Lambda droping to zero
Discussion
I'd started trying to get the adaptives even a very long time ago but the problems with the cams got in the way of that some what.
It seems to be running very well now at last though, I only just got it out on the road in time to pick the kids up today but as it was warm when I got back with them I though I'd plug the old lap top back in to see, the adaptive were some way out about -7 & 20, I tried resetting them but then noticed one of the Lambda's seemed to keep dropping to zero a lot at low revs (up to 2.5k), the other did a bit but not as much.
I'll take it out for a decent spin tomorrow to see if that makes any difference, but thought I'd ask now if a Lambda dropping to zero a lot would point to a fault with it?
I'm thinking if it is faulty there not much point in wasting time trying to get the adaptives even.
It seems to be running very well now at last though, I only just got it out on the road in time to pick the kids up today but as it was warm when I got back with them I though I'd plug the old lap top back in to see, the adaptive were some way out about -7 & 20, I tried resetting them but then noticed one of the Lambda's seemed to keep dropping to zero a lot at low revs (up to 2.5k), the other did a bit but not as much.
I'll take it out for a decent spin tomorrow to see if that makes any difference, but thought I'd ask now if a Lambda dropping to zero a lot would point to a fault with it?
I'm thinking if it is faulty there not much point in wasting time trying to get the adaptives even.
Straight from the diagnostic help file
LAMBDA 1 & LAMBDA 2
Lambda 1 & 2 show the signals from the lambda sensors on each bank. After starting the car the sensors will take at least 30 seconds to warm up, before they will read correctly. Once warmed up, the lambda signals should switch between approx 0 & 1-1.5 Volts. 0 volts shows the mixture is lean, 1 volt shows the mixture is rich. The rate of switching will alter with engine speed, and the percentage of time spent rich relative to lean will alter with engine load. If reading is constantly 0V, check sensor for proper connection, and sensor heater wires for continuity (6 ohms between white & red wires)
When running correctly the lambda voltages should switch between 0 & 1V (approx)
When under light load these voltages will stay predominantly 0 V with rich spikes 1V.
They should switch regularly. Rs-AJP app is
The best for showing switching on the analouge
Dial.
Sounds lazy to me hth
LAMBDA 1 & LAMBDA 2
Lambda 1 & 2 show the signals from the lambda sensors on each bank. After starting the car the sensors will take at least 30 seconds to warm up, before they will read correctly. Once warmed up, the lambda signals should switch between approx 0 & 1-1.5 Volts. 0 volts shows the mixture is lean, 1 volt shows the mixture is rich. The rate of switching will alter with engine speed, and the percentage of time spent rich relative to lean will alter with engine load. If reading is constantly 0V, check sensor for proper connection, and sensor heater wires for continuity (6 ohms between white & red wires)
When running correctly the lambda voltages should switch between 0 & 1V (approx)
When under light load these voltages will stay predominantly 0 V with rich spikes 1V.
They should switch regularly. Rs-AJP app is
The best for showing switching on the analouge
Dial.
Sounds lazy to me hth
Supateg said:
Straight from the diagnostic help file
LAMBDA 1 & LAMBDA 2
Lambda 1 & 2 show the signals from the lambda sensors on each bank. After starting the car the sensors will take at least 30 seconds to warm up, before they will read correctly. Once warmed up, the lambda signals should switch between approx 0 & 1-1.5 Volts. 0 volts shows the mixture is lean, 1 volt shows the mixture is rich. The rate of switching will alter with engine speed, and the percentage of time spent rich relative to lean will alter with engine load. If reading is constantly 0V, check sensor for proper connection, and sensor heater wires for continuity (6 ohms between white & red wires)
When running correctly the lambda voltages should switch between 0 & 1V (approx)
When under light load these voltages will stay predominantly 0 V with rich spikes 1V.
They should switch regularly. Rs-AJP app is
The best for showing switching on the analouge
Dial.
Sounds lazy to me hth
Doh I've read that before but forgot where I'd seen it, thanks LAMBDA 1 & LAMBDA 2
Lambda 1 & 2 show the signals from the lambda sensors on each bank. After starting the car the sensors will take at least 30 seconds to warm up, before they will read correctly. Once warmed up, the lambda signals should switch between approx 0 & 1-1.5 Volts. 0 volts shows the mixture is lean, 1 volt shows the mixture is rich. The rate of switching will alter with engine speed, and the percentage of time spent rich relative to lean will alter with engine load. If reading is constantly 0V, check sensor for proper connection, and sensor heater wires for continuity (6 ohms between white & red wires)
When running correctly the lambda voltages should switch between 0 & 1V (approx)
When under light load these voltages will stay predominantly 0 V with rich spikes 1V.
They should switch regularly. Rs-AJP app is
The best for showing switching on the analouge
Dial.
Sounds lazy to me hth
I found Joolz's old guide to setting it up now so combined with the above it sounds like it may just be doing what it should, I'll have a fiddle.
I'm finding this very useful so thought I'd post up again what Joolz said quite some time ago on here:
"If the banks are balanced, then the adaptives should be equal. It really is that easy.
The garage should have an airmass meter ... it comes supplied in the cerbera kit that every dealer got along with their computer / software / cam timing gear. I very rarely use the meter and just balance it from the values on the screen which is much quicker, but if you had a SP6 then I'd be worrying...that engine has individually adjustable butterflies for each cylinder and you cannot balance this without the meter.
The basis for setting the V8 is to firstly slacken the linkrod between the banks, then by interpreting the adaptive value relative to the throttle pot value bank to bank you can deduce which bank is sucking more air (assuming equal t-pot values this is the bank with the higher adaptive value) there is no other way of interpreting the values .. it's as simple as that. You just adjust the link / throttle pots and idle screw to get the adaptives equal, but also AS CLOSE TO ZERO AS POSSIBLE! If one bank of your engine has adaptives of around 30% then this is running about on the rich limit ... it won't enrich any more than that. So this means your engine needs a good tune up.
It's also wrong to say that the ECU throws up spurious fault codes in the garage such as the AFR error ... there's a logged fault because there IS a fault ... it's the difference in airflow that's causing it. The imbalance is more noticeable at small throttles / light cruise conditions and getting it right can give much better town driving - there's also a Joospeed mod for the 4.2 version to cut out the slop in the cross link that gives a lot of poor running at part throttle."
Here is the process I use: -
Note that the 4.2 suffers from slop in the linkage. So check the adjustments for % adaptives with the throttle very slightly open and thus any slop in the linkage already taken up.
Get the engine properly warmed up so there is no ECU fuel/ignition compensation for cold temp
Go to the logging page and record a minute of idle and occasional throttle opened a bit for a few seconds (nothing wild, 2kish is is fine)
Examine the throttle pot readings to see if there is any spiking of the readings. If there is then replace the offending throttle pot.
Examine the lambda readings - they should be varying. If not then it may be wild adjustment or duff lambda.
I also capture the adaptive maps just for interest so I can compare with the results after adjustment.
Reset the adaptives to clear any existing values before starting any adjustments. You may find this affects the idle and you may need to open the stop to get a reasonable idle.
If I think its all wildly out I remove the airboxes and use the air mass meter to check the banks are pulling about the same air flow. You can also compare what each inlet is flowing. If on a bank the air flow is significantly different across the 4 inlets the only way to adjust it is to get some drilled screws. You replace the screws that clamp the butterfly to the throttle rod with a drilled screw to increase airflow down that inlet.
Otherwise I leave the air boxes on and don't clamp the purge line.
Get the TPot readings even first - adjust idle if necessary.
Highest adaptive reading is the bank flowing too much air - so close that bank down a bit. if its the odd bank (near side) then use the link rod to do it and then adjust the TPot back to same reading as the even bank
If its the even bank thats flowing too much air then note the odd bank tpot reading and close down the idle stop a bit and then adjust the link rod to get the odd bank back to the tpot reading you noted. then adjust the even bank tpot to the same as the even bank.
Its a lot easier to do than describe but you should be able to get towards zero on the adaptives with the tpots reading about 16.8% and idle at 950ish.
If you can't get anywhere near zero adaptives and you have checked both banks with a mass meter then the unvarying lambda is most likely goosed. Cheapest way to check is to unplug and remove the lambdas from the manifolds and swap them over. If the unvarying reading moves the lambda is duff. If the same bank shows the unvarying reading then its something else - possibly wiring but definitely not the lambda.
Other things to make it easier
Replace the TPot screws with cap heads and use a T handled allen key to slacken/tighten
Replace the link rod ball joints with rod ends from Ondrive
Fit a second throttle stop on the odd bank and use that for idle adjustment - that way the slop in the linkages doesn't have any effect - the even bank is always under tension being pulled back by its return spring.
"If the banks are balanced, then the adaptives should be equal. It really is that easy.
The garage should have an airmass meter ... it comes supplied in the cerbera kit that every dealer got along with their computer / software / cam timing gear. I very rarely use the meter and just balance it from the values on the screen which is much quicker, but if you had a SP6 then I'd be worrying...that engine has individually adjustable butterflies for each cylinder and you cannot balance this without the meter.
The basis for setting the V8 is to firstly slacken the linkrod between the banks, then by interpreting the adaptive value relative to the throttle pot value bank to bank you can deduce which bank is sucking more air (assuming equal t-pot values this is the bank with the higher adaptive value) there is no other way of interpreting the values .. it's as simple as that. You just adjust the link / throttle pots and idle screw to get the adaptives equal, but also AS CLOSE TO ZERO AS POSSIBLE! If one bank of your engine has adaptives of around 30% then this is running about on the rich limit ... it won't enrich any more than that. So this means your engine needs a good tune up.
It's also wrong to say that the ECU throws up spurious fault codes in the garage such as the AFR error ... there's a logged fault because there IS a fault ... it's the difference in airflow that's causing it. The imbalance is more noticeable at small throttles / light cruise conditions and getting it right can give much better town driving - there's also a Joospeed mod for the 4.2 version to cut out the slop in the cross link that gives a lot of poor running at part throttle."
Here is the process I use: -
Note that the 4.2 suffers from slop in the linkage. So check the adjustments for % adaptives with the throttle very slightly open and thus any slop in the linkage already taken up.
Get the engine properly warmed up so there is no ECU fuel/ignition compensation for cold temp
Go to the logging page and record a minute of idle and occasional throttle opened a bit for a few seconds (nothing wild, 2kish is is fine)
Examine the throttle pot readings to see if there is any spiking of the readings. If there is then replace the offending throttle pot.
Examine the lambda readings - they should be varying. If not then it may be wild adjustment or duff lambda.
I also capture the adaptive maps just for interest so I can compare with the results after adjustment.
Reset the adaptives to clear any existing values before starting any adjustments. You may find this affects the idle and you may need to open the stop to get a reasonable idle.
If I think its all wildly out I remove the airboxes and use the air mass meter to check the banks are pulling about the same air flow. You can also compare what each inlet is flowing. If on a bank the air flow is significantly different across the 4 inlets the only way to adjust it is to get some drilled screws. You replace the screws that clamp the butterfly to the throttle rod with a drilled screw to increase airflow down that inlet.
Otherwise I leave the air boxes on and don't clamp the purge line.
Get the TPot readings even first - adjust idle if necessary.
Highest adaptive reading is the bank flowing too much air - so close that bank down a bit. if its the odd bank (near side) then use the link rod to do it and then adjust the TPot back to same reading as the even bank
If its the even bank thats flowing too much air then note the odd bank tpot reading and close down the idle stop a bit and then adjust the link rod to get the odd bank back to the tpot reading you noted. then adjust the even bank tpot to the same as the even bank.
Its a lot easier to do than describe but you should be able to get towards zero on the adaptives with the tpots reading about 16.8% and idle at 950ish.
If you can't get anywhere near zero adaptives and you have checked both banks with a mass meter then the unvarying lambda is most likely goosed. Cheapest way to check is to unplug and remove the lambdas from the manifolds and swap them over. If the unvarying reading moves the lambda is duff. If the same bank shows the unvarying reading then its something else - possibly wiring but definitely not the lambda.
Other things to make it easier
Replace the TPot screws with cap heads and use a T handled allen key to slacken/tighten
Replace the link rod ball joints with rod ends from Ondrive
Fit a second throttle stop on the odd bank and use that for idle adjustment - that way the slop in the linkages doesn't have any effect - the even bank is always under tension being pulled back by its return spring.
Hmm well, I managed to make a car that was running fine run badly again 
The adaptives were about -9 & 20 (but the car was running nicely), so I shut down the bank that was high, but now no matter what I try it just ends up with both adaptives going up to about 30. Its hunting & generally crap to drive now when it was fine before
I think I'll have to get an air flow meter to check that properly first.
I've seen the many other threads on this subject so don't really want to start another, but...
I do have one question though, the pots values both go up the second I start the engine, one a bit more than the other, so I have one reading about 14 & the other about 16 with the engine off then the second I start it they both jump to 18, I can shut the engine off with them reading 18, then just switch the ignition back on & they are reading 14 & 16 again, it totally repeatable. I'm assuming as they seem to read fine & with out any glitches when the engine is running they are ok, but it seems a bit odd.
The adaptives were about -9 & 20 (but the car was running nicely), so I shut down the bank that was high, but now no matter what I try it just ends up with both adaptives going up to about 30. Its hunting & generally crap to drive now when it was fine before
I think I'll have to get an air flow meter to check that properly first.
I've seen the many other threads on this subject so don't really want to start another, but...
I do have one question though, the pots values both go up the second I start the engine, one a bit more than the other, so I have one reading about 14 & the other about 16 with the engine off then the second I start it they both jump to 18, I can shut the engine off with them reading 18, then just switch the ignition back on & they are reading 14 & 16 again, it totally repeatable. I'm assuming as they seem to read fine & with out any glitches when the engine is running they are ok, but it seems a bit odd.
FarmyardPants said:
The jump in pot values may just be down to increased voltage while the engine is running, e.g. 13.8V rather than 12V. I'm not an expert, it's just a suggestion!
Thanks yes, that's about the only reason I can think of too. I was searching though the old threads & found a few people saying the battery was the main culprit for different running problems, I really hope it isn't as the optima wasn't cheap! It does keep recording battery as a fault, but as I have to keep staring it so much to get the pots to jump up to there normal running values the battery is taking quite a pounding, so I'd been ignoring that till I saw the other threads. I've never noticed the battery fault staying on at any time, unlike the AFRs I've now managed to get to come on
Oddly, though I've seen the AFRs come on the screen they don't seem to get logged as a fault.
I tried fully charging the battery & leaving connected to the charger (on float mode) when tuning on the ignition, switching on the ignition pulled the measured voltage on the battery down from about 13.8v to 13.2v, but the diags screen showed 12.8v. I'd need to find the right pins on the ECU to test thats not a voltage drop. Either way it made no difference to the pots read out, they stayed at their same engine not running low readings.
I'll try connecting it up to another running cars battery to try to get it up to running voltage & see if that make any difference.
I'm certainly no expert on the subject but as far as I can see as long as they stay constant when it is running it should be fine, it just makes setting up a real pain with them jumping every time I tuning it off as I can't really run the engine with the connecting rod off to get the pot values...
Jhonno said:
The TP's are supposed to be set whilst the engine is hot I have been told..
If they are right when running I would work with that. I thought they had to be 18/19% oon idle though?
My burnt fingers will testify it was all hot! Possibly a little too hot actually (not the engine its self oil & water temps were fine) but the air temp was reading over 40, it is hot a the mo, but it couldn't have been over 30 then. I thought it probably just heat soak, again from having to start the engine so much to get proper readings from the pots, also the sun shining on the black air box wouldn't have been helping - an unusual British problem! But to be sure I checked the next morning when every thing had cooled & it read much more reasonable 17, the water temp was reading 15.8 or so too & they day was staring to warm up by then so they seemed spot on. If they are right when running I would work with that. I thought they had to be 18/19% oon idle though?
Thinking about it my test run of drifting round the field again did get a reasonable load on the engine but wouldn't have got much air flow though the air box to cool it down, so may be that why the adaptives kept climbing.
I guess I should really go for a proper test drive between each adjustment to be sure, I've used 1/4 of a tank of fuel testing already, I guess I'd be better off wasting fuel out on the road as it been off the road for so long anyway!
Edited by Luckyone on Saturday 4th July 10:44
dkwiscrazy said:
It sounds like you're having the same problem as me. I've just ordered a new lumber sensor. Hopefully that will fix the fault.
Thanks for the link, I hope I won't need it though. I don't think mine are playing up the readings seem to be in speck. Cracked manifolds are possible though, thats one problem with exhaust wrap, I didn't want to take it off to check for cracks when I had them off.
From only reading Joolz's post I'd miss understood how it all works, I read though the whole help file that came with the original program & the started using the old MEB program too, it made more sense then.
I think I may need new pots, they don't spike but my readings to fluctuate a bit (more so at slightly different idle settings) I can see dThrottle kicking as a result making the revs clime or causing the engine to hunt. Though I think it should be possible to get them to a happy point where they don't fluctuate, I had got them like that last time I set it up.
I'm not sure the air temp is the cause but it can't be helping, I pushed the Cerb out of the garage yesterday to connect it up to a running car to check the pots, that got it to read 13.8v fine (the pots were no different though still reading low with the cerbs engine off) but in the few mins it took to hook up the jump leads of just the sun shining the Cerbs air temp was reading 30, with the water reading 19 & the old BM it was hooked up to said it was 21. After a run I drove into the garage before opening the bonnet, but it was still over reading a little, but that time I could feel the bottom of the air box & temp sensor were hot from the heat from the rad below.
As long as the adaptives are not maxed out, the banks ARE balanced. That's what the on-the-fly adaptive adjustment gives you.
HOWEVER
My S6 had similar symptoms and I realised that an air leak on one throttle will cause one cylinder to run lean. A lambda sensor reads oxygen content in the exhaust, it does NOT directly read the mixture. So one throttle running (very) lean can cause the adaptives to be sent skyward in turn causing the remaining throttles on teh same header to run rich.
I removed and re-sealed my throttles to the head. Massive improvement. I also added the lambda grounds and re-set the throttles with my flowmeter.
HOWEVER
My S6 had similar symptoms and I realised that an air leak on one throttle will cause one cylinder to run lean. A lambda sensor reads oxygen content in the exhaust, it does NOT directly read the mixture. So one throttle running (very) lean can cause the adaptives to be sent skyward in turn causing the remaining throttles on teh same header to run rich.
I removed and re-sealed my throttles to the head. Massive improvement. I also added the lambda grounds and re-set the throttles with my flowmeter.
Stunned Monkey said:
As long as the adaptives are not maxed out, the banks ARE balanced. That's what the on-the-fly adaptive adjustment gives you.
HOWEVER
My S6 had similar symptoms and I realised that an air leak on one throttle will cause one cylinder to run lean. A lambda sensor reads oxygen content in the exhaust, it does NOT directly read the mixture. So one throttle running (very) lean can cause the adaptives to be sent skyward in turn causing the remaining throttles on teh same header to run rich.
I removed and re-sealed my throttles to the head. Massive improvement. I also added the lambda grounds and re-set the throttles with my flowmeter.
Yes seems my problem was similar, once my new flow meter turned up I found some of the bodies were flowing more air, lots of tweaking of the butterfly's later & it was much better. Its amazing how much the flow can differ with butterfly's that look like they have the same gap. After all the problems I've had with cams I'm paranoid about unwanted noises from the engine, before I got the flows right I kept hearing a slight knocking noise, after tentatively listening to the cam covers with a screwdriver (which sounded fine) I realised it was "the knock" or det due I guess one cylinder running lean with more air than the others on that bank. HOWEVER
My S6 had similar symptoms and I realised that an air leak on one throttle will cause one cylinder to run lean. A lambda sensor reads oxygen content in the exhaust, it does NOT directly read the mixture. So one throttle running (very) lean can cause the adaptives to be sent skyward in turn causing the remaining throttles on teh same header to run rich.
I removed and re-sealed my throttles to the head. Massive improvement. I also added the lambda grounds and re-set the throttles with my flowmeter.
The old throttle linkage rod problem was causing me some pain. So one morning I tested the butterfly to body gap on the bank opposite the stop bolt (the one being held open by the rod) the thinnest feeler fitted nicely. I then warmed the rod with a hot air gun & tested it again, the feeler was very tight & the butterfly snapped shut as it came out. So I kept the link rod hot with the air gun after that till much later when the engine was hot enough to keep the rod hot.
I need to plug the laptop back in to be sure but it's certainly running much better now, pulling much more cleanly.
There is a slight hesitation at 2k rpm but I was just talking to Julez about remaps & he said its something of a feature of the AJP
Edited by Luckyone on Saturday 11th July 21:06
Gassing Station | Cerbera | Top of Page | What's New | My Stuff