rcwilbur

I have Texas Speed 5.3 heads and 224R cam in LS1 Corvette with Manual transmission. AFR is 12.5 at WOT and I'm getting a lot of knock retard. PCM is taking away as much as 10 degrees timing. I started with stock tune which had about 26 degrees of timing at WOT and have steadily reduced timing in about five different steps. Last run I set timing to 16 degrees at WOT. It doesn't seem to matter what the timing map is set at, the knock retard is always removing about the same thing 10 degrees.

I'm wondering whether this might be false knock. I've looked through the tuning repository on HPTuners and it looks like some people modify the Tip In TPS table, and the Knock Sensor Level table. Could this be just a matter of the knock sensors being too sensitive?

jimmyblue

iTrader: (11)

Could be. But rather than speculating, look at what
precedes knock-detects in the logs. If you see a
lean dip on the NBO2 sensors then don't blame a
false knock when something's there that will cause
a real one.

Halve your attack rate and double your decay rate,
and you can get a much better idea of lower level
activity (less cover-up).

rcwilbur

Thanks very much. I pretty much concluded yesterday that it was false knock. There were no lean spikes and I tried making AFR very rich 11.5, all with no effect.

When it was producing knock retard it would be a single spike during WOT and then long decay. I looked at the Ls6 knock sensor settings and noticed that they are different than my 99 LS1. Specifically, they have higher Tip In vs. TPS settings and the Cylinder Knock Levels are a little higer.

I started with timing that I assumed would be safe (about 21). I then copied the LS6 tables I metnioned above into my knock sensor settings. I ended up cutting the attack by about 15% and increasing the decay by 50%. I then increased the Global Sensor Gain for 3600 RPM and up by 10% (went from 1.0 to 1.10).

After all of this the knock retard is virtually gone ( I get a fraction of a degree once in a while). I don't hear any knock and everything seems to be good. It doesn't seem that I did anything that drastic, but I am very new to this.

I didn't know enough about the attack rate to feel comfortable halving it, but will remember that for the future.

rcwilbur

I wrote the above from memory. Here is more specifically what I changed.

Hope I've done everything okay--I made minor tweaks to:

1) the Tip In vs TPS Map (Used the LS6 map);
2) knock sensor level vs. RPM vs. Cylinder Map (used the LS6 map);
3) knock attack table;
4) knock decay table; and
5) a small adjustment to the Knock Global Sensor Gain Table above 65Kpa (went from 1.00 to 1.15).

Thanks very much for your help.

fire67bird

I know this is an ancient thread but I wanted to say thanks for posting.

I have a turbo Silverado that was bogging down during WOT shifts and I couldn't figure it out. Turns out it was a rich condition during/after shifts that was being caused by false knock retard that was throwing my timing way back. The timing was so bad it couldn't burn the fuel. I did what you said to do and now my truck is running better than it ever has. It's a 4.3 and has never been able to break the tires loose on dry pavement but with this correction it's doing it easily.