RangeRovers.net Forum banner

1 - 20 of 21 Posts

·
Registered
Joined
·
48 Posts
Discussion Starter #1
Has anyone experienced a long term trim reading of -100% (less fuel) i.e. over fuelling. ECU was reset and it always comes back to this.

Short term readings fluctuate ok, lamba probes are new ntk/ngk not cheap ones - MAF is new (lucas overhauled NOS), injectors are overhauled, fuel regulator and pump both new and fuel pressure verified at rail. Background is I have an intermittent miss at idle when warm. Cooolant and fuel temp senders give good and sensible readings all through the warm up. TPS is smooth and idle mode activates okay. Have been through everything and yes changed everything as it's all mainly original so I'm not at all averse to replacing items even if they aren't broke. The engine has even been overhauled by Turner!

Looking for some thought analysis as the long term trim seems to be steering me to something, just can't think what...
 

·
Registered
Joined
·
313 Posts
I've not got a closed loop system on mine as it predates cat fitment, a 1991 year. So experience isn't direct on this for me.

Presumably if it's all ok sensor wise it's responding in trying to get the exhaust parameters within range?

Have you got a exhaust emmisions readout / test to see what is coming through to an independent measurement device to cross check what it's putting out?

Again presumably, any missing will result in raised level of unburnt fuel going out, which would trigger ecu response in trying to mitigate that feedback.
 

·
Registered
Joined
·
198 Posts
If you have a misfire. you will end up with unburned fuel in the exhaust stream. the O2 sensor measures oxygen potential in the exhaust, it is incapable of measuring the amount of fuel present. since a misfire will dump unburned fuel AND unburned oxygen in to the exhaust the O2 sensor will interpolate the unburned oxygen as the engine being lean.

You may want to verify and clean the computer and engine ground locations. If the ground for the engine or computer aren't making a good connection it can cause similar issues.
 

·
Registered
Joined
·
48 Posts
Discussion Starter #4
Thanks guys, on the short term the odd bank is -30 to -40 at idle whereas the even floats around 0. As it's trying to reduce fuelling it would seem I'm experiencing a rich condition - would the miss make the lamda see that bank as rich or lean?

Where's the earth for the ECU? Bought another used ECU from eBay which should arrive tomorrow. Jumping the gun but I've discounted mechanical issues, vac leaks, sensors, entire ignition system, fuel system and the hot miss persists...

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
48 Posts
Discussion Starter #6
A miss fire will typically show up as a lean condition since there will be unburned oxygen in the exhaust stream.
That was my feeling on it - so the logic was the miss is due to the ECU aggressively leaning the mixture rather than the leaning being symptomatic of the miss. Does that follow?

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
313 Posts
If it's leaning one bank out significantly more than the other, wouldnt that indicate the report from each exhaust branch was leading to that decision? Ie one being offset from the other.

I guess the major question here is will a ignition derived misfire cause or be the originator of decision making to lean it out, or would sensor ecu driven logic lean it out until it starts to misfire?

Swapping parts Is potentially going to let you see that. If another ecu does the same, then it would suggest it's other vehicle hardware giving origin to the fault you see.

.

Above Link gives interesting discussion, not necessarily to follow their mods but to give more view to what is happening. Interesting about the different maps used.

What plugs and HT leads are you using?
 

·
Registered
Joined
·
48 Posts
Discussion Starter #8
The car was very original so all items looked original except the alternator, hence why I'm quite ok with throwing new parts at it. Other than fuel and coolant temp sender's and TPS I think everything you could name is new. Those are behaving in rovergauge.

I'm using NGK iridium plugs, single point not silly four prongs or some such, fresh silicone leads from Rimmers (thinking of magnecore but current ones were only changed to try to cure the hot idle miss and they ohm test fine so comfortable that they are not the issue. New (proper) red rotor and the cap with the 123 distributor seems great quality over the cheap feeling ones (had original Lucas on the car before it). Think it's beru and has the enclosed contacts and the density of the plastic feels right if you follow. Performed all of the ignition tests in rave plus some of my own and will swear it's not ignition related. Fine while driving and cold, coil is new Bosch, amp was replaced but now surplus due to the 123.

ECU is here so I will swap it tomorrow - all parts changed have been in an effort to rid me of the hot idle miss so while I understand it might induce a new issue the original issue remains so it kind of points to old units being ok. The 123 ignition was most recent bid as I had slight timing scatter, didn't fix issue but did perk up things greatly (recommended if you don't fancy megajolt or squirt)

Truly at wits end now, base idle was reset, all vac ancillaries removed and discounted hot miss remains. New stepper was actually inferior to the original so I might go back with that...

Only item yet to be replaced is the ECU, my theory is exactly as you say, rather than reacting to the miss, it is in fact causing it as it only happens when at operating temp so only when the ECU is using lambda. Now I think of it what harm would running it with a non cat tune resistor do in the short term? Just to prove the point rather than run it like that long term??

Will give that pH thread a read now - I can't think of any thread I've not read on this . Thanks for sticking with it, just need someone to mention something I've not tried and it will cure it for sure!

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
313 Posts
Reading your original post, about engine built by Turner comment. Was it doing this before that rebuild? Or just since? Leading to a question of checking rocker shaft shimming, it's conceivable that with borderline clearance on one bank of cylinders it could be leaking compression down exhaust manifold. I know it's a remote chance but possibly worth checking hot vehicle compression readings, warm up, all plugs out, remove fuel pump fuse and crank with throttle wide open, to see if you have parity from one bank to the other.

Plug leads, I've never used anything but generally available silicone leads without any problems at all. Plugs I've not tried iridium so no experience, but use NGK BPR6ES just standard electrode plugs. I use resistance plugs as well as leads as it appears to give more tolerance of running weaker/accurate mixture in these engines. I know this essentially contradicts most views on forum generally as they always get stuck on bigger, hotter, flasbang sparks, also reduction of resistance too. These chambers are never going to be fast burning and I feel the resistance elongates the spark duration and makes better use of cylinder squish effects when burning optimum mixture.

Mine is currently running MAF voltage check of 0.75 but it has got close to 160,000 miles on it.

I don't advance static ignition timing either, it's got plenty of total advance as far as I see in use, mine is static 6degree but it wanders slightly with the wear in cam chain so it's a slightly aggregate reading ;)
 

·
Registered
Joined
·
48 Posts
Discussion Starter #10
HI RR,

It had the hot idle miss before going to turner - it had 110k on the clock so cam wear, slipped liner potential etc made the decision for me. All ancillaries were replaced, new pump, radiator, idler, oil pump gears, pressure relief spring and so on. I replaced the ram housing and inlet gasket (valley) with vac leaks in mind. In any event the hot idle miss is exactly as it was before. New plugs and harness at this time too.

Subsequently the fuel pump, fuel regulator, coil, AFM, distributor, base idle and so on. Looking through the system I'm left now with the harness or the ecu. It's only when hot but this could be either due to the open vs closed loop or the idle target rate being lower after warm up and the higher idle masking it...

Kind of agree with you on the leads and plugs but after enough reading through all the lumpy idle threads I knew I'd be chatised for not having done the rotor arm cap leads etc!

Couldn't get ecu fitted yesterday so will have a go today - may just slave it in as I don't believe the chassis ground is used on the ecu itself...

Might update the spare with the later software so the whole exercise is not a complete loss! Do you know if there are de-cat y pipes with the lambda ports still present?
 

·
Registered
Joined
·
48 Posts
Discussion Starter #11
Well replacement ECU seems to yield same result, will drive more to confirm.


Read that with interest, going to try a non cat resistor and play with the co trim for idle mixture. Theory is if I can get it to idle without the miss then I'm not mad and it could be that the car had a performance cam in before (did look like someone was in the timing cover before me) and does once again causing the rough idle/miss. Am I correct in thinking a cat equipped vehicle dismisses the co trim value and uses lamda alone?

Will update as I go just for record keeping if of little interest to the community.

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
313 Posts
I don't have definite answer regarding co trim, but it looks to me that it runs that as neutral /baseline operation and trims from there as yours is doing. It would either have to have that or an arbitrary figure in ecu surely?

Mine had all of anything adjustable "optimised" when I got it. Someone had just turned most things in the hope of whatever they thought they were after.

The blue red report from my MAF was reading 2.1volts! It was definitely rich, smelly exhaust and you could smell petrol dilution in the oil. Crap torque as well as you either had to ring it's neck uphill with it laden or manually knock it back to 3rd to maintain speed.

Ending up with 0.75volts on MAF trim, my HC reading is aggregate 0.380 and CO 80 now, both readings from mot testing over last six tests.

Have you got any emmisions data for yours?

If you want to try adjust, do as your thread nz thread advises, take voltage read first then test it yourself. Higher voltage is richer, and obviously the inverse for leaner. Be interesting to see what you've got now.
 

·
Registered
Joined
·
48 Posts
Discussion Starter #13
No emmissions data, fuel consumption seems about right, getting 15mpg with mailnyl crap town mileage, up to 18-19 if on a run. Smells right when at the exhaust on idle and plugs look spot on, could be mildly lean if anything...

I'll make a reference of existing trim value and adjust with the green tune resistor and then move back to white to see what affect that has. Starting to wonder if the torque cam needs a modified tune for the 14cux and I unwittingly had one already fitted upon purchase. Doubt turner would still have the old cam now but worth an ask.
 

·
Registered
Joined
·
313 Posts
Been thinking through the sequencing of that ecu decision, plus the combination of hardware that's been changed.

Is it possible to swap the Lambda sensors with each other?

Also suspect the integrity / resistance of loom to Lambda sensors.

It's broadly the same as mine for economy, mine reached a peak of 19.7 mpg long run (ooh so close to 20) and worse just punting around at more or less 17. So not much difference.
 

·
Registered
Joined
·
48 Posts
Discussion Starter #15
Swapping the lambda from left to right or swapping the harness to leave lambdas in there current position but have them measure the wrong bank so to speak?

Lambda replacement was another one that was done in an attempt to rid me of the idle miss, plus they look very original so were due anyhow...

I can check resistance at the ECU connection if that's what you mean which would I guess show any relative difference from side to side.

Got my 470 ohm resistors today but doubt I can play tonight, tomorrow I've time so will do as you suggest and also play with co trim on original AFM whilst in non cat tune to see what that highlights

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
313 Posts
Yes, swap the sensors over. No change would suggest parity of the two, error swaps with sensor would indicate a difference. As they are new you'd expect them to be the same, so I doubt if it would cure as it was not fixed when you replaced them.

But the loom, if you swapped the plugs on sensors then it wouldn't show anything as having sensor report the same wouldn't give you a change. If wiring to right hand bank sensor was high resistance it would still report to the ecu the same as now so wouldn't confirm the fault, if that makes sense.

You'd just have to meter through each Lambda to ecu route to find impairment as far as I can see.

Be interesting to see what you get with 470ohm tune resistance installed.
 

·
Registered
Joined
·
48 Posts
Discussion Starter #17
I swapped the sensors over but the disparity doesn't follow. I did a resistance check at the harness for the ECU and the odd bank lamda read open circuit while the even read a sensible figure (have it in my notes) checked over and over and couldn't get anything but an open/infinite resistance to the odd. Despite this they both dance on the short term trim at idle in rover gauge so I need to do more digging there.

The home made tune resistor worked fine, showed 1.7v stock for the original AFM and 2.5v for the NOS Lucas overhauled later jaguar one (with no co adjust). Once the original was leaned out to 0.35 the idle seemed less lumpy but the very short drive I did showed a lack of power in comparison to how it was.

Bought an sdt 202 smoke tester (tidy piece of kit for £150) and this showed a vac leak on both sides of the throttle butterfly spindle. Nothing life changing but I've order new seals anyway. Did an exhaust test while in town and showed a leak at the downpipe on the even side and a gusher further back (where rear box mounts - aftermarket stainless exhaust so has a split there). Worth doing but I'm quietly confident neither are my issue but will resolve and report back.

Keep going around the combustion circle, air, fuel, spark and should get there.

Can anyone answer if they have a lumpy idle with the Kent h180 cam? Also does anyone with rover gauge and a cat 3.9 see long term trim less than 100%?

Sent from my LG-H930DS using Tapatalk
 

·
Registered
Joined
·
313 Posts
Agree that if you keep eliminating clear faults it should eventually get there.

Thoughts on the trim display question. Speculation from my point but worked on producing something similar taking readings from flow driven impeller data for processing machinery as roverguage is doing.
We had data flow in raw numbers streamed out of impellers, then collated into logfile, then distilled and rung though sampling to drive screen graphic, which we had trouble stabilising to give a competent readout. Decisions about sample interval and damping etc have to be made to give decent graphic, which maybe happening here in effect giving display of granularity of decision stream, if that makes sense.

Presumably 14cux is being questioned by roverguage of trim levels and not 02 sensor stream? So already one step back from the coalface in logic terms, which would be cleaned up somewhere.

Presumably if O2 sensor reports out of range it'll throw a 44 or 45 code from the ecu, but faced with no feedback from O2 what can it do? If default within ecu is to trim out to leanest position to protect cat, that could be safely caught by the general map plus MAF trim co input. So it wouldn't be dangerously lean but would run without the input of 02 sensor report.
Would the ecu do a "handshake " with both 02 sensors on boot up? And default if it didnt reach logical criteria?

Roverguage may just be giving notice of the above as it cycles looking for data from that channel. Could you use a jumper on ecu input plug to duplicate the good 02 input into both channels so the ecu thought all was OK? Just as a test.
 

·
Registered
Joined
·
313 Posts
Also found culprit on mine of rough tickover and cold start roughness.

This was only slight and controlled by IACV loop in maintaining idle rpm correctly, but sometimes you could feel it start with a cylinder out but then chime in shortly after.

Starting and running for 30secs, switch off and check temps of exhaust ports (std issue thumb on metal method) it was number one cylinder repeatedly. Switched plugs with another but still the same.

Took off that HT lead and found inside the cap that lead was released from the metal crimp with the spark jumping from lead core to metal and then obviously to plug electrode. There was black carbonisation / sooty marking where it's being doing it for some time.

To test it, I cut off and re-established the connection with all symptoms now gone. It's as smooth as could be again.
 

·
Registered
Joined
·
313 Posts
I've been reading through more fine tuning forum posts for these. See what you think of following.

It looks that mine (non cat) is solely dependant on that co trim pot of MAF for its mixture. Your cat and 02 equipped has baseline set for mixture with MAF pot, and then trimmed for each bank according to loop closing with sensor in each pipe.

So, I set my co pot trim (now at 0.645v) and yours is set to median of 1.7v start point, then yours can trim itself on demand even to where mine is if necessary. Net result is they can potentially end up in exactly the same place but yours MUST start at that setting if it's not to overlay it's trim value on top of an already leaned out start point.

Further to that, it looks like yours has one bank trimmed out permanently from having no closed loop on that faulty 02 report, giving no parity with the other bank and hence a rough idle.

It does look like you've got to correct that 02 report to get anywhere when running a cat tune resistor. OR non cat resistor and just pot trim it for parity on both banks.
 
1 - 20 of 21 Posts
Top