PV - adaptive knock retard

[Sunny Jim]

PV Knock table
Say My Knock control is switched on and my
Adaptive Knock retard is zero'd - effectively switched off,
Can I assume my knock control/ knock count is still working correctly?
Thank you.

[hrdtail78]

PV Knock table
Say My Knock control is switched on and my
Adaptive Knock retard is zero'd - effectively switched off,
Can I assume my knock control/ knock count is still working correctly?
Thank you.

I look at these as long term and short term.  So, you shut off long term and left the short term running.  Will it work correctly?  Yes, the ECM will detect knock and react to it by pulling timing.  You will be able to see the reaction in the logs and be able to adjust your spark table based on this reaction.  The ECM will not learn or remember where this spark happened.  The ECM wont adapt or keep pulling timing in this area. It will have to see knock in order to react to it. 

I shut off long term for tuning and I believe it is recommended by tuning manufactures as well.

[Sunny Jim]

For the record - So I ran my bike yesterday with adaptive switched off.
I had spark knock control switched on along with spark knock count.
I logged data and recorded a small amount on knock above 4k @ 90kpa
But didn't record any knock count.
 

[hrdtail78]

PV puts this information out in two different data traces.  One is knockCntF and KnockCntR.  This is where adaptive is shown.  The other trace is Spark Knock F, and Spark Knock R.  This is the trace that shows actual ECM's reaction to knock detection, or short term.  How do you view your data?

[Jamie Long]

For the record - So I ran my bike yesterday with adaptive switched off.
I had spark knock control switched on along with spark knock count.
I logged data and recorded a small amount on knock above 4k @ 90kpa
But didn't record any knock count.
 

Late model Delphi CANbus ECM's store learn knock retard in eprom, this is not cleared at key off or upon reflash. The signals Front & Rear Spark Knock on the data stream represent learned knock retard the ECM has already populated & stored rather than individual events on these models. Based on what you are seeing I would expect you zero'd the adaptive knock table, if this is the case the data stream will still have the previously learned knock values in place (until they are unlearned), if you are looking to completely disable knock control you can do so under Spark>Knock Control = off   

[1stop]

So,I'm not understanding. Ft & rr knock is learned value that ecm is pulling out compared to spark table. (Revised spark table,ill call it).And knock count is knock detected that happened running with learned/revised spark advance. Jamie mentioned learned/revised table is still there after reflash. How do you reset learned knock? Looked in quick tune, only see reset adapt fuel values. Also, I use megalog viewer to view logs. I don't see knock count in logged data, just knock ft & rr. Could somebody confirm if my understanding is right, and how to rest knock learned? MLV log viewing?

[hrdtail78]

You can't reset learned knock.  It decays over time with key cycle if no other knock is detected in this area.  If you are tuning the bike.  Adaptive knock should be turned off.

[1stop]

So turn off adaptive knock while tuning. That will take away learned knock retard and run the values in spark table. Leave knock on and use what log setting to view knock events, ft & rr knock retard or knock count log settings to view knock events? Then turn adaptive on after tuning. Correct? I assume previous learned knock before tuning will become effective until allotted time to decay back to values in spark tables. Sound right? So, if you log with adaptive knock on ecm is retarding timing from set values, what can you log to see it?