Obdtuner Calibration

Check for knock sounds in the engine bay. Loose exhaust, worn mounts, engine movement when suddenly WOT etc. which can trigger knock due to frequencies. Had knock on a R53 mini due to loose exhaust.

Check for knock sounds in the engine bay. Loose exhaust, worn mounts, engine movement when suddenly WOT etc. which can trigger knock due to frequencies. Had knock on a R53 mini due to loose exhaust.

 

Logging sessions show that I was detecting knock retard when i had the standard exhaust system and the custom exhaust system. No parts of the exhaust system are touching chasis/body.

 

Logging sessions show that I was detecting knock retard when i had the standard engine mounts and the black series engine mount inserts.

 

Ive determined that when operating in WOT (open loop) i get more KR than when operating in closed loop (during fuel learning). So confusingly less fuel results in less KR.

Ive now tried a more viscous oil (10W60) to see if its piston slap causing false KR. It has made no difference to the KR readings.

 

Really struggling on where to go from here. Done another fuel learning session and the fuel table isnt changing much at all. Why is less fuel helping to reduce KR?

MoT limits for lamda are 0.97-1.03.

MoT limits for lamda are 0.97-1.03.

That’s at idle/fast idle not at WOT

 

What I have just noticed is that during the logging sessions at WOT the lambda readings are around 1.2. Correct me if I'm wrong but I thought that a lambda value greater than 1 meant lean conditions? Are these values correct?

 

For a target AFR of 11.8 I was expecting a lambda reading of 0.8?

 

 

Lambda at WOT should be around 0.8 for an SC. If it's really 1.2 than it far too lean!

You did do a full fuel learning session? Did you also do a wideband learning?

Yes, I’ve done multiple fuel learning sessions. I don’t have a wideband sensor fitted nor do I have the obdtuner pro.

The thing that’s confusing is when carrying out a fuel learn the fueling is in closed loop control and adjusting the injector duration based on the readings from the lambda sensor to achieve a lambda of 1 (14.7:1). In closed loop I’m getting reduced KR.

But when operating in open loop I’m getting a lambda of 1.2 (17.6:1) when trying to achieve a lambda of 0.8 (11.8:1). It’s almost like the correction of the fuel table for open loop control isn’t working right?

Edited by Fordy, 25 July 2020 - 08:25 PM.

Your 0-1V lambda dont account during wot. If your 0-1V lambda correspondes with your 0-5V widebandlambda (or vica versa) during normal drive, then you know your 0-5V wideband is OK. Use only your 0-5V wideband for wot/boost tuning, dont rely on your 0-5V lambda.

Edited by TFD, 25 July 2020 - 11:24 PM.

Your 0-1V lambda dont account during wot. If your 0-1V lambda correspondes with your 0-5V widebandlambda (or vica versa) during normal drive, then you know your 0-5V wideband is OK. Use only your 0-5V wideband for wot/boost tuning, dont rely on your 0-5V lambda.

As I’ve mentioned before, I don’t have a wideband sensor.

The lambda sensor is showing that I’m running lean at WOT rather than rich. I don’t need a wideband sensor to tell me that

Tuning without wideband and relying on your narrowband...

Good luck.

Tuning without wideband and relying on your narrowband...

Good luck.

Thank you for the response.

At this moment in time I’d be happy with a safe tune. Just because I don’t have a wideband fitted doesn’t mean that I don’t agree that it is the better option for tuning. Unfortunately, I’m not currently in a position to upgrade to obdtuner pro and a wideband kit.

At the moment I am trying to address the current issue with what I have. If you have any further advice it would be greatly appreciated

Apologies, I think I've realised why I've created some confusion.

 

Peter has corrected me that the lambda readings that I am seeing on the log files are the voltage rather than the calculated lambda value.

 

So, at wide open throttle I'm getting a lambda voltage of around 1.2 V which corresponds to a lambda value of less than 1 therefore there is indeed a rich mixture as there should be.

 

So it does seem that i'm getting KR with rich conditions.

You must understand that an OE ECU knock detection is tuned for its stock pistons.

 

With you forged Wisecos (I have some too), it will barely work, the internal sounds of the engine are way different.

 

I see you degraded the ignition, but by how much ?

You should try a degraded second calibration with 3 or 4 less advance at least everywhere, if you see no change at all with that so it is unlikely to be real knock, just noises. (or eventually what it could be if one cylinder with real knock due to a bad injector ?)

You must understand that an OE ECU knock detection is tuned for its stock pistons.

 

With you forged Wisecos (I have some too), it will barely work, the internal sounds of the engine are way different.

 

I see you degraded the ignition, but by how much ?

You should try a degraded second calibration with 3 or 4 less advance at least everywhere, if you see no change at all with that so it is unlikely to be real knock, just noises. (or eventually what it could be if one cylinder with real knock due to a bad injector ?)

 

Hi alanoo,

 

Sorry I must have missed your post.

 

Does that mean that you have KR issues with forged pistons too? If so, what have you done to reduce them?

 

I reduced the ignition by around 50% in the boost region up to 4000 rpm as a test. This reduced the peak KR but didn't remove it completely.

 

I had previously tried swapping injector#1 (least KR detected on this cylinder) and injector#3 (most Kr detected on this cylinder) around and this made no difference as to the worst case cylinder.

 

Thank you

Not really, I did use a standalone ECU at the time.

We had made some knock recordings and by comparing the curves (it's basically just an audio recording) you could see the rising frequencies were way different.

 

I had some higher than usual piston to wall clearance on my build too, we built it that way.

 

I'm building a new engine now with 8.9 Wisecos and still not certain if going OBDtuner or Motec