Lambda voltage readings?

[chris52]

Still trying to sort out the lean running on the Griffat high RPM so decided to check all the ECU sensors. All checked out Ok except the Lambda below are the readings I got.

Both lambda readings taken at ECU plug and gave pretty much the same readings

Cold start and idle = 1.1 to 1.28 volts
once warmed at idle they were cycling between 0.02v to 1.32v cycling every couple of seconds with various readings in between ie 0.02v then jumping to 1.15 down to 0.04v then 1.22v down to 0.08v up to 1.14v etc etc.

In 5th gear starting at low rpm and full throttle a steady reading of around 1.22v only did this for a few seconds.

All it says in the bible is you should get a reading between 0.5v to 1v so I am guessing something is wrong but with exactly the same readings from both lambdas is it more likley to be ECU related or that both lambdas are Bu66ered

Cheers in advance
Chris

[blitzracing]

Those readings are spot on, and the bible is wrong, the lambda's switch between 0 and 1 volt under tight lambda control, hitting an upper limit of 1.2 volts or fractionly more as larger corrections are made. The switching point of the probes is so rapid the ECU will never manage to hold the lean part of the cycle at .5 volt, it will drop lower than this, around 0 volts. Full load tests (above 3400 rpm, or about 2/3 throttle) will give you between 1 and 1.3 volts but this upper limit but is not a reliable measure of mixture as it alters as the probes age, and the probes are used purely as a switch by the ECU, unlike wide band sensors where a voltage will represent a particular AF ratio. The switching point is 14:35:1 AF ratio. Full load tests are difficult as it really needs to be flat out for many seconds at warp speed, as the moment you drop off full throttle, even fractionly, the ECU leans the mixture back off.

Edited by blitzracing on Saturday 28th August 21:16

Edited by blitzracing on Saturday 28th August 21:17

[chris52]

Thanks Mark. I have just rechecked the differences between the two and one lambda is cycling between 0.02 and 1.35 the other is cycling between 0.02 and 1.15 is this OK?

Chris

[blitzracing]

It just means that one bank of the v8 is fueling fractionly differently than the other, its quite normal. The problems you have is not lambda related in any way, as its occurring outside of the closed loop area under full load. Im still sticking to my origional diagnosis of the fuel starvation of some sort, pump, filter, restrictive fuel lines or leaking regulator, as even with a change in fuel pressure the mixture still ends up lean at the same point, where the fuel delivery to the injector rail cant keep up with demand so the pressure is dropping. As you are using the lambda probes for reference under load, you will see the voltage peak under hard acceleration as the ECU puts a fixed amount of extra fuel in during this "fuel enrichment stage" from whatever RPM, and the voltage should stay at 1.2 volts or more once you reach full load and you are still at full throttle. If you drop from a constant full power the ECU will lean the mixture off as it enters cruising mode and drops back into closed loop. You can run at lot leaner under this mode, but the emission control system (the lambda feedback) will then pull it back to 14.35:1 after a second or two, just to keep the catalysts working correctly, even though it uses more fuel than needed. The basic issue with road testing it is very difficult to tell the difference between the fuel enrichment stage and static full power mode, unless you have a long steep hill to load the engine up, although both these stages the lambda output should not drop below about 1.2 volts. In your case Id expect the lambda output to drop after a few seconds under full load if its getting as lean as 14.35:1.

[TVR Beaver]

Mark, Chris... can't help with this issue but one thing I get (or can get if i create the right conditions as it does not usualy come into normal driving) is a slight switching up and down as I drive... If I try explain...
If I run under very light load at exactly 1700 RPM I can feel a slight drive... off drive... drive... off drive... this cycles about every second... do you think this is the ECU cycling? As I say.. I have to sort of make it do it but thought it was a duff lead at one point, but it's not.. it a defo on / off for the same time period...

Chris... if you want a referance I'll bring my car around.. see what that does?...

John

[chris52]

Thanks Mark for the info. I thought it best though to check all the sensors before starting to replace parts that may or may not be ok. I am thinking of getting a wideband lambder reader to check the AFR, at least then I will know if something I do will have fixed the problem without lots of RR sessions. First job will be to get the injectors ckeaned and tested followed by a new fuel pump (any recomendations of which one to get). If neither of these work then I will be pretty lost as to what to do next.
So far everything checks out Ok lambda,fuel temp sensor, collant temp sensor, AFM, etc.
Is there anything else I should check prior to starting replacing things. The car runs great no issues with power or exessive fuel consumption. It will idle very steadily and although I do get some shunting it is not exessive. If it wasn't for the RR sessions I wouldn't have suspected that anything was wrong, I have had it now on two different RR just to make sure I actually did have a problem and both AFR readings showed the car leaning off at top end
First graph this was with no adjustable regulator

second graph as above different RR

third graph with adjustable regulator and different fuel pressures

Just to note on all of these Graphs I do have a very slight manifold blow on the 4 into 1 section of the right manifold would this make much difference its not major and you can only hear a slight ticking when the engine is cold once up to temp it goes away.
thanks again for the help.
Chris

[blitzracing]

TVR Beaver said:

Mark, Chris... can't help with this issue but one thing I get (or can get if i create the right conditions as it does not usualy come into normal driving) is a slight switching up and down as I drive... If I try explain...
If I run under very light load at exactly 1700 RPM I can feel a slight drive... off drive... drive... off drive... this cycles about every second... do you think this is the ECU cycling? As I say.. I have to sort of make it do it but thought it was a duff lead at one point, but it's not.. it a defo on / off for the same time period...

Chris... if you want a referance I'll bring my car around.. see what that does?...

John

This is the exact shunting many people refer to, and yes the lambda cycling does have something to do with it in some cases, although the shunting is not in sync with the lambda cycle, so its not the only factor. Easy test is to use a 400 tune resistor to switch off the lambda cycle as a test only. As you have a Bosch AFM Im not sure if you have the modification to put a CO trim resistor back into the loom, as you need this to set the mixture at idle as you have no lamda feedback. All it needs is a 1k ohm variable resistor between the blue red and red black wire in the lucas part of the AFM loom.

[blitzracing]

chris52 said:

Thanks Mark for the info. I thought it best though to check all the sensors before starting to replace parts that may or may not be ok. I am thinking of getting a wideband lambder reader to check the AFR, at least then I will know if something I do will have fixed the problem without lots of RR sessions. First job will be to get the injectors ckeaned and tested followed by a new fuel pump (any recomendations of which one to get). If neither of these work then I will be pretty lost as to what to do next.
So far everything checks out Ok lambda,fuel temp sensor, collant temp sensor, AFM, etc.
Is there anything else I should check prior to starting replacing things. The car runs great no issues with power or exessive fuel consumption. It will idle very steadily and although I do get some shunting it is not exessive. If it wasn't for the RR sessions I wouldn't have suspected that anything was wrong, I have had it now on two different RR just to make sure I actually did have a problem and both AFR readings showed the car leaning off at top end
First graph this was with no adjustable regulator

second graph as above different RR

third graph with adjustable regulator and different fuel pressures

Just to note on all of these Graphs I do have a very slight manifold blow on the 4 into 1 section of the right manifold would this make much difference its not major and you can only hear a slight ticking when the engine is cold once up to temp it goes away.
thanks again for the help.
Chris

Before you go and spend your hard earned cash, you really do need to do a full load fuel pressure test. I can lend you a gauge, but you will need to buy a good liengh of fuel hose and a couple of clips so you can attach to the pressure take off point, and have the gauge where someone can read it when you are driving. No bodging please, I dont want to be responsible for you filling the cockpit with fuel it it leaks!! I had the same problem with the Ginetta and it took ages to find, I ended up replacing the fuel pump, both fuel filters, and finally the fuel line as it had becomed partly kinked under the car. (Ginetta used a plastic pipe that kinked easily so I replaced it with a bigger bore steel pipe).

[chris52]

Hi Mark I have done the fuel pressure test with John. We did both mine and Johns one after the other at different loads, speeds and full throttle right through the rev range. We both got pretty much the same results, John car is fuelling ok so I guess the pressure on mine must be OK aswell If I remember correctly we both got around 36 to 37 psi at full power in 3rd gear. But thanks for the offer anyway.
I have just invested in this http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&am... so I will be able to find out its AFR at differnt road speeds. It comes with the exhaust clamp so I can just clamp it the tail pipe and go out for a spin and see what sort of AFR readings I am getting. I know the LM-1 is not the best on the market but it should give me a fairly reliable reading as a base measure to find out if things improve.

John if you fancy having a test day we could hook this up to yours aswell and see if we get similar readings.

Chris

[TVR Beaver]

blitzracing said:

TVR Beaver said:

Mark, Chris... can't help with this issue but one thing I get (or can get if i create the right conditions as it does not usualy come into normal driving) is a slight switching up and down as I drive... If I try explain...
If I run under very light load at exactly 1700 RPM I can feel a slight drive... off drive... drive... off drive... this cycles about every second... do you think this is the ECU cycling? As I say.. I have to sort of make it do it but thought it was a duff lead at one point, but it's not.. it a defo on / off for the same time period...

Chris... if you want a referance I'll bring my car around.. see what that does?...

John

This is the exact shunting many people refer to, and yes the lambda cycling does have something to do with it in some cases, although the shunting is not in sync with the lambda cycle, so its not the only factor. Easy test is to use a 400 tune resistor to switch off the lambda cycle as a test only. As you have a Bosch AFM Im not sure if you have the modification to put a CO trim resistor back into the loom, as you need this to set the mixture at idle as you have no lamda feedback. All it needs is a 1k ohm variable resistor between the blue red and red black wire in the lucas part of the AFM loom.

Thanks Mark... I think the trim is part of the Bosch set up... you put a trip pot in when you install it by cutting a wire and tapping into another.. you then set it to one voltage for the cat's and another for pre-cat's... will give it a go at some point... but thinking about it.. it only does it when warming up... can't get it do do it when fully warm... does this follow?...I also have a spare ECU... may swap this over and see if it makes any differance... everything I'm on about is very small now anyway.. it's just interesting what makes these things happen?

[TVR Beaver]

chris52 said:

Hi Mark I have done the fuel pressure test with John. We did both mine and Johns one after the other at different loads, speeds and full throttle right through the rev range. We both got pretty much the same results, John car is fuelling ok so I guess the pressure on mine must be OK aswell If I remember correctly we both got around 36 to 37 psi at full power in 3rd gear. But thanks for the offer anyway.
I have just invested in this http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&am... so I will be able to find out its AFR at differnt road speeds. It comes with the exhaust clamp so I can just clamp it the tail pipe and go out for a spin and see what sort of AFR readings I am getting. I know the LM-1 is not the best on the market but it should give me a fairly reliable reading as a base measure to find out if things improve.

John if you fancy having a test day we could hook this up to yours aswell and see if we get similar readings.

Chris

Hi Chris.. yes.. will do.. always good to get a comparison... looks a good bit of kit... (he should have taken your offer!!)... see you at the next meeting?... John

Edited by TVR Beaver on Sunday 29th August 18:02

[blitzracing]

chris52 said:

Hi Mark I have done the fuel pressure test with John. We did both mine and Johns one after the other at different loads, speeds and full throttle right through the rev range. We both got pretty much the same results, John car is fuelling ok so I guess the pressure on mine must be OK aswell If I remember correctly we both got around 36 to 37 psi at full power in 3rd gear. But thanks for the offer anyway.
I have just invested in this http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&am... so I will be able to find out its AFR at differnt road speeds. It comes with the exhaust clamp so I can just clamp it the tail pipe and go out for a spin and see what sort of AFR readings I am getting. I know the LM-1 is not the best on the market but it should give me a fairly reliable reading as a base measure to find out if things improve.

John if you fancy having a test day we could hook this up to yours aswell and see if we get similar readings.

Chris

You've got me there then, thats not far out. The best test is to disconnect the vacume pipe on the pressure regulator at idle and note what the base fuel pressure is (well mark it on the gauge anyway). Then reconnect it and do the full power test, and see if it drops below below the mark on the pressure gauge. I was seeing about a 3 psi drop that threw the fueling right out. There is nothing on the ECU side of things that will cause that type of convergence of your 3 different base fuel pressures to the same final lean AF ratio. The injector pulse width will be the same in all three cases under the same conditions, and with a constant higher fuel pressure behind the injectors you should get a richer mixture, across the rpm range once in open loop. The ECU has no way of trimming the mixture in open loop at all and the amount of fuel trim previously applied in closed loop is ignored as well, so there is no underlying variables to consider. The only other possibility I can see is the some vaccume in the plenum is making the regulator drop the pressure, but this would only happen at WOT if the inlet tract was restricted in some way, and you would see this as a fuel rail pressure drop anyway. I assume your take of point was the fuel rail itself? One other point is unburnt mixture looks like a lean mixture to a lambda probe, so if you have some sort of combustion problem it looks lean voltage wise. I doubt however that the rolling road session would simply use unburnt oxygen as a mixture measurment, I think the gas analysers are a lot smarter than just a lambda probe.

[chris52]

blitzracing said:

You've got me there then, thats not far out. The best test is to disconnect the vacume pipe on the pressure regulator at idle and note what the base fuel pressure is (well mark it on the gauge anyway). Then reconnect it and do the full power test, and see if it drops below below the mark on the pressure gauge. I was seeing about a 3 psi drop that threw the fueling right out. There is nothing on the ECU side of things that will cause that type of convergence of your 3 different base fuel pressures to the same final lean AF ratio. The injector pulse width will be the same in all three cases under the same conditions, and with a constant higher fuel pressure behind the injectors you should get a richer mixture, across the rpm range once in open loop. The ECU has no way of trimming the mixture in open loop at all and the amount of fuel trim previously applied in closed loop is ignored as well, so there is no underlying variables to consider. The only other possibility I can see is the some vaccume in the plenum is making the regulator drop the pressure, but this would only happen at WOT if the inlet tract was restricted in some way, and you would see this as a fuel rail pressure drop anyway. I assume your take of point was the fuel rail itself? One other point is unburnt mixture looks like a lean mixture to a lambda probe, so if you have some sort of combustion problem it looks lean voltage wise. I doubt however that the rolling road session would simply use unburnt oxygen as a mixture measurment, I think the gas analysers are a lot smarter than just a lambda probe.

Mark could all this be just down to an exhaust leak giving an incorrect reading on the analyser?
I don't think that there can be much wrong as the BHP and torque are very respectable for a 5.0ltr with only the smoothbore, pre-cats removed and carbon trumpets modified.
When I get the wideband reader I will take some readings as it is, then replace the manifold with the 4 into 1 leak (I have a replacement ready to go on ) and see if the AFR changes much.
John is also gonna come round so we can do some side by side testing.
From what I can gather the car should show under normal driving conditions an AFR of around 14.5-1 and under full acceleration this should richen up to around 12.5-1 is that right?

Chris

Edit to say that the fuel pressure readings were taken directly from the fuel rail via the Schrader valve

Edited by chris52 on Sunday 29th August 21:19

[blitzracing]

Ive just checked on this. Idealling the RV8 needs 13.5:1 at idle, 12.5 under accelleration, and 13:1 at full load. You cant alter the idle much under lambda control,(thats why the tickover is a bit rough on the catalysts fuel map- its too lean) but both the acceleration stage and full power will depend on fuel pressure and mapping. Im not sure about the air leak, it would make sense if the rolling road measured excess oxygen in the exhaust, but you need to ask them.