I say, it doesn't bother me as the petrol is used very rarely but I have thought it might be nice to get it running as it should.
Marty: The original maf was broken, it crapped out when it got hot. i've replaced it with a used one from ebay. I also tried unplugging it. When the old MAF was broken, the engine ran fine with the MAF unplugged. Infact it ran BETTER with the MAF unplugged. With the new MAF on, it ran fine, until i did the above work and reset the adaptions. Now, it wont even run properly with the MAF unplugged, as i tried that earlier. The MAF values reported by nanocom seem feasible, so i'm struggling to see it as a MAF issue.
The IAT values are odd, showing -3 i think on nanocom, but even with that, a low IAT should be enriching, and i'm pretty certain this is lean.
Gilbertd: when its running in proper closed loop, you can see the STFT's cycling up and down by a few % and the voltage swinging up and down. Lambdas act a bit like a switch, So the ECU slowly enriches the mixture until the lambda switches rich, and then slowly leans off until it switches lean. If its stuck at 38.75, it means its run out of adjustment and still hasnt managed to get the lambdas to switch. My lambda voltage stays low when the trims are stuck at 38.75. When i rev the engine up to 3200ish and it starts switching, it starts cycling between 1 and 5ish and the STFT's drop down into the high 20's. So whatever is going on, its running out of adjustment at lower enginespeeds, and cant compensate by enough to keep it running.
Your saying you get 5v and a rich smell. but your reporting +38.75? I'm getting a low voltage (it fluctuates a bit, usually 0.5-1ish) and no rich smell and +38.75. They cant both be true? I would expect -38.75 if it was rich?
Current thoughts are fuel pressure. Ordered a test guage. Really clutching at straws though. I've fitted new lambdas, so i can trust the readings, and they are both doing the same thing, which suggests whatever the fault is, its something global, rather than per-bank. I was considering pulling the wideband lambda out of my track car and grafting it in, but i really cant face trying to fit an sensor bung into a pipe under the car.