False KR with Turbo

[TGP37]

Turning the false KR into a spark retard for spool. Pull decay back and reduce the spark pulled vs rpm.

I finally figured it out, though I suspected this for some time.

The turbo spooling up will trigger false KR. Being the KR module looks for a specific frequency (sound) the turbo will hit that frequency as it spools up and then pass it by.

This is what I found, the turbo will not trigger false KR unless it is spooling up. No false KR once fully spooled so I can keep the standard settings past 3600 RPM. For a turbo, thats cool.

Now comes the cool stuff. False KR is bad, obviously for lack of power, though retarding spark while spooling is somewhat beneficial. As long as the spark isn't retarded too much.

Now, I kept regular KR values above 4000 RPM, since the turbo is hardly ever spooling above that RPM. So I have knock protection when it counts, in boost.

What I did was reduced the amount of spark the KR system can pull between 2800-3600 RPM. Just enough to let it act as an anti-lag system.



Anyone suffering false KR with a turbo, this might be the issue.

I turned a problem into a benefit. Using the false KR from spool as anti-lag. And it works, as far as the logs show. MAF delta rate of change is slightly higher with the anti-lag concept on, immediately after the false KR.

[AleroB888]

If I understand this correctly, you have effectively increased the potential spark advance while spooling, and the result is more airflow while spooling.

If you had no false knock, do you think you could program an increase in the timing while spooling? If so, have you tried damping material on the knock sensor? Sometimes they are a little over-sensitive to the boosted whirring, whining and even loud exhaust noise.

[TGP37]

I am trying to find a good spot with the attack rate and decay vs rpm.

Ultimately I'd rather kill all false KR and just go anti-lag the proper way. Dampening the sensor may have to happen.

It seems the airflow spikes up faster per 0.1/sec tick while KR is subtracting during spool. As long as the KR pulled isn't excessive. After a few degrees it drops off.


But it DOES seem a stock KR setup can be tweeked to work friendly with turbo. It started by my reducing of attack rate in the range I had false spool induced KR. Took it for a drive and my butt dyno really felt the more aggressive pull going into boost. I tried reducing spark in the same region, it is tricky. It helped but when not in boost it didn't and became more sluggish.

I wouldn't of known this as well had I not did a massive spark tune project. Logging hours of drive while altering the over all spark 0.5 a step for a total of +/- 6 degrees. Running tables in excel I found what spark degree made the greatest torque on average per cell in the spark map. Worked well.

That was how I found the torque anomaly during mild KR. ( I had reduced all max KR to no more then 2 degrees while tuning this stage below 4400 RPM) So the 2 degree reduction added power only just before boost.



I think the biggest trick will be getting the attack/decay rates just right. ANd knowing the system was designed for KR and not anti-lag, it may never work as I like. But it does have its good moments.

Last log had high attack and decay, spark was jumping (no more then 2 degrees max) all over. Proving how tricky this may be to pull of reliably.





Programming anti-lag system to intentionally trigger the KR sensor? Possible, though I have no idea the complications that could arise.

[Z26-T]

If so, have you tried damping material on the knock sensor? Sometimes they are a little over-sensitive to the boosted whirring, whining and even loud exhaust noise.

I have false KR whenever my turbo is spooling.. I've tried different damping materials, different knock sensors.. no change on my system. Gave up on it for now and just run with the sensor unplugged.

TGP, do you have a .bin I could look at to see what you did with your spark tables?

[Dave96z34]

I get false kr from the 284 getrag at low low rpms under load on my scans. The trans makes a noise and it triggers a false misfire and flashes the ses light aswell.

[Dave96z34]

Could try a knock module from a different engine.

[3400-95-Modified]

I get false KR and I think it comes from the sound my headers make... Unless it was picking up noise from my shot timing chain and that's what it did in the past as well... Not quite sure on that yet.

[Z26-T]

Yeah, I tried an 01 knock module with matching sensor and then a 96 module with matching sensor. No change thus far. May try to see if a module from a obd1.5 module will fit in a obd2 module and try that with the matching sensor.

[TGP37]

I have false KR whenever my turbo is spooling.. I've tried different damping materials, different knock sensors.. no change on my system. Gave up on it for now and just run with the sensor unplugged.

TGP, do you have a .bin I could look at to see what you did with your spark tables?

Sure do, I'll get them up soon. If you are curious where to reduce spark during spool, it is around 0.44-0.48 g/cyl area. Spark map is pretty much standard until 0.4 g/cyl where it dips and rolls into spark for boost.




At the moment, I have all knock disabled until 4000 rpm. This way the spool doesn't trigger false knock but still having knock protection in boost. I could nudge it down to 3600 but a few times the spool is caught and pulls timing when flying down hill.

If you suffer from false knock from turbo spool, this does help a little. Or you can reduce the max timing pulled vs RPM. But I found sometimes the false KR pulling only 1 degree can extend into higher RPM from the decay rate. As long as the spark below 0.52g/cyl isn't too aggressive it should be okay.

Pulling timing with the KR sensor is too chaotic. It works somewhat for spooling but it isn't stable and that alone makes it hard to tune. So that is where I am at.

I am sure I can come up with better antilag system.

[TGP37]

If I understand this correctly, you have effectively increased the potential spark advance while spooling, and the result is more airflow while spooling.

If you had no false knock, do you think you could program an increase in the timing while spooling? If so, have you tried damping material on the knock sensor? Sometimes they are a little over-sensitive to the boosted whirring, whining and even loud exhaust noise.

It reduces the spark, which dumps more of the heat energy onto the turbine instead of the piston.

The airflow itself isn't higher. The rate of change was higher. The increase rate of airflow was faster, by a small margin.