Notices
ECU Flash

Need some tooning help to rid knock.

Thread Tools
 
Search this Thread
 
Old Apr 2, 2008 | 06:44 AM
  #1  
tkklemann's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: Jul 2005
Posts: 1,228
Likes: 0
From: Charleston, SC
Need some tooning help to rid knock.

Hey guys,

I recently installed a heavily ported IX turbo on my 05, and have been having lots of low load/low rpm/part throttle knock that I simply can't seem to get rid of. FWIW, the wastegate area was ported out, the exhaust outlet was ported, the inlet was ported in sneaky ways, and the compressor housing was ported out in sneaky ways. (And until I get the knock to go away, the turbo sucks! Under WOT, it friggen rocks!) The wastegate also has a spring contraption I came up with that allows me to hold 24-24.5 psi by redline (Don't think, but this might be contributing to the problem) Spool up is rediculous though, Just pulling away from a stoplight with 20-25% throttle sometimes nets me 2psi only going 9mph.

The problem is soo repetitive, and that's why I am worried about it. The slow knock eating away at my motor evertime I pull away from a stoplight.

Anyhow, attached is a log that shows a few of the knock events I am talking about. Here are the cliff notes of the knock events:

Always between 2500-4000rpm
Always from a light start (Pull away from a stoplight)
Knock if usually between 1-8 counts (And yeah, Octane No. goes down sometimes)
Boost is between -0.5 and 4 psi. (Yeah, just from pulling away from a stoplight)
TPS is always around 25-35%
Load is always between 80-115
AFR's are between 14.7 to 10.5 (Yipes, but only for now, until resolved)
Timing varies based on the knock count.

It is severely noticeable when the ECU pulls the timing. It's surprising how different 3* vs. 8* is when taking a light start from a stoplight.

I have since re-scaled my load columns to add in resolution in that area, like the following shows, to try and gain more precise control over this area:

20
40
60
80
90
100
110
120

Here is what I have done tro try and alieviate the problem:

1. Kill timing. This didn't work very well because I eventually got down to 1-2* in that area, and the knock still would show up, pulling timing sometimes to -2*.

2. Richen the fak out of that section of the map to see if that would make it go away. As the data log shows, even with an AFR of 10.5:1, it would still knock occasionally.

3. Rescale the map's as shown above. The problem isn't as frequent now, but it's still there as show in the attached data log.

4. Speed independent. Makes sense..

5. Playing with my Open Loop Load/throttle Maps now. From the data log, you can see in about the middle of the knock event, the O2 trim goes to 100, showing that the car, in the middle of the knock event, starts to go off the maps. I am trying to change this by playing with the Open Loop Load and Throttle Low Maps to see if making the car transition into open loop earlier helps.

The attached data log shows a few of the knock instances. I deleted all the other usless areas between the knock like idling, and crusing areas. The areas are separated by a blank line.

Anyhow, what do you guys think? I can't seem to hit the nail directly on the head to aleviate this problem. It's driving me nuts, and makes pulling away from a stoplight smoothly next to impossible.

If anyone wants to see my actual ROM, I can send it to your email or anything. Just let me know..

Opinions, thought and advice are all greatly appreciated..
Attached Files
File Type: zip
Knock log.zip (7.7 KB, 60 views)

Last edited by tkklemann; Apr 2, 2008 at 06:47 AM.
Reply
Old Apr 2, 2008 | 08:16 AM
  #2  
logic's Avatar
Evolved Member
iTrader: (2)
 
Joined: Apr 2003
Posts: 1,022
Likes: 7
From: Berkeley, CA
The only change (other than the debugging you've been doing with timing, fuel, etc) was the addition of the turbo? It looks like it's all low-RPM, mid-shift (or from a light) noise? Perhaps you're getting mechanical/vibrational "knock" rather than actual detonation? How is shaft play on the turbo? Everything anchored well (exhaust, turbo, intake, etc)?

One good way to confirm: if your tuning looks good otherwise, check your plugs for signs of detonation...
Reply
Old Apr 2, 2008 | 08:42 AM
  #3  
tkklemann's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: Jul 2005
Posts: 1,228
Likes: 0
From: Charleston, SC
There was no shaft play in the turbo at all. My stocker had more that this one.

I have changed the plugs three times already, so I don't think that is the problem.

I checked all hardware, and it is all tight. FWIW, all new gaskets as well...
Reply
Old Apr 2, 2008 | 08:46 AM
  #4  
PureDsm's Avatar
Evolving Member
iTrader: (1)
 
Joined: Nov 2007
Posts: 435
Likes: 1
From: DMV
i feel it somthing that not tightend down properly
Reply
Old Apr 2, 2008 | 10:00 AM
  #5  
roger smith's Avatar
Evolving Member
iTrader: (4)
 
Joined: Dec 2003
Posts: 357
Likes: 0
From: Ventura County, CA
Looks like something to do with open-closed loop transitions or lean spool. Do you have lean spool on or off? My car does this same thing, but I haven't tried to fix it yet.
Reply
Old Apr 2, 2008 | 10:50 AM
  #6  
tkklemann's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: Jul 2005
Posts: 1,228
Likes: 0
From: Charleston, SC
Originally Posted by roger smith
Looks like something to do with open-closed loop transitions or lean spool. Do you have lean spool on or off? My car does this same thing, but I haven't tried to fix it yet.

Lean spool is actually on now. I had trouble with tuning the car when I disabled it initially, so I turned it back on. I am wondering if I turn it off, if the AFR at the beginning of the knock will actually go away, and run off the maps instead of being in open loop.

I have adjusted my Open Loop Load to read 65, so that it will be reading off my map when it get's into the 2500-4000rpm/80-120 load. It didn't seam to help because the O2 Trim's didn't go to 100 during that section of the log to indicate open loop operation.
Reply
Old Apr 2, 2008 | 11:10 AM
  #7  
lemmonhead's Avatar
Account Disabled
 
Joined: Nov 2006
Posts: 1,296
Likes: 2
From: wexford,pa
closed/open loop knock defi should not be causing 8 counts of knock, only 1-2 at most. Im pretty sure you got some kind of vibration problem. I would slowly increase throttle from a stop to try to make as little fast movement thrust as possible. if it dont know them something is vibrating.
Reply
Old Apr 2, 2008 | 11:39 AM
  #8  
roger smith's Avatar
Evolving Member
iTrader: (4)
 
Joined: Dec 2003
Posts: 357
Likes: 0
From: Ventura County, CA
Question, why is your timing dropping so low at 2500 rpm 100 load area? I have about twice your timing, but I'm only running a turbo-back. I'm not sure what other mods you have.

Your AFR in that area looks to be in the 14.0-14.7 range. Fuel is still burning slow there.

Also short, drastic changes in timing or AFR can cause knock. It seems your car is doing that.
Reply
Old Apr 2, 2008 | 12:46 PM
  #9  
tkklemann's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: Jul 2005
Posts: 1,228
Likes: 0
From: Charleston, SC
Originally Posted by roger smith
Question, why is your timing dropping so low at 2500 rpm 100 load area? I have about twice your timing, but I'm only running a turbo-back. I'm not sure what other mods you have.

Your AFR in that area looks to be in the 14.0-14.7 range. Fuel is still burning slow there.

Also short, drastic changes in timing or AFR can cause knock. It seems your car is doing that.

The timing is because I was pulling it to try and see if that made the knock go away. Is it possible that too little timing can cause knock? In some areas of my map, i have pulled in excess of 15* trying to get it to go away to no avail. I set it back timing wise to where it was because the knock didn't get better or worse with the timing change.

The sharp change in knock is only shown on this specific log. I have logs that show 14-14.7 throughout the entire knock event, as well as ones showing it at 10.5 throughout the entire knock event.

The drastic change was caused by the closed loop/open loop transition. If you look at the O2 change to 100, it corresponds to where the AFR suddenly drops. But, the knock started before the change in AFR.

Make sure when you look at the data log, check out the JDM MAP numbers. I am curious if the car can handle the 1-2psi, 8* (What its suppose to be) and 14-14.7 afr's numbers. Could this be caused by the 1-2 psi I am getting at such low rpm numbers?
Reply
Old Apr 2, 2008 | 01:17 PM
  #10  
JohnBradley's Avatar
Evolved Member
Shutterbug
Liked
Loved
Community Favorite
iTrader: (30)
 
Joined: Jan 2004
Posts: 11,406
Likes: 78
From: Northwest
Open loop load and open loop TPS both need changed. I know this started right after the turbo change, but what condition is your clutch in? Mileage, abuse, etc.

Remember that as long as your knock control is on and the counts arent staying at lets say 8 and increasing, the counts afterward are echos if you will and the car is adding timing back in but remembering what knock level there was. Like a ping on an oscilliscope. Its not going to eat away slowly at your motor, becuase if that were true what about all the Brand X and Brand Y flashed cars that drove around with 15-20 counts for years and didnt know it till they got into logging and flashing themselves?

I am not trying it minimize or poke fun or anything but I think were all too worried about knock. If it knocks under boost fix it, but at NA or low boost levels who cares? What about your grandpa's small block that has AUDIBLE knock everytime it goes up a hill because he puts 87 in it? Those things go for 100K or better without any knock control to drop timing. Too low of timing can cause knock as well as being too rich.

What would be cool is to take a 100% stock Evo for a drive and see how much it knocks in just normal unboosted to low boost driving. I know they knock under full boost so I wouldnt be surprised to see part throttle as well.

Sorry I am not really much help in your problem though
Reply
Old Apr 2, 2008 | 01:52 PM
  #11  
tkklemann's Avatar
Thread Starter
Evolved Member
iTrader: (6)
 
Joined: Jul 2005
Posts: 1,228
Likes: 0
From: Charleston, SC
Originally Posted by JohnBradley
Open loop load and open loop TPS both need changed. I know this started right after the turbo change, but what condition is your clutch in? Mileage, abuse, etc.

Remember that as long as your knock control is on and the counts arent staying at lets say 8 and increasing, the counts afterward are echos if you will and the car is adding timing back in but remembering what knock level there was. Like a ping on an oscilliscope. Its not going to eat away slowly at your motor, becuase if that were true what about all the Brand X and Brand Y flashed cars that drove around with 15-20 counts for years and didnt know it till they got into logging and flashing themselves?

I am not trying it minimize or poke fun or anything but I think were all too worried about knock. If it knocks under boost fix it, but at NA or low boost levels who cares? What about your grandpa's small block that has AUDIBLE knock everytime it goes up a hill because he puts 87 in it? Those things go for 100K or better without any knock control to drop timing. Too low of timing can cause knock as well as being too rich.

What would be cool is to take a 100% stock Evo for a drive and see how much it knocks in just normal unboosted to low boost driving. I know they knock under full boost so I wouldnt be surprised to see part throttle as well.

Sorry I am not really much help in your problem though
It did indeed only start after the turbo swap, and injector change.

The clutch is a Fidanza segmented flywheel, ACT super huge pressure plate with a 6-puck unsprung ( ) center. It was never a problem before, as the old turbo with tune was stellar, no knock in those areas. The clutch only has about 8 launches on it, and about 20K miles.

True about the Brand X & Y knock counts. Good point. I guess when you are chasing a problem, sometimes you don't think of the surface reasoning behind it, if that makes sense.

The big reason that I am wanting to fix this is you can really feel the timing being pulled. When pulling away from a light, it feel like this: Having your car tied to the car behind it, then when the knock starts, the rope that ties your car to the one behind it looses all the slack, and you start to pull the car behind you. Then, once the slack is out, the car takes off like you shot it with say a 15hp shot.

I have made progress with it when I resorted to putting in the old tune. It worked well for the most part, except this problem came up. It sucks because I haven't really tweaked the upper WOT parts to a tee yet because I ca't ge the car past 9mph wthout it knocking.. I just want to get to 88mph with no knock!
Reply
Old Apr 2, 2008 | 03:16 PM
  #12  
steexz's Avatar
Newbie
iTrader: (4)
 
Joined: Mar 2007
Posts: 91
Likes: 0
From: Las vegas
I have the same problem as you .. I'm running full tbe with cams. O2 housing, intake, mbc.... Basically there are times when I will be cruising in any gear between 80 to 120 load between 2500-4000 rpms and basically I get knock... Sometimes I will get knock in neutral if I give it gas to let's say 4500 rpms...e friend who is tuning has never seen this as well... I'm running 100 octane with 29psi spike...I have the cel light flash when I get 5 counts of knock so I know when it happens...in running the act street hdss with the prolite flywheel.... It just seems like it is random when this happens and seems to happen even more after a couple wot runs.... The crazy thing is that in evo scan right where it starts to flash u can see. Where timing was pulled... We can't figure it out...but yes generally wot its fine... Subscribing bcuz any help is great...
Reply
Old Apr 2, 2008 | 03:19 PM
  #13  
steexz's Avatar
Newbie
iTrader: (4)
 
Joined: Mar 2007
Posts: 91
Likes: 0
From: Las vegas
also the worst part is that I have seen 35 counts of knock in this area at times ....
Reply
Old Apr 2, 2008 | 03:59 PM
  #14  
Slo_crx1's Avatar
Evolved Member
iTrader: (9)
 
Joined: Sep 2007
Posts: 821
Likes: 1
From: Simpson, PA
Originally Posted by steexz
also the worst part is that I have seen 35 counts of knock in this area at times ....
Seeing 35 counts has got to be a mechanical cause, that's extremely high for any knock count. I too get the occasional knock count in the range up to 100% load, but honestly even if I do spike upwards of 7-8 counts it really never seems to make a "feelable" difference in power production. Once the car passes the 100% load range into boost all knock goes away and maintains a strong pull through the rpm range. I think JB is right on this, we're getting way too picky in trying to resolve the low load part-throttle knock issue.
Reply
Old Apr 2, 2008 | 04:04 PM
  #15  
20psiMR's Avatar
Evolved Member
iTrader: (6)
 
Joined: May 2006
Posts: 731
Likes: 0
From: From Vegas to Reno
Im the one tuning Anthony (Steexz) car. Ive tried a combination of things. Even flashed it back to a stock map. It seemed to really come on strong when I added in a modified mivec map (he has cosi cams). On a bone stock map its not so bad at all. What Ive found so far is a STRONG dip in timing that seems to be causing the knock. Same areas TK, between 80 and 120 load. Im running 2byte on his car as well. Whenever that timing drops, it throws crazy knock. Car pulls hard running a ton of timing up top too, for 29psi on 100.

What Ive found so far is seriously increase the timing in the high octane maps. It wants timing, and when it gets pulled down to the 20's it throws 35 counts of knock. Right now we have it running pretty solid running like 35 degrees of timing in that area. Its really weird, because my car has basically the same parts and doesnt do it using the same settings. Both cars are nines. I'll attach a copy of the timing map Im using right now, it seems to have quelled the part throttle knock quite a bit.

Attached Thumbnails Need some tooning help to rid knock.-anthony-timing.jpg  
Reply



All times are GMT -7. The time now is 03:03 PM.