Transient ignition and knock count

[brn12345]

So after testing out various fuels to see the impact on IAM and Knock I found that the local premium fuel we have still makes the car knock and hence I end up with a low IAM (0.2 to 0.3). I also figured out that on full throttle there was minimal to no knock using the local premium fuel. In a bid to find out where the knock was being detected I did long logs (15min plus) of point A to point B and tried to include most driving conditions. As a picture is worth a thousand words, I attach a picture of a log I did with BRZedit using the custom feature. I looked at RPM and Engine Load as the axis and Sum of Knock Correction to populate the table. As you can see it is very clear that the problem area is in the 2250-3000 RPM and 0.6-0.7 Grev. So I started looking at map tables to try and see where this problem could be resolved and I found the Transient Igntion Retard table (attached). I went ahead and circled in green the area on this table that corresponds to the "knocky" area on the log and as is clear there is no transient retard being applied.


My next test will be to increase the Transient Ignition Retard in the 0.6 to 0.7 Grev range from 0 to 3200 RPM to -7 degrees. I will then log a similar log and report.


Comments would be great especially if you have tried something similar, I know a few more proficient guys on here have tried to resolve the knock issue before.

[jamesm]

Oddly I tried exactly that yesterday, though I only retarded it 5 degrees. It had no effect on the knock situation. I've spent more time than I should have driving around with det cans on and I've never actually heard knock while watching it pull timing at low load. Also, timing adjustments in that area have no repeatable effect.

If you figure it out I'll buy you a case of beer, but I'm guessing it's either phantom knock or pekc.

[wootwoot]

Keep up the good work!

[steve99]

Originally Posted by jamesm

Oddly I tried exactly that yesterday, though I only retarded it 5 degrees. It had no effect on the knock situation. I've spent more time than I should have driving around with det cans on and I've never actually heard knock while watching it pull timing at low load. Also, timing adjustments in that area have no repeatable effect.

If you figure it out I'll buy you a case of beer, but I'm guessing it's either phantom knock or pekc.

Probably a noob question

I am assuming the transient knock table pulls timing when their is a sharp transient in revs/load/throttle position.

However I have noticed on other cars knock at low rev under light load constant throttle, if this was the case here would you then need to change a different timing table possibly base timing in this troublesome area.

I have heard knock on my BRZ under low load light throttle occasionally (stock mapping, waiting for OFT to arrive) althought it difficult to hear.

[jamesm]

Originally Posted by steve99

Probably a noob question

I am assuming the transient knock table pulls timing when their is a sharp transient in revs/load/throttle position.

However I have noticed on other cars knock at low rev under light load constant throttle, if this was the case here would you then need to change a different timing table possibly base timing in this troublesome area.

I have heard knock on my BRZ under low load light throttle occasionally (stock mapping, waiting for OFT to arrive) althought it difficult to hear.

yes, if you had knock under stead throttle or load you would take timing out of the base or advance, depending on your preference.

[ztan]

Originally Posted by brn12345

So after testing out various fuels to see the impact on IAM and Knock I found that the local premium fuel we have still makes the car knock and hence I end up with a low IAM (0.2 to 0.3). I also figured out that on full throttle there was minimal to no knock using the local premium fuel. In a bid to find out where the knock was being detected I did long logs (15min plus) of point A to point B and tried to include most driving conditions. As a picture is worth a thousand words, I attach a picture of a log I did with BRZedit using the custom feature. I looked at RPM and Engine Load as the axis and Sum of Knock Correction to populate the table. As you can see it is very clear that the problem area is in the 2250-3000 RPM and 0.6-0.7 Grev. So I started looking at map tables to try and see where this problem could be resolved and I found the Transient Igntion Retard table (attached). I went ahead and circled in green the area on this table that corresponds to the "knocky" area on the log and as is clear there is no transient retard being applied.


My next test will be to increase the Transient Ignition Retard in the 0.6 to 0.7 Grev range from 0 to 3200 RPM to -7 degrees. I will then log a similar log and report.


Comments would be great especially if you have tried something similar, I know a few more proficient guys on here have tried to resolve the knock issue before.

Will be very keen to know what you find - I was about to try this also.
I've been getting quite a lot of knock in the same range (2400-4000) at loads from 0.6-1.0.

Seems to be temperature dependent, what kind of heat do you get in Bahrain?? Around 25C ambient, I get FBKC, but my IAM doesn't drop, between 30-35C ambient, I'll get FBKC with IAM dropping occasionally. Besides monitoring FBKC, also look at the range where FLKC or IAM drop as FBKC won't register if KC is in fine or rough learning mode.

[brn12345]

Originally Posted by ztan

Will be very keen to know what you find - I was about to try this also.
I've been getting quite a lot of knock in the same range (2400-4000) at loads from 0.6-1.0.

Seems to be temperature dependent, what kind of heat do you get in Bahrain?? Around 25C ambient, I get FBKC, but my IAM doesn't drop, between 30-35C ambient, I'll get FBKC with IAM dropping occasionally. Besides monitoring FBKC, also look at the range where FLKC or IAM drop as FBKC won't register if KC is in fine or rough learning mode.

Right now its 14-16C ambient and I do agree with you that there is a great deal of temperature dependent variables in the mapping (I have a feeling some not defined yet). Having said that though this particular knock in the 0.6/0.7 range has been with me from day one in summer and winter. It is greatly reduced when I run VP Motorsport 103 fuel but is still present! I am just starting to get my head around the knock control process with these ecu's (coming from Honda where the knock sensor usually goes in the bin) but I can say I am getting the hang of it. It still baffles me why IAM suddenly drops sometimes. Anyways, I just installed headers (see my other thread in the engine subforum) and have other issues to deal with before I can go back to testing this theory.

[steve99]

Originally Posted by ztan

Will be very keen to know what you find - I was about to try this also.
I've been getting quite a lot of knock in the same range (2400-4000) at loads from 0.6-1.0.

Seems to be temperature dependent, what kind of heat do you get in Bahrain?? Around 25C ambient, I get FBKC, but my IAM doesn't drop, between 30-35C ambient, I'll get FBKC with IAM dropping occasionally. Besides monitoring FBKC, also look at the range where FLKC or IAM drop as FBKC won't register if KC is in fine or rough learning mode.

Was looking at the Ignition timing compensation tables , it appears that the IAT timing compensation is disabled at engine loads between 0.6 and 1.0 at engine RPM between 3000-4000 and somewhat limited at 2000-3000.

Might be worth looking in this area

I often see IAT climbing above 50 C (120F) on 35 C days when in traffic.

[jamesm]

that's interesting, i haven't noticed a correlation with intake air temps but i'll look at the logs tonight and see if i notice anything. i'm still experimenting here with no real success.

[brn12345]

[QUOTE=steve99;1420589]Was looking at the Ignition timing compensation tables , it appears that the IAT timing compensation is disabled at engine loads between 0.6 and 1.0 at engine RPM between 3000-4000 and somewhat limited at 2000-3000.QUOTE]



Thanks! I am aware of this and I have seen IAT go above 50c in traffic too (goes up very quickly). The knock in my logs though is always with IAT below the retard threshold so it is not a factor.

[solidONE]

So what's the word on this? Were you guys able to eliminate or reduce tip in knock?

I was able to stop my IAM from getting trimmed back by reducing the KC advance from .70 g/rev and up at 800~3200 across the board by .35 degrees. Not sure if I over did it but I could not induce IAM to be reduced by running the car under mid to high loads in the lower rpm range after this simple change (with cool weather and IATs). Although the tip in knock is still present. Not enough to cause the IAM to be reduced but still there.

http://datazap.me/u/solidone/maft100...zoom=4375-6880

You might also notice my fuel trims to be mostly in the negative since I dialed this maf scale to around 100+ degrees IATs. While I did not log the temps, the IATs when this log was taken should be in the low 80's. The IAT temp compensation in the maf scale is another thing I'd like to address, but I guess that's for another thread.

[steve99]

Originally Posted by solidONE

So what's the word on this? Were you guys able to eliminate or reduce tip in knock?

I was able to stop my IAM from getting trimmed back by reducing the KC advance from .70 g/rev and up at 800~3200 across the board by .35 degrees. Not sure if I over did it but I could not induce IAM to be reduced by running the car under mid to high loads in the lower rpm range after this simple change (with cool weather and IATs). Although the tip in knock is still present. Not enough to cause the IAM to be reduced but still there.

http://datazap.me/u/solidone/maft100...zoom=4375-6880

You might also notice my fuel trims to be mostly in the negative since I dialed this maf scale to around 100+ degrees IATs. While I did not log the temps, the IATs when this log was taken should be in the low 80's. The IAT temp compensation in the maf scale is another thing I'd like to address, but I guess that's for another thread.

I pretty much did the same as you have done for Australian fuel which is rubbish compared to yours, so had to pull 0.7 and then more up to 1 at high rpm and load, did a thread on it. IAM sloid at one but still get the odd bit of tip in knock if your aggressive on throttle low rpm high gear.
Also have made the temp compensation cut in one cell earlier and added a bit to the 3000 rpm range where it was zero seems to help on hot days in traffic, it winter now so carnt test it any more

[86drift]

The knock at 0.6/0.7 isn't transient knock. Typically happens in closed loop when driving up a steady hill. Try lowering the closed loop AFRs. Pulling timing doesn't help.

[steve99]

Originally Posted by 86drift

The knock at 0.6/0.7 isn't transient knock. Typically happens in closed loop when driving up a steady hill. Try lowering the closed loop AFRs. Pulling timing doesn't help.

Thanks will try that when it warms up again united 100 ron seems to help as does cooler weather worst i was getting was 0.6 degres on 100 ron.

will try afr adjustment