Knock Jump
 

I have gathered and reviewed a ton of data logged from my 2 recent track days. As I am trying to sort out and fine tune my mini 'build' I have found some scary knock values logged only on track. Any home running doing 3rd and 4th gear pulls does not produce this. Because of the TPS value I wasn't immediately concerned with the values, but seeing how consistent I produce this type of knock on track, I wanted to spend some time troubleshooting and getting some input. Is this just some mechanical noise causing the knock sensor to go nuts abruptly, or is this an indicator of something wrong with my tune. What do you think, and why.

I have dozens, maybe hundreds of examples, but I'll just provide this one for now, in case someone is interested in more data.

https://dl.dropbox.com/u/41298765/Ev...llow080411.jpg

im interested to see input on this. i havent logged my car on the track but...

i believe the MAX count is 10? so to go from zero to max with real knock is a little odd to me, although i have seen my car do it when lifting at high rpm on regular third gear pulls, traced it on the map and took a degree out to be careful. i also have suspicions that it was my driveshaft snapping around on lift too...

Max knock sum is 36.

well then...dunno where i got that from lol.

That driveshaft comment is interesting. I am suspecting some very mechanically loud noise that the knock sensor is picking up, but on track, at speed, I can't here it. I was thinking exhaust, clutch, new built trans & FD, etc. I also did the BS delete, and Cosworth Springs for my 272s.....

However years ago I did a tuning/troubleshooting session at Road Race Engineering when I first got the car (salvage title) and there was an odd and violent bucking when just getting off throttle. We never figured out what that was, and I didn't have that occurrence in my logs again. However, I never logged track days so thoroughly as recently, so I don't know for sure if I had it before my build and didn't know it, or just created with this build.

Here is another example of higher knock values, and I have some with 30s too. :confused:
https://dl.dropbox.com/u/41298765/Ev...dway111056.jpg

I also have, what I believe is the more common 'Shift Knock'. Here you can see it and the TPS goes to 13% as I am likely shifting. Not sure this type of data would relate to the above, but just in case. I don't have any other kind of knock that is material (0,1, and 2s for high load and WOT).

https://dl.dropbox.com/u/41298765/Ev...ShiftKnock.jpg

those 8x 10 couns of knock, what was the time interval between them all? .2 seconds, 1 second 5 seconds?

my understanding is that an Evo IX pulls 1* on 1 count of knock and then another 1* every additional 3 counts which correlates to your 5* drop from 22* to 17* @ 10 counts of knock (roughly)

also notice that when it hits 15* it holds that timing. so it could also possibly be that although there are 8x 10 knock counts, there may be a delay in logging to get it back to showing 0 knock hence why it is not continually pulling timing and actually start to increase again.

cheers

are you hitting negative timing values at 50% tps and 17psi due to timing being pulled? that almost seems like why your count would jump to me...

two things greatly contributed to my jerky behavior on lift off. prop shaft bushings, and adjusting the clutch.

I don't have negative values in my timing map for those conditions. I also don't have negative values in the logs when the jump of knock comes in. Its a bit of chicken or egg thing, but in the OP you can see postive timing values and then boom 10 counts of knock.

I believe I have the clutch adjusted perfectly, finally. however, could you explain more about that comment. What adjustment exactly impacted your lift off driving behavior.

And, what exactly about the prop shafts. I am curious about troubleshooting that one. I did blow up my TC last year, and it was pretty violent. I just don't know what or how I need to check those bushings. When under the car it all seems correct and tight, etc.

For your shift knock:
Your timing rise to 22* at 5000rpm 130load. It surely doesn't help.
Try lowering the timing on lift off, I use this with success:

http://i262.photobucket.com/albums/i...iftknock-1.png

For your knock at part throttle, it seems to be a little rich : 10.6 at 50% TPS. It's because your BOV seems to hold boost until 48% TPS... How is tuned your TPS WGDC correction table? This one is very important for reducing part throttle bucking and helping the BOV. Something like that, and even lower depending on your BOV/turbo:

http://i262.photobucket.com/albums/ii118/domyz/wgdc.jpg

Knock could be something else, but there is no harm setting it right.

I think I understand this and decided to work on my timing map to not only give some low end TQ back, but smooth this area of my map. When I did this my map starting looking weird and breaking the rule of oscillating in one column. As I might have higher values in the 130 load column at lower RPMs than I would at higher RPMs because I was trying to do what you noted - smoothing the lift off transition. So going down the map in the same column as RPMs rise I was lowering timing to deal with the movement through the map due to lift off, (right to left)

The extra timing woke up the car, but I then logged cruising on the freeway to make sure it was safe before I did some pulls. Yikes. I had knock popping up just at cruise.

I am going nuts with this. I can't really fine tune for power without knowing what is going on with the knock sensor.

I don't have this table, and don't have tephra Rom. Do you know what other name it would be. I don't know anything about that yet. (add that to a rather long list)

Can anyone expand on where I would find this table in a non tephra ROM, or how it's named? also any more detail on how it should work as domyz implies I need to tune it. I suspect it's stock, the turbo and BOV are stock 9mr as well.

They're part tephra's added code for boost control.

[QUOTE=tephra;7597106]TephraMod V7

[...]

FEATURES

[...]

TPS compensation for WGDC
- Basic multiplication table for WGDC

[...]

100% WGDEC under a certain load
- Keeps BCS from openning wastegate under a certain load (to aid in spool up)

[...]