Lambda wiring / circuit diagrams?
Discussion
I'm not convinced my lambda sensors are working too well. The car "hunts" a lot at light throttle, as if the ECU is trying to force the lambda signal across the stiochromatic point but causes a misfire doing so.
As soon as the throttle is opened it clears and pulls strongly.
I know TADTS to some extent, but I'd like to stick a scope on the output to see if the lambdas are reacting at all/reacting quickly enough.
Since they are 3 wire sensors, I guess there is a signal, a +12v, and a ground, but is the ground for the heater or for the signal or both?
Any ideas on the voltages/frequency of a good sensor? Are we in the 0.2-0.8v, 1Hz standard kind of thing?
Does anybody have circuit diagrams for a pre-serp Grif 500? Or the colour codes for the lambda sensors? I've got the 4.3 pre-cat from the SH book which doesn't help a lot!
Thanks!
As soon as the throttle is opened it clears and pulls strongly.
I know TADTS to some extent, but I'd like to stick a scope on the output to see if the lambdas are reacting at all/reacting quickly enough.
Since they are 3 wire sensors, I guess there is a signal, a +12v, and a ground, but is the ground for the heater or for the signal or both?
Any ideas on the voltages/frequency of a good sensor? Are we in the 0.2-0.8v, 1Hz standard kind of thing?
Does anybody have circuit diagrams for a pre-serp Grif 500? Or the colour codes for the lambda sensors? I've got the 4.3 pre-cat from the SH book which doesn't help a lot!
Thanks!
Hi Scoule,
Mine had this problem. I sorted it by swapping my white tune resistor for a green one (i.e the ones that pre-cats have)it basically tells the ECU that you have no cats. I have kept the lambdas connected and once swapped just reset the ECU, tweeked the AFM and after a blat it just got better and better. I have done about 200 miles since and it drives like a different car!! I have heard lots of different views on disconnecting by passing the lambdas some say it will now damage the cats others say that the lambdas never work properly as they are cheap and not tuneable fast enough to keep up with the ECU and injection system. So who knows?? All I know is that it has worked for me as it was always bugging me everytime I hit a 30 zone, Griff asking for more fuel but the lambas saying NO!! but not any more!!
Lig.
Mine had this problem. I sorted it by swapping my white tune resistor for a green one (i.e the ones that pre-cats have)it basically tells the ECU that you have no cats. I have kept the lambdas connected and once swapped just reset the ECU, tweeked the AFM and after a blat it just got better and better. I have done about 200 miles since and it drives like a different car!! I have heard lots of different views on disconnecting by passing the lambdas some say it will now damage the cats others say that the lambdas never work properly as they are cheap and not tuneable fast enough to keep up with the ECU and injection system. So who knows?? All I know is that it has worked for me as it was always bugging me everytime I hit a 30 zone, Griff asking for more fuel but the lambas saying NO!! but not any more!!
Lig.
Reading the bible the sensor should have three wires white/orange +12v, black ground, and an unspecified other colour which is the sensor lines... These should be easily identifiable as they should be shielded wires with additional 'insulation'. With the engine running you should read between 0.5 and 1.0v (0.5 being weaker mixture 1.0 being rich but a scope would be needed to get a decent reading I think), 0v being knackered..
From the little experience I have if you remove the sensor from the exhaust ( not sure if you have yet) and look at the 'tip', if its damaged either missing entirely or pitted/wholly it a good indication as to its functionality.
I would try to find out if they're working or not, before you decide whether to change the tune resistor to a precat type or replace the sensors. Although they aren't the best lambda sensors in the world they do offer closed loop control of fueling, and without them the car will probably run rich eating the cats away! Not a problem until the MOT when I would have thought it would fail on emissions and you'd then need to fix the lambda sensors and cats..
Lig may have a way around this or know better than my guess?
Also by-passing the closed loop control means you could be just hiding another fault i.e. faulty throttle pot which I read somewhere can 'do battle' with the working lambda sensors as to how much fuel is required causing light throttle running problems.. Don't get me wrong it probably is the sensor not something else, but I would fix them not by-pass them..
Mines a pre-cat (so you can't trust me) and the idea of closed loop control of fueling is very appealing to me. Otherwise its a clock work system, the ecu not knowing what effect the amount of fuel its metering out is really having, it keeps putting the same amount regardless too much or too little..
An engine boffin may have a different opinion I'd love to hear it. cos I've wondered what effect it would have fitting lambda sensors to a pre-cat apart from costing me a tidy sum. not really something I'm prepared to do apart from with my 'Imaginary / one day' bank account..
Good luck and post a conclusion when you have one, I always want to learn more and no doubt be proved wrong.
From the little experience I have if you remove the sensor from the exhaust ( not sure if you have yet) and look at the 'tip', if its damaged either missing entirely or pitted/wholly it a good indication as to its functionality.
I would try to find out if they're working or not, before you decide whether to change the tune resistor to a precat type or replace the sensors. Although they aren't the best lambda sensors in the world they do offer closed loop control of fueling, and without them the car will probably run rich eating the cats away! Not a problem until the MOT when I would have thought it would fail on emissions and you'd then need to fix the lambda sensors and cats..
Lig may have a way around this or know better than my guess?
Also by-passing the closed loop control means you could be just hiding another fault i.e. faulty throttle pot which I read somewhere can 'do battle' with the working lambda sensors as to how much fuel is required causing light throttle running problems.. Don't get me wrong it probably is the sensor not something else, but I would fix them not by-pass them..
Mines a pre-cat (so you can't trust me) and the idea of closed loop control of fueling is very appealing to me. Otherwise its a clock work system, the ecu not knowing what effect the amount of fuel its metering out is really having, it keeps putting the same amount regardless too much or too little..
An engine boffin may have a different opinion I'd love to hear it. cos I've wondered what effect it would have fitting lambda sensors to a pre-cat apart from costing me a tidy sum. not really something I'm prepared to do apart from with my 'Imaginary / one day' bank account..
Good luck and post a conclusion when you have one, I always want to learn more and no doubt be proved wrong.
I had similar problem which was driving me mad, checked all the obvious things such as throttle pot etc, eventually Rob at V8 developments (which is just a few miles from me) had a look, he changed the tune resistor to take the lamda sensors out of circuit and what a transformation.
On the fuel consumption, on run to Le Mans and back we were'nt hanging around but still returned 22-23mpg overall (there was a fair bit of steady motorway running though which is bound to help fuel consumption)
On the fuel consumption, on run to Le Mans and back we were'nt hanging around but still returned 22-23mpg overall (there was a fair bit of steady motorway running though which is bound to help fuel consumption)
I tried the LH lambda with a multimeter/scope before posting the initial message but saw no signal whatsoever. To qualify this, I didn't piggyback onto the connectors, but removed the connector from the loom so it's entirely possible that the heater circuit being disconnected meant that the lambda wasn't hot enough.
The car was at full operating temperature - I don't know whether it should've worked under that scenario, anyone know?
I'll pierce the wires with pins and check again, but I thought I'd give the neighbours a night off from half an hour of Grif tickover!
The car was at full operating temperature - I don't know whether it should've worked under that scenario, anyone know?
I'll pierce the wires with pins and check again, but I thought I'd give the neighbours a night off from half an hour of Grif tickover!
The second edition has the wiring diagrams for the ECU (traced and redrawn by myself) See page 84.
If it is hunting at idle, have you tried cleaning the stepper motor?
As for bypassing the lambdas.... it will kill the cat and the emissions tend to go out of the window which makes it difficult come MOT time. The problem is that by putting the resistor back may not reset the system as by that time the damage to the cat and lambda sensors has been done.
The lambda sensors are tested by the ECU and it will generate an appropriate fault code. Worth going to a dealer and asking them to plug in their fault reader. Only takes a minute.
Steve
www.shpub.co.uk
If it is hunting at idle, have you tried cleaning the stepper motor?
As for bypassing the lambdas.... it will kill the cat and the emissions tend to go out of the window which makes it difficult come MOT time. The problem is that by putting the resistor back may not reset the system as by that time the damage to the cat and lambda sensors has been done.
The lambda sensors are tested by the ECU and it will generate an appropriate fault code. Worth going to a dealer and asking them to plug in their fault reader. Only takes a minute.
Steve
www.shpub.co.uk
SteveC
We meet again :smile:
Been tagging the same thing in the Chimp forum.
Pete H and co have put me onto the vac advance falling of the dizzy first, which it was loose so have a nosey.
Tigtened up loose vac screws on dizzy, sucked the vac pipe to check holding vac, will see on run to work in the AM.
We should have a Steve and Bob Tivver/Lotus forum soon !!
We meet again :smile:
Been tagging the same thing in the Chimp forum.
Pete H and co have put me onto the vac advance falling of the dizzy first, which it was loose so have a nosey.
Tigtened up loose vac screws on dizzy, sucked the vac pipe to check holding vac, will see on run to work in the AM.
We should have a Steve and Bob Tivver/Lotus forum soon !!
Anyone else would think we never do any work ;-)
I checked the lambdas and they were oscillating from 0-1v at idle so are looking pretty good. They weren't oscillating very fast but I don't know how fast they should flip at tickover - any ideas anyone?
I'll replace the ticking manifold gasket on 7 or 8 (rear most passenger side) and see if that helps the light throttle hunting.
I checked the lambdas and they were oscillating from 0-1v at idle so are looking pretty good. They weren't oscillating very fast but I don't know how fast they should flip at tickover - any ideas anyone?
I'll replace the ticking manifold gasket on 7 or 8 (rear most passenger side) and see if that helps the light throttle hunting.
They should flip about once a second in closed loop mode. They'll get slower with age and eventually become unresponsive, so if it's significantly slower than this it might be worth replacing them anyway. However, be aware that other faults in the engine management system can also prevent the ECU from achieving lambda 1 so a steady value from the lambdas indicates a fault, but not necessarily a fault in the lambda sensor.
Pete,
Can you spill the beans on your last post a bit, I'm keen to keep up with the Lambda check that Steve's following through, same applies to the Chimaera I guess ?
If only Andy Freescan had been a Tivver man eh Steve ?
What mileage has the Griff done Steve, could the Lambda sensors be burnt tips ?
>> Edited by s4s280 on Friday 1st July 22:38
Can you spill the beans on your last post a bit, I'm keen to keep up with the Lambda check that Steve's following through, same applies to the Chimaera I guess ?
If only Andy Freescan had been a Tivver man eh Steve ?
What mileage has the Griff done Steve, could the Lambda sensors be burnt tips ?
>> Edited by s4s280 on Friday 1st July 22:38
This site has all the details of wiring colours and how to test the sensors.
The main failure mode of lambda sensors is that they gradually take longer and longer to react to changes in the exhaust gas. This means that the ECU takes longer to realize when it has passed through lambda 1, so the regular oscillations between rich and lean get slower and slower. Eventually it stops working altogether and the ECU just heads over to the maximum lambda correction (which iirc is about 25% correction, massively wrong).
The main failure mode of lambda sensors is that they gradually take longer and longer to react to changes in the exhaust gas. This means that the ECU takes longer to realize when it has passed through lambda 1, so the regular oscillations between rich and lean get slower and slower. Eventually it stops working altogether and the ECU just heads over to the maximum lambda correction (which iirc is about 25% correction, massively wrong).
GreenV8S said:
[url]... eventually it stops working altogether and the ECU just heads over to the maximum lambda correction (which iirc is about 25% correction, massively wrong).
To sum that up then Pete, if the ECU heads over to max correction, I guess that means massive overfuelling rather than leaning.
Also, if two similar aged slower sensors, whilst not working "as new" were in place, would it be less of an issue than 1 sensor failing. i.e. if one (cylinder bank) sensor fails and the other is working correctly does the ECU fuel to the working one or do you get a big inbalance and obvious poor running.
Bob
I believe that each bank has its own feedback loop based on its own lambda sensor, so one unresponsive sensor could cause that bank to head off out of control while the other carries on OK. I don't know the details but the ECU has some heuristics to detect sensor failure so I think if one sensor stopped working completely the ECU would detect this and drop into 'limp home' mode rather than just sit at max correction indefinitely. But if it's working but very very slowly the lambda control algorithm will suffer from massive overshoots causing the fuelling to jump up and down, perhaps by as much as 25%.
Gassing Station | Griffith | Top of Page | What's New | My Stuff