My knock retard problem.. My heads/cam V6 is slower than stock.

[Magnus]

I posted this on clubgp.com and mfba.org.. It's been forever and I just don't know what to do anymore. I have a heads/cam car that runs like crap.

Ever since heads/cam my knock sensors have been going crazy. My cam isn't that agressive either. 212/212 .520/.520 112LS

Piston to valve clearance is ok.
Knock is false, I've used race gas.
No exhaust pipes are hitting the frame.

Knock sensors are good. I have 2 that are in the bottom of the block. Even if I have only one plugged in I have the same results.

I had 6 broken inner dampeners on the CC 941 springs. Now I just run the springs without the dampeners. Installed load is 96ft/lbs. Rockers clear the retainers fine.

The problem:

The knock sensors pick up a lot of noise below 2000 RPM's

In cruise if I"m below 2000 RPM's I'll have KR. It will stay. I can watch the knock count just increase.

In park if I rev it I will see a spike of KR for a split second but it goes away.

The knock count does not increase when I'm over 2K rpm's.. However, If I see 15 degrees of knock on my way to 2K rpm's (Like when I launch), it takes 12 seconds or so for the PCM to go back to zero degrees of retard.

I don't know what to do anymore. The valvetrain is QUIET. I hear ZERO pinging. There is no exhaust pipes hitting frame.

[/b]Here are the only things I can think of that would be causing the knock sensors to be going crazy:[/b]

1) My harmonic balancer has some how gone bad. It's an idea. I doubt it though because its a solid piece. It's not on wrong either cause the damn thing is keyed.

2) My poly trans mount vibrates the whole car. I doubt its this though because I've had the mount on for years with no problems. Problems happened immediately after heads/cam.

3) The knock sensors are picking up the misfire kicks from the cam as knock. The engine does shake a little at idle sometimes and kinda kicks from the misfires.. This is what I'm leaning toward as the cause for the KR sensors to go nuts.

4) Perhaps bad flywheel.

Here are possible solutions going AROUND the problem which is something I don't want to do.

1) Transbrake. Stall the engine up bast the problem area and just be done with it.

2) Buy 2 knock sensors and bring them inside the car and put them on a toggle switch with the ones in the block. Switch to the block ones in the upper RPM's for safety.

3) Start searching PCM code for KR tables and zero things out.

4) Drive the car off a cliff.

During lunch I took some autotap logs..

http://www.dxsoftware.com/misc/startup.htm

That is a cold start. You can watch the engine coolant temp rise and the car switch from open loop to closed loop. In that long MOST of the time the knock sensors go crazy is 1600-1700 RPM's. I really think its a frequency/resonance issue. Something is causing the knock sensors to go off.. Something is at the right frequency.

http://www.dxsoftware.com/misc/run.htm

That log shows how bad it really is... As you can see by that log, its impossible for me to start the car from a stop with timing advance... As soon as I accelerate it pulls all timing and I have to wait for the damn car to recover. If I keep the RPM's over 2K when I drive, I'm good to go.. I can punch it from a roll with full timing and pull nicely. However if I am cruising along at 1500 RPM's, I'll have full knock and if I punch it the knock will remain and slowly go back to zero.


The autotap logs speak for themselves. I currently have a 3.23 rear end with an open differential. I can't break the one tire loose with my car. Thats how bad it is.

Anybody have any ideas here?

[AZ3.8Camaro]

I don't know too much on this subject. But maybe those dampeners are causing it? Why did they break? What do they do? How are you motor mounts? Maybe the lumpy Idle from the cam is shaking the motor?

[Magnus]

The dampeners are now gone. They broke cause they were binding.

Motor mounts aren't that bad.. the engine does shake from the cam.

[jersey]

stuff like this is over my head, i usually come to you for answers. have you contacted tom@SSM
He might have some ideas, also i would suggest pulling the cam and run the stock one to see if the KR goes away. Is your cam a regrind? maybe the specs are wrong or the cam is messed up. this is really all i can think of, good luck keith

[black98V6]

I am kinda liking this theory about a resonance..
wonder if some sort of counter weight (somehow) attached to a knock sensor might change the resonance properties..

I'll explain..
read one time where someone had a side mirror on their car that would vibrate only at idle making it useless.. well it had to do with resonance and the vibrations caused by the engine at idle were just right to make the mirror vibrate.. the suggested fix was to attach a small weight to the backside of the glass thus changing the resonance properties..

do you think something like that might be possible?? I have never seen a knock sensor so I just don't know..

not to talk down but have you tried a mechanics stethescope? maybe you can track something down..


could something be shorting out?? maybe some wiring got messed up upon installation of the heads and cam?

too bad you can't get some other knock sensor and say hook them up in the interior of the car.. I'd just really want to see if that changes things.. gonna stew on this a bit..

man I am sooooo thinking this is something painfully simple..

[RiceyRS]

I think it's your cam alignment. I'm thinking the cam is causing vibration similar to an unbalanced tire that shakes at say 55-65 mph but no where else. Also did you replace the bearings/pushrods/ or anything else that moves or rotates at different speeds?

[Only4U]

even though i don't know THAT much, i do know that brad's theory seems to be somwhat to pretty likely and his solution, if doable, sounds good.

now, for my elementary solution:

is there some way that you could possibly lengthen the throttle cable (attached to the TB) so that your engine always idles around 2kish? don't have any idea whatsover what this would do to the engine/torque converter/performance and such, but since you say your problems come only below 2k rpms, wouldn't keeping the engine above the 2k fix the problem?

like i said, it's elementary, but i figured i'd give my .005 cents worth anyway [img]tongue.gif[/img]

[slacker69]

Could it be something with your plugs i was reading somewhere that certian pluds help to eliminate KR here are some of those plugs:

http://www.zzperformance.com/zzp/pro..._sparkplug.htm

i am refering to the autolite ones.

[scottblack]

Sorry to hear this man. I am going to look around right now and see what i can come up with to maybe help you out. :(

Scott

[John_D.]

<blockquote>quote:</font><hr>Originally posted by Magnus:
The knock sensors pick up a lot of noise below 2000 RPM's
The knock count does not increase when I'm over 2K rpm's...
it takes 12 seconds or so for the PCM to go back to zero degrees of retard.
I don't know what to do anymore. The valvetrain is QUIET.

Here are the only things I can think of that would be causing the knock sensors to be going crazy:
1) My harmonic balancer has some how gone bad.
2) My poly trans mount vibrates the whole car.
3) The knock sensors are picking up the misfire kicks from the cam as knock.
4) Perhaps bad flywheel.

Here are possible solutions going AROUND the problem which is something I don't want to do.
1) Transbrake. Stall the engine up past the problem area and just be done with it.
2) Buy 2 knock sensors and bring them inside the car and put them on a toggle switch with the ones in the block. Switch to the block ones in the upper RPM's for safety.
3) Start searching PCM code for KR tables and zero things out.

...In that log MOST of the time the knock sensors go crazy is 1600-1700 RPM's. I really think its a frequency/resonance issue. Something is causing the knock sensors to go off.. Something is at the right frequency.
<hr></blockquote>

My first thought would have been valvetrain noise. If you can definitely rule that out, then that mainly leaves a resonation or the cam/lope. Since lope eases out above idle, that may be it.

Either way, resonanace or lope, I don't know a way to eliminate it, so you will probably have to work around it.

Trans mount probably is not it. That transmits vibrations from the engine to the frame, but not vice versa, and the sensors are picking up engine vibrations.

If it takes 12 seconds for the retard to go away, I would shy away from the transbrake approach, unless you are shooting completely past the kr issue when slamming the rpms from idle to stall. Otherwise you will still be in the 12 second countdown when the light turns, unless the other guy stages really slowly...

Spare knock sensors sound like a good idea, with switching back to the real ones above 3k. Maybe you could even use a window switch?

Zeroing out kr in the pcm would do basically the same thing, but I am not sure if you can keep it intact at high rpms and eliminate it only at the lower rpms.

You could bypass the stock timing advance mechanism entirely with an aftermarket unit, but then you still don't have the knock sensors working in your favor above 3k. You could still log the kr readings and tune from them though.

[ September 05, 2002: Message edited by: John_D. ]</p>

[Dominic]

That is most definetly cam lope causing it since its BELOW 2000RPM. Remember the stock cam is not split duration, not that split duration is bad, but going with a split duration cam would most likely solve that issue if its indeed cam lope.

The lope is unexpected and the slight misfires would cause a knock-like vibration. I've seen LS1s do this on really big cams before - it sucks.

Don't drive her off a cliff yet [img]smile.gif[/img]

[black98V6]

dunno if this helps but some owners of those $30k LS1s so many of us drool over complain of over sensitive knock sensors.. I'm just wondering why nobody else with a cam is having the same problem.. (in a v6 of course)

man if I just knew more of how a knock sensor worked this could be so much easier..

pretty much all a sensor does is act like some form of adjustable resistor.. something causes the voltage flowing through something to change be it heat.. pressure.. amount of oxygen in the air.. whatever causes a change in voltage.. the pcm is desiged to read these changes in voltage as changes in something and adjust for it..

so what exactly happens inside a knock sensor to change this voltage?? I never was that great in digital design in college.. but it's not THAT tough..

what I'm trying to get at is there has to be some way to "desensitize" the knock sensor.. if we could find a measurable way to do this.. then we should be able to "tune" the knock sensors to ignore the cam lope.. hell listening to Ironman's vid clip of his car lopeing.. I'd think he would be having the same problem.. IF INFACT cam lope is causing false knock..

ahh well.. more stuff to stew on.. man with the collective brains here.. we should be able to come up with a solution.. or if nothing else locate the problem.. [img]graemlins/thumbsup.gif[/img]

[pvtschultz]

The KR is caused by the idle lope that the cam is creating. Thousands of people before you have seen the same thing. My last enging project didn't smooth out until around 3000 RPMs, thankfully there were no sesors in the car.

Anyways, an old trick used on the older OBDI V8s, is to put teflon tape around the knock sensor and not screw it in all the way, making it less sensitive to knock. That is what I would try. Make sure that when you do do it, that you aren't getting real knock and your sensors are not sensitive enough.

[jmorgan]

theres a knock desensitizer the LS1 guys use Nineball (tony) guy has first 434 MTI sroker i think made them and sold them look at his profile on ls1tech or maybe email him. allso i know lt1's used to get LT4 knock sensoers (they are less sensitive) maybe they would work?


if not i like the idea of engaing the knock sensors with a window swtch at maybe like 2500 rpm so basicly the only time your sensors wouldnt work would be right off the line and since you know its not knocking there i think that would work good if theres no alternitives.


umm i think it could be the cam but if ironman isnt getting knock i dont really see why you should

[v6maro]

Like someone said, if it sends a voltage to the PCM, find out what voltage it sends when it's normal, then find a resistor to counter-act the KR at those specified RPM's with a window switch.
just an idea
-Steve