Notices
ECU Flash

Don't think it's the tuning - trying to find the culprit

Thread Tools
 
Search this Thread
 
Old Oct 10, 2006 | 06:24 AM
  #1  
kapolani's Avatar
Thread Starter
Evolving Member
iTrader: (6)
 
Joined: Feb 2004
Posts: 442
Likes: 0
From: Maryland
Don't think it's the tuning - trying to find the culprit

I started this thread earlier because I've been datalogging my car and noticed some strange onset of knock.

See this thread for previous discussion:

https://www.evolutionm.net/forums/sh...d.php?t=227401

What has me confused is the knock occurs RANDOMLY. All log runs are done back to back - same conditions - same stretch of road - in 3rd gear.

I would think if something were definitively wrong with the tune - I would be able to induce the knock with every single pull.

I've searched and read posts from many tuners and what I have gathered is this:

Stiffys, aftermarket motormounts, exhaust resonance (rubbing), the WRONG TYPE of spark plugs, mechanical differences in chamber cc,
coolant variations, IAT temp variations, carbon buildup, excessive EGT recirculation, improper fuel distribution....alot more, cause or help induce phantom or REAL knock activity.

Ive done some more logging and had some very strange and troubling results. The attached logs were done on the same stretch of road - back to back - same gear.

The run labeled _High_Knock_ was really unsettling. If the tune were the culprit - and caused such excessive knock count one would think that this behavior would have exhibited itself on the other pulls.

Could it be my DP rubbing when the engine is torqued? Only caused when certain factors are happening at once...

Or, could my knock sensor be faulty and causing this extreme random behavior...

This post is not meant to create a _tuner bashing_ thread. Intelligent discussion and constructive posts are welcomed.

I admit I'm a newbie when it comes to tuning etc., but have a willingness to learn from you all here.

Thanks!
Attached Files
File Type: zip
Kapolani_EvoScanDataLog.zip (19.8 KB, 53 views)
Reply
Old Oct 10, 2006 | 08:04 AM
  #2  
justchil's Avatar
Evolved Member
iTrader: (44)
 
Joined: Oct 2003
Posts: 4,409
Likes: 0
From: Charleston, WV
I haven't read your previous post but to totally rule out the tune u could fill up with some GT100 (if you can get it) and see what the logs show. That would rule in/out gas/tune for the most part.

If knock sensors are cheap I'd say swap it out for a new one... that HIGH_KNOCK log is the most know I've ever seen posted :\
Reply
Old Oct 10, 2006 | 08:08 AM
  #3  
caLi4G63's Avatar
Evolved Member
iTrader: (40)
 
Joined: Nov 2004
Posts: 621
Likes: 8
From: so. caLi
Can you log and view timing? If so, does it change accordingly when the knocking occurs? I would try what justchil said and put some higher octance fuel in the gas tank and see how that goes.
Reply
Old Oct 10, 2006 | 09:30 AM
  #4  
kapolani's Avatar
Thread Starter
Evolving Member
iTrader: (6)
 
Joined: Feb 2004
Posts: 442
Likes: 0
From: Maryland
Originally Posted by justchil
I haven't read your previous post but to totally rule out the tune u could fill up with some GT100 (if you can get it) and see what the logs show. That would rule in/out gas/tune for the most part.

If knock sensors are cheap I'd say swap it out for a new one... that HIGH_KNOCK log is the most know I've ever seen posted :\
You know what?

I've given others this advice as well.

I don't know how I've forgotten about it. I CAN get 'race gas' in my area.

I'm going to pick up a few gallons and run that then do some datalogs.

Most - if not all traces - of knock should be non-existant when using the high octane fuel.

Originally Posted by caLi4G63
Can you log and view timing? If so, does it change accordingly when the knocking occurs? I would try what justchil said and put some higher octance fuel in the gas tank and see how that goes.
If you look at the _high_knock_run_ you can see that timing AND knock count is logged. I admit that I'm not well versed in reading logs, but I do see that a lot of timing is being pulled.
Reply
Old Oct 10, 2006 | 09:42 AM
  #5  
kapolani's Avatar
Thread Starter
Evolving Member
iTrader: (6)
 
Joined: Feb 2004
Posts: 442
Likes: 0
From: Maryland
Originally Posted by justchil
that HIGH_KNOCK log is the most know I've ever seen posted :\
When I saw that log I was pretty amazed as well. I have never logged one as high as that.

But, if you look at the run right before it all is well.

AND - the two runs after it are perfectly fine too.

Not sure what the heck is going on....

Last edited by kapolani; Oct 10, 2006 at 10:47 AM.
Reply
Old Oct 10, 2006 | 09:57 AM
  #6  
justchil's Avatar
Evolved Member
iTrader: (44)
 
Joined: Oct 2003
Posts: 4,409
Likes: 0
From: Charleston, WV
I've read some stuff here and there where a knock sensor wire was damaged so you may want to give that a look just in case. (shot in the dark)

Last edited by justchil; Oct 10, 2006 at 10:09 AM.
Reply
Old Oct 10, 2006 | 10:26 AM
  #7  
mchuang's Avatar
Evolved Member
iTrader: (11)
 
Joined: Sep 2005
Posts: 2,180
Likes: 1
From: h town
Maybe you have a faulty injector. If the higher octane dont fix it then go to get your injectors flow tested.
Reply
Old Oct 10, 2006 | 11:00 AM
  #8  
DynoFlash's Avatar
Account Disabled
iTrader: (91)
 
Joined: Sep 2003
Posts: 16,850
Likes: 0
From: 2003 Evo VIII - Silver
A/F should also be logged obviously

In the past I have also seen bad injectors that radomly heated up and went lean resulting in this kind of extreme knock which comes and goes
Reply
Old Oct 10, 2006 | 11:32 AM
  #9  
RazorLab's Avatar
EvoM Guru
20 Year Member
Liked
Loved
Community Favorite
iTrader: (8)
 
Joined: Aug 2003
Posts: 14,094
Likes: 1,092
From: Mid-Hudson, NY
In all honesty, with 8* advance at peak torque you are probably on the edge or over MBTT. Which is probably why the knock isn't consistant. Some pulls everything is nice and cool or feeling great, clean pull, next pull things are a little hotter, or you are on a incline, or other factors and it knocks and pull timing.

Try putting your timing at 3000 to 4500 to 4-5 degrees and see if your pulls become more consistant and knock less or not at all.
Reply
Old Oct 10, 2006 | 11:34 AM
  #10  
RazorLab's Avatar
EvoM Guru
20 Year Member
Liked
Loved
Community Favorite
iTrader: (8)
 
Joined: Aug 2003
Posts: 14,094
Likes: 1,092
From: Mid-Hudson, NY
Originally Posted by justchil
I haven't read your previous post but to totally rule out the tune u could fill up with some GT100 (if you can get it) and see what the logs show. That would rule in/out gas/tune for the most part.
How would adding 100 octane to a 93 octane tune rule out a bad tune? The tune is for 93 octane, not 100 octane. If it doesn't knock with 100 octane that means the 93 octane tune is too aggressive.
Reply
Old Oct 10, 2006 | 01:51 PM
  #11  
C6C6CH3vo's Avatar
Evolved Member
iTrader: (9)
 
Joined: Feb 2005
Posts: 4,223
Likes: 4
From: sc
The luxury of a logged A:F would be a nice thing to have.

As a troubleshooting tool, If symptoms go away once high oct fuel is used then it is likely ping
Reply
Old Oct 10, 2006 | 01:59 PM
  #12  
C6C6CH3vo's Avatar
Evolved Member
iTrader: (9)
 
Joined: Feb 2005
Posts: 4,223
Likes: 4
From: sc
Originally Posted by razorlab
In all honesty, with 8* advance at peak torque you are probably on the edge or over MBTT. .
So it is confirmed 8* at peak is too much mechanical strain?

I'd like to know cuz if so, I'll need to back up a little. It's not pinging or anything and 8* keeps things quite, but if it is proven that it's too much at less than 400whp then it would be good to know.

THanks

I just looked at the logs and in one you had counts at low throttle, one with no knock at all, another at peak, and another etc. Looks like it could be anything, ie, fuel starvation, enging or drivetrain noise, or heat + peak timing. Who knows throw some high oct in there and see what happenes

Last edited by C6C6CH3vo; Oct 10, 2006 at 02:07 PM.
Reply
Old Oct 10, 2006 | 02:17 PM
  #13  
justchil's Avatar
Evolved Member
iTrader: (44)
 
Joined: Oct 2003
Posts: 4,409
Likes: 0
From: Charleston, WV
Originally Posted by razorlab
How would adding 100 octane to a 93 octane tune rule out a bad tune? The tune is for 93 octane, not 100 octane. If it doesn't knock with 100 octane that means the 93 octane tune is too aggressive.
You just answered your own question I'm not saying the tune is bad but if you're getting knock and you up the octane and it still knocks then you may have issues other than the tune to deal with.. if not then like you said.. it's probably a bit aggressive.
Reply
Old Oct 10, 2006 | 02:19 PM
  #14  
KOEvo's Avatar
Evolved Member
iTrader: (5)
 
Joined: Aug 2005
Posts: 1,878
Likes: 0
From: Florida
It appears in almost every log that you are pulling at least some timing, so 8-9* between 3-4k rpms just isnt cool in your case.

The high knock run has to be a mechanical problem, that wouldn't just happen out of nowhere. Phantom knock was going to be my guess in the first thread, but the counts weren't high enough in those logs to rule out bad tuning; now this kind of solidifies it for us.
Reply
Old Oct 10, 2006 | 02:46 PM
  #15  
DynoFlash's Avatar
Account Disabled
iTrader: (91)
 
Joined: Sep 2003
Posts: 16,850
Likes: 0
From: 2003 Evo VIII - Silver
Originally Posted by KOEvo
It appears in almost every log that you are pulling at least some timing, so 8-9* between 3-4k rpms just isnt cool in your case.

The high knock run has to be a mechanical problem, that wouldn't just happen out of nowhere. Phantom knock was going to be my guess in the first thread, but the counts weren't high enough in those logs to rule out bad tuning; now this kind of solidifies it for us.
In the logs I was reviewing the knock was starting at 2750 rpms and in one log it hit 23 knock count there and held steady knocking all the way through the power band


The interesting thing to note is that at that exact map location the STOCK TIMING is slightly HIGHER than the flash tuned timing so razorlab's theory is not accurate

This knock is starting OFF boost or at a very low boost at 2700 the turbo is just starting to spool



2656.25 89.80392157 0 8
2750 100 7 5
2875 100 23 0
2968.75 100 24 -2
3156.25 100 25 -2
3281.25 100 25 -2
3437.5 100 25 -2
3625 100 26 -2
3781.25 100 29 -3
3937.5 100 29 -3
4062.5 100 28 -3
4218.75 100 28 -3
4343.75 100 28 -3
4468.75 100 28 -3
4625 100 27 -3
4750 100 27 -3
4906.25 100 27 -3
5062.5 100 26 -2
5187.5 100 26 -2
5312.5 100 26 -1
5437.5 100 26 -1
5593.75 100 25 -1
5718.75 100 25 -1
5843.75 100 25 -1
5968.75 100 24 -1
6062.5 100 24 -1
6187.5 100 24 0
6281.25 100 24 0
6375 100 23 2
6531.25 100 23 2
6625 100 23 3
6718.75 100 23 3
6812.5 100 22 3
6937.5 100 22 4
7000 100 22 5
7125 100 21 5
7218.75 100 21 8
7312.5 100 21 8
7406.25 100 21 10
7531.25 100 20 11
7593.75 90.19607843 20 11
7406.25 20 3 37





The condition is very random and only happens in 1 out of 4 passes

It is unusual and could be one of several factors

I have sent the customer a diagnostic ROM to use to try and rule out some factors and further isolate the condition

One more point - when I was tuning this car, I never saw any knock at all at any time

Last edited by DynoFlash; Oct 10, 2006 at 02:51 PM.
Reply



All times are GMT -7. The time now is 07:56 AM.