M8 Lambda Sensor

[Olivert]

I've tried to search the forum archives, but am told that the search term "lambda" is not allowed as it is too common!....

I have an old M8 ecu. This is running a 4-litre Judd EV. I have struggled with a lack of power above 9,500 rpm and at the weekend swapped to a map from a different engine - all sensor settings were from the original map, with the fuel and ignition maps above 4500rpm spliced together.

Suddenly, I have more power than I can deal with! In addition, a strange rythmical cutting of the engine (about every 6 seconds for about 0.25 secs) at 3,000 to 6,000 rpm when above 65C has gone...

I am however concerned about the AFR and I would like to check and maybe make map changes on the track.

From what I have read, I can only use a Bosch LSM 11 (0 258 104 002) sensor with the M8. I already have the wiring in the loom for it. I hope I can log lambda in the ADL I have.

Can anyone tell me whether the LSM 11 will give me what I want? As a novice... what lambda values should I be aiming at, what ones are dangerous?

Thanks,

Oliver

[RossB]

The LSM 11 is the only sensor that can be wired directly to the M8. You can actually run 2 of them.

I am not really sure what your missfire could have been. Did you use a file from another Judd EV engine?

I am guessing that this a hillclimb car. You should be looking at mixtures of around 0.90 Lambda at full throttle. You can run it richer than this if you want to make it safe but I wouldn't go any leaner.

[Olivert]

Ross,

thank you for the reply. It is indeed a hillclimb car. The cutting of the engine was very strange. Some engine people suggested a corrupt map - after changing the map I am inclined to agree.

Please excuse the dumb question, but if 0.9 is my target, is lower richer and higher weaker?????

Do you know whether the LSM11 gives an instantaneous readout? If I log a lambda of 1.1 at an rpm of say 10,000 at 100% throttle, is this the true value? Or should I compensate for a delay in the sensor? My logged data is likely to be at highly transient conditions. Perhaps I should go testing and attempt to log at near constant rpm?

Ideally I remove the engine and return to the dyno. Time and money do not allow for this.

And finally - Bosch appear to offer 3 different LSM11s - the LSM 11 (0 258 104 002), the LSM 11-PM (B 261 209 105 01) and the LSM 11-RM (B 261 209 101). When comparing all three, I cannot see any difference in their performance, just the connector type. The simple LSM 11 is by far the cheapest and I need to remove the connector anyway to match it to my loom. Can you advise?

Thanks,

Oliver

[RossB]

Normally if a file gets corrupted you just end up with a junk file that won't work. Of course anything is possible but it is an unusual problem. We could test the file on a simulator if you want.

This engine is a 76 degree V8, is it an oddfire engine?

WIth lambda the lower the number the richer the mixture, so 1.1 is lean, 0.9 is richer.

There is a slight delay in lambda response time, roughly 0.3 -0.4 seconds, you can measure it on a dyno. Unfortunately the LSM sensor is not as good at responding to transients as the more modern 5 wire sensors. Your idea of doing some testing would probably be sensible.

As far as I know the only variation on the LSM 11 is the connector.

[64hardtop]

Sorry to hijack this thread but....I have an M8 with dual lambda enabled. I also have a third party dual wide band controller that outputs 0-5v on each channel. Unfortunately both channels share a common -ive connection where as the wiring diagram for the M8 shows separate -ive connections for both lambda inputs (pins 36A and 10B).

Can I common both these pins together to be able to use this controller or will that cause a problem (earth loop?) with the ECU?

Thanks.