PDA

View Full Version : Knock knock.... who's there? KR



Whippled 496
September 22nd, 2006, 05:19 PM
First of all I would like to thank DCJustin for all the help he has been giving me over the last few weeks with helping me understand getting around in EFILive. So let me give you the quick run down.

My Truck:
2003 2500HD 4x4
8.1L/Allison
WHipple Supercharged @ 7psi non-intercooled
No Whipple electronics
42lb/hr marine injectors
cool plugs gapped at .035
160* thermostat

I can not for the life of me get the knock to go away. It is there at low rpm, low load, and everywhere in between. We have tried lowering the timing to as much as 3* and I still get 3*-5* of timing being pulled. I have tried changing gas and using octane booster and it has not helped. I know my IATs are a little high without the I/C, but when its pulling 4* and i only have a total of 3* in there to begin with, something is wrong. The one thing that crossed my mind was the fact that i have an Allison trany, and I was told a while back that the Allison TCM commands it own torque management seperate from that of the PCM. Well when i opened the TCM file, i found nothing referencing TM in the parameters. I figured that maybe the timing that was being pulled was from the Allison TCM, but now I dont know. I even drove beside jersey walls to see if i could hear the sound of knocking or pinging from outside my truck...i heard nothing. Even when it says its pulling 5* of timing i hear nothing at all. I am at a loss as to what could be causing this and like "O2" said, 100 people looking at it is better than 1 or 2, so i figured i would post up my tunes, logs and Allison file for you all to look over if and when you get time. AFR seems right on the money too. Well here is a zip file with some tunes and logs associated with those tunes as well as the stock Allison TCM tune. You will need EFILive v7.3 to open the Allison tune i think. Look how at even part throttle, low load conditions, i still get knock. I gave the logs the same or similar names as the corrosponding tune it came from. If anything jumps out at ya, let me know. Thanks a lot.

http://www.pasadenatownhall.com/tunes.zip

Tordne
September 22nd, 2006, 05:41 PM
Not knowing about trucks at all, cause we don't get them down here really, I'm working on the assumption that this is not a DMAX and has the same tables as "normal" calibrations...

I've experimented A LOT with various things in my early days to try and identify and remove Knock. My findings are basically:
1) The Burst Knock Retard sux big time. I have experimented with various settings and have settled on 0'ing the B6210 & B6212 tables to stop it doing anything. I ALWAYS find that if enable this I always log more knock that without it, which ironically is the exact opposite of its purpose/intention.

2) The PCV system on cars vents oil vapor from the crankcase into the intake manifold which dilutes the charge and causes Knock also. I installed a PCV Catch Can to reduce the effects of this.

3) Running a heat range colder spark plugs also reduced Knock, especially on tip in.

I guess you need to be careful being FI.. But these few things basically eliminated all my Knock dead!

Good luck :)

Whippled 496
September 23rd, 2006, 02:18 AM
I dont have a B6210 or a B6212 table, maybe those are custom OS tables? I have no PCV valve on my truck its integrated into the intake manifold on the 8.1L's and I am running one range colder AC Delco plugs. My burst knock table is also zero'd out.

Whippled 496
September 23rd, 2006, 02:53 AM
I see reference to "torque reduction mode", or "TRQ Red" in the "shift timing" parameters of the Allison file. I dont see anyplace to reduce or eliminate TRQ Red anyplace though. Anyone have any insight to where the adjustment for torque management is in the Allison tune?

SinisterSS
September 23rd, 2006, 02:59 AM
For more data samples per second, limit the number of channels in the scan to 24 or less. Below is a screenshot of the PIDs from the last test run with all the miscellaneous stuff removed.

For a blown big-block, the timing seems a big aggressive. Try the attached tune below and see if you still get KR. Starting low and working your way higher is easier and safer than starting high and working low.

The more accurate your VE table is, the less chance for KR as well. It's apparent you've been working on your VE under boost. How accurate is the VE in the rest of the table?

On the 8.1 is there a hose routing the crankcase gases back into the intake or throttlebody? If so, Tordne's suggestion of a catch can is a very good idea. On my LS6, adding a can took care of random KR.

Whippled 496
September 23rd, 2006, 03:54 AM
Thanks Sinister, i will give this tune a try and do some logging. Thank you very much.

Tordne
September 23rd, 2006, 06:30 AM
I dont have a B6210 or a B6212 table, maybe those are custom OS tables? I have no PCV valve on my truck its integrated into the intake manifold on the 8.1L's and I am running one range colder AC Delco plugs. My burst knock table is also zero'd out.

LOL, well there goes all my suggestions :frown: BTW, those tables I mentioned are in the stock OS for F-Body, Y-Body etc.

ringram
September 23rd, 2006, 08:10 AM
Try commanded fans on earlier for cooler ECT too, that helps a lot.
I run 80*C Stock is like 95*C

Whippled 496
September 23rd, 2006, 10:15 AM
I run a constant ECT of 167*. That should be plenty cool enough.

Whippled 496
September 23rd, 2006, 10:25 AM
Hey Sinister, i tried that tune out and the KR went nuts! I couldnt tip in with out pegging my KR limit so I had to reflash it. Here is the log from that tune, the only thing i did was raise the knock limit to 8*. Its short because i was afraid to run it to long.

jfpilla
September 23rd, 2006, 12:35 PM
I'm seeing -40 ect in your log. Is that correct? If so it could be adding timing. My file would add 8* at -40.

Whippled 496
September 23rd, 2006, 01:03 PM
sorry JFpilla, i was experimenting with my PIDs and I forgot to enable ECTs i think. I definately did not have ECT temps of -40*

SinisterSS
September 23rd, 2006, 01:45 PM
Looks like your original tune have have been running on the low octane table. In test 3, the PE fueling comes in more quickly to help with the tip-in knock.

jfpilla
September 23rd, 2006, 01:47 PM
I didn't think so in MD. I as thinking something was reporting wrong and acting like it was -40. I guess not. With the knock occuring at low timing levels it could be false knock?
I lived in Randallstown when they first started on the Harbor project. We were there a month ago. What a great change. They've lowered the quality of the crabcakes though.

SinisterSS
September 23rd, 2006, 01:53 PM
It could very well be false knock. If it is still getting KR with version test_3, there's something afoot.

Whippled 496
September 23rd, 2006, 02:05 PM
I will give it a go tomorrow morning Sinister, as long as there is no rain. Thats some serious false knock if thats what it is.

dc_justin
September 24th, 2006, 01:10 AM
It could very well be false knock. If it is still getting KR with version test_3, there's something afoot.

Exactly the line of thinking we've come to over the past couple of weeks... His original tune had the IAT spark table zero'd out. :Eyecrazy: Pegging the KR limit in boost...

Next version I added in a correction table for him, timing dropped significantly as a result... KR still existed. He changed to higher quality gas and added octane booster... same result.

Whippled... if I end up moving back to this area as a result of what happened this weekend, we'll definitely have to meet up. :rockon:

Whippled 496
September 24th, 2006, 01:50 AM
Whippled... if I end up moving back to this area as a result of what happened this weekend, we'll definitely have to meet up. :rockon:


As a result of what happened? Are you kidding? I set the whole thing up just so you would. It cost me a lot of money, but its the only way i knew of to get this tune fixed....:muahaha:.

Im getting ready to go try this tune out, but like Justin said, he gave me a tune very similar to this, only difference was he had timing being pulled through the IAT table like he said. So in a sense, it was even more conservative than this one. That wont stop me from trying and logging it though. I will report back when i got something. Thanks guys.

Whippled 496
September 24th, 2006, 01:54 AM
Looking over this tune i see you made a few other changes. Looks like you have reduced timing through the entire RPM range and you also decreased the PE mode enable down to 25% TP in some places. HMMMM, is that so it goes into PE mode during those low load, low RPM situations?

dc_justin
September 24th, 2006, 03:00 AM
As a result of what happened? Are you kidding? As a result of the g/f happenings of Friday :)What are you seeing for KR with the 8 degree limit now?

Whippled 496
September 24th, 2006, 03:19 AM
It was going higher than 5*, but i dont know if they ever hit 8* or not. I think it did once. Here is a real long log (18 minutes) of some varied driving. This was with a different tune with WOT timing of 14* (its the one i been using for a while). You can see that the KR comes in most of the time around 2200-2500rpm and or 1.10-1.20 g/cyl. Not sure if that means anything, im just looking for a pattern of some sort. This is not the one Sinister just sent me, i have not got out to test that one yet. I will soon though.

Whippled 496
September 24th, 2006, 06:15 AM
okay Sinister, loads of KR on tune_3 as well. Here are the logs. You know whats crazy, is i loaded up an older tune that i had done when the intercooler was on. Its more aggressive in the spark table than the one you gave me and it yeilded much less knock. I am going topost those logs and tune in a second. Here are the logs of the last tune you gave me.

Whippled 496
September 24th, 2006, 06:19 AM
Here is the latest tune i have loaded up and a few logs from it. Its spark table is more agressive and it resulted in LESS knock than the other one. I am getting very confused now. I am just learning how to tune, but for Gods sake this is making me nuts. Here is the tune and the logs from it....

Whippled 496
September 24th, 2006, 08:15 AM
Looking over some other tunes i noticed a few things that may or may not have to do with anything but i figured i would ask.

The first thing is table B6227 (knock sensor filter). I have read that GM uses frequencys in the 4-5kHz range for thier knock sensors. All the other tunes I have looked at have the cells filled with 4.70. My cells are filled with 7.90, what effect would this have? Is the raising of that number actually filtering more outside noise -OR- causing it to hear more (knock) than it should?

Second thing. Can the VE table have anything to do with knock, or can the data being incorrect in those cells effect KR? I ask because in looking at table B0101 (main VE table), and when the "UNITS" (cells) are populated with "%" instead of g/cyl, there is a dip at 95kPa between 2800-4000rpm. I dont see this dip in any other tunes and am wondering why it would be there. You can see these things in that last tune i posted. I got my 2bar MAP sensor, but i dont want to go F'in with that until i get my knock cleared up.

Thanks guys

Whippled 496
September 24th, 2006, 09:59 PM
glad im not the only one stumped here. Another thing i noticed in the logs is i have some times right after WOT where my AFR spikes (over 20 sometimes). What would cause that?