What is Acceptable Knock on LS3 and LS7?

[jdd123]

I'm encountering random 1 and 2 degree knock on my mildly modified Camaro and bone stock vette. I am not sure what is an acceptable knock. Maybe better stated, what is real vs false knock. The runs made are on black top roads. So what is your opinion are they road noise or real knock?

I have a 2010 Camaro with American Racing Headers with free flow cats, Corsa mufflers and a intake that put the MAF on a straight section. So they are really mild changes. Altitude is 5000ft. I followed the MAF part throttle and and wot throttle tuning that is from the tuning school. MAF definitely needed calibrating. At this time, I left the PE table stock even though the low 11's afr in WOT seem rich. But that is what the OEM called for. Anyway, the long term fuel trims were low so I decided to leave the stock speed density VE algorrithim at low rpm's rather than enabling the MAF down to a low RPM. I am not sure how to adjust the virtural VE anyway, but it appears to be still close from the logs.

So next on my agenda was to tune spark at part throttle. I was trying to use delivered torque at part throttle on the road. Before I could realy get into the spark tuning I did some test runs and noticed this knock. For fun, I even set the Camaro down to 10 degrees (in the part throttle section) as a test and still had a 1 degree blip. I attached both runs and the tune. I also included a sample wot run I did with the wide band connected. I did not install the wide band for the part throttle runs because I was not going into power enrichment.

I also attached my stock zo6 part throttle and wide open runs and noticed it had knock. By the way if you see the afr jump leaner on the vette 3k rpm into WOT that is probably because the flaps on the muffler opened up and the wide band was on the muffler tip. Eventually, I will install a bung on the vette. The Camaro has a bung after the driver side cat that I attach to.

I purchased/read the basic and advanced courses from the tuning school and read several fuel injection tuning books including Greg Banish's.

Sorry for the long post. I was trying to do a little pratice spark tuning before taking it to a dyno and blowing money because I am so slow. The stock knock is a concern. I understand the delivered torque method is not delivering repeatable data and I don't know how to change the virtual VE table. But these issues are seperate and not a priority versus determining what is accepatble knock from road noise.

[mav_7417]

im having this same problem on a 2006 z06. stock tune with 90 oct.
please tell me what you find. people were telling me its false knock but i dont want to risk it

[juyanith]

Me too.

I have a somewhat modified 07 Z06 and I get consistent knock that doesn't go away despite pulling timing and adding fuel. I'm still trying to eliminate any mechanical cause but it does seem that the knock sensors are, well, sensitive. I hesitate to reduce the knock settings because I don't want the car to miss any real knock should it happen. However, I have read that others have had to do just that. Does anyone have a link to an explanation of the right way to do this?

[streetsweepergto]

A rich afr isn't always better.

You have some modifications.

Your best bet is to get fueling dialed in.
Then adjust spark accordingly.

If you still encounter knock
Add in some octane booster and double check.

If it still knocks then it is probably false.
If you retard the spark by half it's retard value or more and it still knocks it is probably false.

I wouldn't recommend dumbing down the knock sensors.


Make sure it is not burst knock your looking at also.
Good luck.

[Chris@HPTuners]

there's no easy way to do it, the knock sensors are effectively a microphone and the ECM and electronics filter and process the "sound" and make a decision based on the background noise level and anything it hears above that level, the louder the knock the more initial timing is pulled. Tuning the sensitivity calibrations require you to actually know what is real knock and what is other noise, the 2 most common ways i have seen people listen for knock on engine dyno's are:

1. Bolting a long length of copper pipe to the block and run it thru the wall to the dyno control room. The pipe amplifies the noise and it's usually easy to hear knock events while sitting in the quiet control room.

2. use a device like "the knock box" which is basically a knock sensor that attaches to the block and comes with its own filtering knobs and buttons and a set of headphones. You literally listen to the engine thru the headphones. It works very well but obviously costs more than a simple piece of tube.

http://theknockbox.com.au/

there is lots of cool info there as well.

The bottom line is that zero real knock is the acceptable limit during all driving conditions, especially WOT. If you are reducing timing and adding fuel and cannot actually hear any knock then it is likely to be false detection by the sensors, but to expect yourself to be able to calibrate the sensors better than GM does in a lab environment is another story...

Chris...

[jdd123]

I want to find an effective way to tune my spark. I never actually heard knock in either stock vette or the maro. I do not want to reduce the knock sensitivity. Just need to know when my spark advance is reaching the true knock limit. I'm ok with the VCM taking off a degree here and there rather than risking missing a true knock.

Yes, I already calibrated the AFR. I calibrated the MAF to deliver the requested AFR at part and WOT. The PE EQ ratio is stock will leave it for now. I next wanted to get the spark tuned at part throttle and that is when I noticed that the 1's and 2's of knock randomly appear even when throttle and engine steady going down the road.

As far as knock goes I guess one possilby two degrees could be false. I've done part throttle runs on the same day where I did a base, advanced 2 from base, retarded 2 from base and final base run. I had less knock in the advanced and more knock in the retarded. It was initially confusing. Sunday, I tried just 1k to 2k RMP's, looking at knock and delivered(another story)torque. By 4 degrees of advance from stock, I noticed that the knock was hitting a few 3's and higher. So in those particular cells, I figured that would be the knock limit to back off from. I was still concerned that the stock base runs were knocking. By setting the base spark map to 10 in the part throttle areas should have made it disappear. It didn't. So that convinced me it may be random noise, but wanted to make sure. I do not have the experience to just assume it. So that is why I posted.

So is there a way to get the VCM to use the main spark table without modifiers when tuning? Do I have to go back to the many individual modifying spark tables and set them to 0?

[streetsweepergto]

Rent a dyno.

Polish on the street.

[jdd123]

That comment really wasn't helpful. With the mild mods that were done, I am only "polishing". Did not address the random knock issue. Looks like I need to make a judgement as to what is real.

[Chris@HPTuners]

if you are only running 10deg final timing you shouldn't be seeing knock retard at light throttle. But looking at your logs itlooks like KR is happening at low RPM at highway speeds (in a high gear i guess) when you open the throttle a little, timing is around 35deg. I wouldn't assume this is false knock. Since you are lugging the engine at low RPM and high timing its more likely real.

[jdd123]

Thanks, I will look at this closer and do some more runs. There is a reason I was keeping the rpm's low.

Most people want to tune wot runs etc. right off the bat. I wanted to try to improve mileage at 55 and above. Have a couple mile stretch where I have to go 45. I was trying to get the cells up to .048 g/cyl filled. The engine did not seem to be lugging and I could not hear any knock. I tried to not to go above mid throttle but to catch a few cells that I might use if I wanted an easy pass closed loop. I noticed that wot runs started around .06 so figured the .048 did not seem that high. Used both low and high gear. Had to start somewhere, so I tried to hit as many cells in the 1 to 2k rpm range as I could but still staying in closed loop. I limited it because those rpm's are the highway speeds and where most of my commute takes place. I plan to work up higher rpm's and eventually PE. First though, needed to pratice and test methods.

Except for the one that said 10 deg, the files I submitted were stock timing with the corrected calibrated MAF table. Tried to do various loads at 1 to 2k rpm in closed loop. I noticed that the TPS changes at the knock were only a few % in change and there were some knocks where tps was constant.

[jdd123]

Thanks, I will look at this closer and do some more runs. There is a reason I was keeping the rpm's low but the engine didn't feel like it was lugging. There were a few times that the TPS was steady and it happened. The files I submitted were stock with the calibrated MAF. The file with "10deg" was where I set the spark from .016 to .052 and 1 to 2K to 10 degrees. So anything out side that rang on that file would be the higher stock.

Most people want to tune wot runs etc. right off the bat. I wanted to try to improve mileage at 55 and above. Have a couple mile stretch where I have to go 45. I was trying to get the cells up to .048 g/cyl filled. The engine did not seem to be lugging and I could not hear any knock. I tried to not to go above mid throttle but to catch a few cells that I might use if I wanted an easy pass closed loop. I noticed that wot runs started around .06 so figured the .048 did not seem that high. Used both low and high gear. Had to start somewhere, so I tried to hit as many cells in the 1 to 2k rpm range as I could but still staying in closed loop. I chose this limited range because those rpm's are the highway speeds and where most of my commute takes place. I plan to work up higher rpm's and eventually PE on a dyno. First though, needed to pratice and test methods.

[streetsweepergto]

That comment really wasn't helpful. With the mild mods that were done, I am only "polishing". Did not address the random knock issue. Looks like I need to make a judgement as to what is real.

renting a dyno is a easy way of tuning spark.

on the dyno you can dial in spark accordingly and see how it helps with performance or not.
in that environment its also a little more easier to listen to the motor without major distractions such as the road and other people.


add to the fact that wot on highways and streets is just down right dangerous and of course illegal.

i use to be a firm believer in attacking all my tuning on roads, but a dyno is where alot of the magic tends to happen.
so much easier to dial in things, less variables etc, alot safer too.

[skythunder83]

if you are only running 10deg final timing you shouldn't be seeing knock retard at light throttle. But looking at your logs itlooks like KR is happening at low RPM at highway speeds (in a high gear i guess) when you open the throttle a little, timing is around 35deg. I wouldn't assume this is false knock. Since you are lugging the engine at low RPM and high timing its more likely real.

Hi I have send you a private message, can you check it pls?
Or I can send it again to your email if you like.

Thx

[IDRIVEAG8GT]

A "Less Safe" way, but something useful I do:

1. Disable BKR completely.

2.Take Knock Fast Attack recovery rate and set all numbers to 5.

3. Take a running log with those settings.

4. Watch closely, if you see any knock when you hit a bump in the road THE SENSORS ARE TOO SENSITIVE.

5. Adjust sensitivity accordingly.

GM set the sensors way the hell up there on the newer vehicles, so my guess is that they are transmitting false data to the ECM.

[ShoddyHog]

Don't know if this helps, but...

I let a "performance shop" install my first cam, and my car had knock everywhere. Some of it was real because the "tuner" dialed in 28deg of WOT advance. Even after fixing it, howerver, I still had knock on every single log...and I do mean every single one without excpetion.

I did all the tests to rule out real knock, and decided most if not all of what I was seeing was false. I drove the car like that for almost 70,000 miles...doing probably in the thousands of WOT pulls to redline in that time).

After the first 25,000 miles, I found I had some valve damage, so I swapped to L92 heads and different injectors. There was no change whatsoever in the KR.

At 70,000 cammed miles (115k on the block) I lost a lifter last fall, and it took out the cam. Installed a new cam (different one), Morel Link-Bar lifters, Lunati dual springs, and had the heads freshened up. Everything else was the same.

While I had everything apart, I MEANT to retorque the sensors, but forgot...put her all back together, and fired it up.

Guess what? No more KR period. Sure, I can MAKE her knock at WOT if I try to give her 28 degreees again, but all of that false knock is now gone. Hell, I can't even get it to knock lugging the crap out of the engine.

So, what fixed it? Beats the living shit out of me. I didn't touch the knock sensors or wiring. My timing chain was grinding on the front cover, and I fixed that. I also did loosen the header collectors and retorque them with new hardware, but they didn't feel loose at all.

HTH...

[IDRIVEAG8GT]

Also, sometimes the exhaust manifold will heat the exterior of the Kock sensors up, and as you all well know things expand with heat. They ALWAYS become more sensitive when warm over cold. Another thing to consider, these are aluminum engines thus magnifying sound much more than that of an iron block, and any detonation will clearly be heard with a relatively quiet environment around the car. You should be able to adjust timing accordingly, but a safe line that I have followed on L92 head applications and stock compression is about 18-20 degrees of timing max at WOT and 29-31 degrees cruising. Screw good fuel economy over a powerful and long lasting engine I say.

[jdd123]

Hi I have send you a private message, can you check it pls?
Or I can send it again to your email if you like.

Thx

I did not get the private message. You can email me at [email protected].

[jdd123]

I should have titled the thread what amount of knock is false. I'm not trying to push the vehicle to the edge or lug it down. I can't lug it anyway because the torque management is enabled.

For fun I logged a drive to work today and ya these little knocks are all over the place at light loads of .28 to .32. The car is stock spark tune. The knocks even happened at steady part throttle. Funny thing is today there were a few higher ones. Knock scares me but in the stock vehicles I have, there is false knock. Had I not experimented with the timing at 10 degrees in my target area, where it still had a knock, I would be concerned. It looks like others seen this too and unless I'm willing to mess with sensitivity,(I'm not) the dyno may be quieter.

Anyway, the dyno was always on my list for WOT. Just wanted to pratice part throttle tuning as laid out in the school books. Plus that is where I would get better mileage and I have more time to safely play with it. However, this false knock mucks it up.

[IDRIVEAG8GT]

Disable BKR..........

[04colyZQ8]

I am also getting a lot of KR mostly at part throttle cruising under load, or accelerating in a high gear between 1200- 1600 RPM. Could changing to a cast iron block over aluminum be part of my problem? As well as ridged mounts, and long tube headers?