[rtkraken]

Hey guys,
Recently reset adaptive values on my 95 P38 4.0 with RCL. The vehilce is new to me and am in the process of going through it. Initially found the long term adaptive value set point at -158. I reset the values and it immediately went to -160 and appears to be staying there... :think:

Engine runs fine, mileage calcs (if you trust them) by the on board computer look fine 16-18 mpg hwy, don't see anything gross and obvious from the input sensors. I did notice the smell of my catalytic converters when I parked it in the garage after an initial highway run (before resetting values).

On my 96, when I've reset values they always have gone to zero and slowly adapted from there. For some reason reset on this vehicle is starting at -160? I'm nervous to run the vehicle too much with the mapping compensation cranked to it's max.

Thanks for your help.

 

[rtkraken]

Clarification... Long term fuel trim is set and holding at -160 after an adaptive values reset.

My bad.

 

[rtkraken]

Update... after a few hours of highway driving the Long Term Fuel Trim is now at -8.3. Much better.

Not sure why the adaptive value reset caused the vehicle to adapt its way from -160 rather than 0, but at this point the results are what counts. Running nicely now. I think all it needed was a bottle of injector cleaner and a few hours at highway speed.

 

[6x9base13]

Thought I'd bump this rather than starting a new thread as I have the same issue, expect a few hours on the highway hasn't sorted mine out.

The history, my 97 GEMS was idling a bit rough and after plugs and leads did nothing and not having any obd tools at the time I left it at that. Eventually got some time to replace the coil pack, put in a new radiator and injectors and had the exhaust replaced. I now have a Nanocom so reset adaptives. Driven nearly a tank now and Nanocom always says short term is 0 and long term -160 and she is massively heavy on fuel !

TPS and MAF readings look like they are in range. IAC is a little tight, will get it in the 15 - 30 range tonight, but I suspect the long term should have changed by now.

One other thing that might help and I haven't found any obvious input on the net about it is that the on board computer display's range and consumption figures don't update either. They are at 999km and 0.0L respectively.

I'll trawl RAVE later tonight, but any advice would be appreciated.

Edit to add - no faults are being logged anywhere.

 

[Gordonjcp]

On anything other than a real genuine Testbook, the LTFT is utterly wack. As is the MAF reading. Weirdly, STFT seems to be about right with even a cheapy ELM327 knockoff.
(edit: hit enter too soon)
If fuel consumption is really heavy, check that the lambda sensors are reading something reasonable, switching up and down every second or so. If they're wrong they won't necessarily log a fault.

 

[Rowant]

Gordon - you are telling us people with faultmates that they are not accurate and can not be trusted with some readings? I wrote to BBS and asked why the LTFT readings did not reset back to zero when I did a reset and they said they were not an adaption - very strange.

Rowan

 

[Gilbertd]

Cars for the US are fully OBD2 compliant as it was mandatory from 1996 onwards but those for Europe, where it didn't become mandatory until 2000, aren't. I'm not sure of the cut off date for South Africa and NZ, but suspect it is the same as Europe. For some odd reason, although the GEMS controller is capable of being compliant, it isn't implemented. For example, as well as the oddball readings given by the MAF and LTFT, there is no misfire detection or MIL activation for most faults. With the engine running you can disconnect the TPS, IAT sensor, knock sensors, lambda sensors, etc and although it might store codes, it won't bring on the MIL. It would on a US car though where the failure of any sensor will bring on the MIL and the GEMS BIS states that it will but as this is published by Land Rover USA to comply with their freedom of information laws it is talking about a US spec version and not any others.

 

[RRToadHall]

THis is where being specific is very important so as not to confuse people or mislead them into false information. Making a blank statement that only Testbook can give accurate readings is point blank incorrect for some markets. It rates right up there with people that believe calibration blocks can only be used with genuine Testbook. Lets try be aware that what may fly in your region may not be accurate for all regions so as not to cloud people attempting to gain information.

 

[Gordonjcp]

As far as I can see, there is very little appreciable difference between UK and NA ROMs for GEMS - certainly not enough to cause the MAF and LTFT trims to work any differently.

From what I gather all the ECU has to do is spit out convincing-looking RPM, coolant temp and lambda readings to pass a US "inspection", which a UK GEMS ECU will do quite happily.

 

[Gilbertd]

I've no idea why but if you look in RAVE, section 19 (Fuel System) pages 20 to 22, it says that failure of the cam position sensor, MAF, IAT, Engine Coolant temperature and Engine Fuel temperature sensors will all bring on the MIL on North American spec vehicles when we know for a fact that you can disconnect any of these on a UK spec car and the MIL doesn't light, all that happens is a code is stored. Why they chose not to implement this on UK cars I've no idea as it would have increased the profits with panicky owners rushing to their nearest main dealer as soon as the light came on. Possibly they had to do it on US spec cars to meet the regs in some States. As the OBD2 spec was mandatory in the US from 1996, that is why you get silly lambda readings as although the GEMS uses 5-0V Titania sensors, the output is fudged so it gives a more common 0-1V reading which is what OBD2 requires.

 

[Gordonjcp]

It's probably just toggled off somewhere in there. I haven't got that far in yet.

 

[rtkraken]

Started this thread again recently on BBS board as it still seems to be effecting my vehicle. Another board member posted the same problem and his vehicle too was an NAS 95 model year. My 96 does no such trick. Resets to zero every time. The 95 resets to -160. I've not heard any good explanation for it. I've settled on -160 being 0 on the 95s. Can't explain my original post cause after full overhaul of fuel and ignition system and adaptive value reset the vehicle no longer smells rich, runs nice, and is registering ltvs around -155... Weird.

 

[6x9base13]

If fuel consumption is really heavy, check that the lambda sensors are reading something reasonable, switching up and down every second or so. If they're wrong they won't necessarily log a fault.

The rest of the discussion around differences and model years is slightly above my pay grade, but I can confirm that my South African model doesn't throw check engine lights when disconnecting the TPS ,IAT or MAF etc.

I can tell from driving that there is some adaptation happening. She's more responsive and whilst I'll need to get further into this tank to get a better feel for consumption, I can confirm she's not spitting black stuff out the exhaust with the condensation on first start up in the morning. It seems to be getting better. Nanocom reading stays -160 though.

I've set the Idle stepper into range so will see where to from here. Haven't had a chance to trawl RAVE at night, but will over the weekend. I have been monitoring the Nanocom and spotted a few other things so I might have more than one problem, one of which is an earth somewhere. The gearbox claims to be getting 10.5V only even though I'm not getting the usual check gearbox errors I suspect it should be fixed before trouble shooting further. Gearbox only sees 74% throttle when I have the peddle to the floor too.

One other thing to add, our SA Rangies run open loop - 00 configuration so no lambda sensors. Would that affect time to start adapting ?

 

[Gordonjcp]

No lambda sensors *at all*? I've no idea how it would set the fuel trims, then. It probably doesn't.

 

[RRToadHall]

If you have no sensors and run only in open loop wouldn;t you be on a static mapping and have no adaptive values?

 

[Gordonjcp]

If you have no sensors and run only in open loop wouldn;t you be on a static mapping and have no adaptive values?

In which case the LTFT would most certainly be out to lunch, yes.

 

[RRToadHall]

Kinda what I am thinking. Running in open loop your only real concern would be timing not fuel mixture.

 

[Gordonjcp]

So why would it be rich? Stuck regulator, bunged up return pipe, some deep sense of confusion within the ECU?

 

[RRToadHall]

or dirty injectors dripping, stuck injector, air flow not reaching minimum.. it's all out of my realm. I think the Australian setting runs a similar set up with no O2 sensors.

Maybe check over on AULRO for info?

 

[Gordonjcp]

Simplest thing is get rid of all that stupid stuff and convert it to gas!