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

[pgriffith]

Magnus, we will find out if the knock issues are from the cam soon enough. My cam will be here tommarow. I will take my time putting it in but when I do We will see if my car does the same thing.

[hou-tex-six]

<blockquote>quote:</font><hr>Originally posted by Jason Morgan:
theres a knock desensitizer the LS1 guys use <hr></blockquote>

good call Jason, that looks like it could work. Heres the link if you dont wanna look for it, reasonably cheap too.

http://users.ev1.net/~ynot_dv8/KOBox.htm

Cp

[Loren]

I have a plan, but first a little theory. Our typical race engine might spin at 6,000 rpms, or 18,000 power strokes per minute, or 300 little explosions per second. When you think in audio terms, 300 Hz is a pretty low tone, a little below the A string on a bass guitar.

A Knock Sensor is effectively a microphone. The PCM expects to hear rumbles, hums and bangs in that 300 Hz, or below, audio range. So how does it detect pre-ignition? Harmonics! Any one of you with a hefty sound system has probably seen a "clipping" light on an amplifier. The amp can detect clipping (distortion, which endangers the amp) via harmonics. A clean 1 Khz tone will show little harmonic content above 1 Khz. A clipped, or distorted, 1 Khz tone will show harmonic content at 2 Khz, 4Khz, 8Khz and multiples there of...

The PCM is using this same harmonic analysis to determine if the engine sounds healthy, or more like a handful of nut and bolts in a blender. If it hears rattle, buzz, clack, slap well above the reasonable 300 Hz range, it assumes something is terribly wrong. Err go, richen the mix, and back off the timing, saving the precious engine!

So, what makes these bad noises besides actual spark knock? Frickin everything! Things like a sloppy valve train, loose parts, a broken motor mount, even tapping on the side of the block with a hammer will trip the sensor. This has caused madness ever since knock sensors were invented.

Fine, another great idea that is often more trouble than it's worth. Let's fix it. The circuit is effectively an audio circuit. What we need, is a volume control. Enter the Ohm's Law geek. The knock sensor is most likely a low impedance microphone, perhaps 75 Ohms. I can make this assumption, because a high impedance circuit would be way too susceptible to interference from the ignition system (think AM radio). The solution is a resistive bridge that looks just like the original knock sensor from the PCMs perspective, but this one is hard of hearing. Consider the following design change:

PCM --&gt; 75 Ohm Knock Sensor --&gt; PCM
To
PCM --&gt; 75 Ohm Resistor --&gt; 75 Ohm Knock Sensor --&gt; PCM
|----------------&gt; 150 Ohm Resistor ---------------|

Placing a resistor, of equal resistance to the knock sensor, in series with the knock sensor will lower it’s output by 50%. Placing a resistor, of twice the resistance of the knock sensor, in parallel with the series circuit corrects the bridge back to the original impedance of 75 Ohms (sans a little inductance theory, which I’m sure the PCM cares nothing of).

Viola, the PCM can’t tell the circuit has changed, but now it can only hear severe (and more likely, genuine) spark knock. The resistor values in the bridge can be adjusted for the desired level of sensitivety. As long as the sum of resistance in the series circuit equals the parallel circuit, the bridge is of valid impedance to the PCM. Lower the sensitivety too much and you’ll turn up a “Failed Knock Sensor” code. Lower it just enough to solve the timing retard problem, and you’ll have a fast car that still has functional (albeit, lienient) knock sensors.

Oh, and in case you’re wondering, YES, I did take a few wedgies for talking like this in the schoolyard. But now I actually get paid to spew this technobabble... [img]smile.gif[/img]

[Magnus]

jersey, I have talked with Tom@SSM a few times, I'll be e-mailing him in a few days as I get more info. My cam is no a regrind, its a COMP 212/212 .520/.520 112LS. I really want to avoid swapping heads/cam right now. Those will be the very last things I do to solve the problem.

black98v6, I have tried a mechanics stethoscope. I couldn't locate a thing. :( I checked for shorts, all the wires are perfect. I highly doubt its something painfully simple. I've tried everything simple already. *sigh*

RiceyRS, it could be the cam is not balanced. Only moving parts I changed were cam, pushrods and springs/retainers.

Only4U, I can't idle at 2K. I did some more testing though and the sensors do pick up noise all over the RPM range. It's just the worst in the lower RPM's.. NEVER does it happen at WOT though. That makes me wonder.....

Slacker69, I'm pretty sure the plugs are good. I change plugs every 5K miles.

Scott, Thanks.

John_D., I can't definately rule out valvetrain noise. It is noiser than stock yes, but nowhere near as noisy as it was before when I had broken dampeners. I might pull the sensors out of the block.. I realy don't know right now.

Dominic, The sensors do pick up noise over 2K rpm's as well... mostly at part/no throttle.. Never at WOT though.

black98v6, your on the right track. I think however the PCM sends an AC signal and reads the DC return signal or Vice Versa??? I'm not realy sure. There are several V6's with my valve springs and MANY 3800's now with cams... I realy shouldn't have to desensatize them. :( I wish I really knew what the heck was causing them to go nuts.

pvtschultz, I do have teflon on the threads right now.. I also torqued them down to 12 ft/lbs instead of the spec of 14 ft/lbs. Perhaps its still too tight.

Jason Morgan, I already added the LT4 knock module. Didn't help. The LS1 cars use a different style knock sensor than we do too.. ours are LT1 style. I'll look more into the voltage aspect of the sensors this weekend perhaps. Motivation is growing thin however.

Loren, GREAT info! [img]smile.gif[/img] I will be doing voltage/resistance tests soon hopefully with the sensors.

I really doubt I'll find out why they are going crazy. I'll probably just end up relocating them out of the block.

Btw, HUGE thanks to CamaroZach for coming out yesterday and helping me with the car. Thanks a ton my friend. [img]smile.gif[/img]

Also, much thanks to all those who have read and replied to this topic. I really do appreciate it.

I did record some mp3's that I should post on monday. I have 3, one from under the oil pain, one from in the engin bay and one on the side of the car where you can here the exhaust. The valvetrain is louder underneath the car than it is from above but I think thats normal.

- Keith

[pgriffith]

Magnus you are using different valve springs right? They are wider diameter could they be causing clearence issues which the knock sensors could be picking up?

[snotty]

There was a guy who made a knock sensor thing to trick the pcm, i emailed him and he said if he had apicture of the knock sensors and the hook ups he could possible make one for the v6, i cant emember his ame at the moment :(

[camaro_zach]

keith, it was no problem man. Anytime you need a hand you know i'll be there. btw, had a blast at horist's house today, glad you were able to come with. hahaha, we gotta go hit for hit again sometime.

[pgriffith]

Magnus I posted this problem on the local Utah Fbody message board. These guys know there stuff. I hope they can give you answers. IF you would like to follow up on the thread you can check it here:

http://www.ufba.org/forum/index.php?...;threadid=1326

[*K*night Rider]

What is knock, retard?

[John_D.]

<blockquote>quote:</font><hr>Originally posted by Magnus:
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.
<hr></blockquote>

Magnus, this one statement of yours has been in my head and bugging me off and on all weekend.

1. Is there a geometry/clearance problem, which caused the dampener springs to go in the first place? If so, this may be creating just enough noise to trip the sensors, even if nothing is breaking anymore.

2. Without the dampeners, there may very well be some harmonics being introduced into the system that would not normally be there.

Just something to think about. This is the one most out of place thing you have described about your combo, and therefore is a highly likely culprit.

[Magnus]

John_D., The inner dampeners broke cause they were binding. Now that they are gone there is no more geometry/clearance issues. About the harmonics, yea.. its possible, However I doubt its that. Several other V6's run 941's. Some LS1 guys use them without the dampeners too. I don't really think the inner dampener is much of a harmonic dampener dealing with audio, but more of a stabilizer to keep the straight wound spring from wobbling in the middle.

Here are some mp3's I recorded of the car and the valvetrain. It's louder underneath the car than it is on top. I assume this is normal. It doesn't sound that bad. I'm still looking for ideas.

Note, there is a slight ping/squeek in these files. It's a small exhaust leak at the EGR. No, its not the noise that is causing the knock sensors to go crazy.

Underneath the oil pan (362KB)

In the engine bay (226KB)

Idle recorded from the side of the car (191KB)

[ September 09, 2002: Message edited by: Magnus ]</p>

[Magnus]

Talking with a local club member this morning.. and an interesting idea was brought up...

Perhaps my lifters are damaged from the binding of the inner dampener. When a spring binds, SOMETHING in the valve train has to give. If the lifter can compress to compensate for the bind, then your ok.. however it if can't compress anymore, what happens?

I'm thinking I might have a bad lifter or perhaps a clogged oil passageway that pumps up the lifters. I don't think I have a fully collapsed lifter though as I believe it would make more noise.

[pvtschultz]

Good point on the lifter idea. If one were collapsed, it would sound like a car that was just cold started and the lifters have settled with all the little clicks and such. When the lifter is good, it goes away shortly after starting. But you may have collpsed a lifter or even bent a push rod in the process of breaking the dampners....Just food for thought.

[Magnus]

Pushrods are straight. Checked them last week.

The whole thing about the dampeners binding.. Something has to give. Either a pushrod bends or the lifter squeezes enough to absorbe the bind... If the lifter can squeeze no more, and the pushrod doesn't bend... what takes the hit?

:(

I'll try to do a compression test this week.

I'm really thinking lifter though.

[John_D.]

<blockquote>quote:</font><hr>Originally posted by Magnus:
The whole thing about the dampeners binding.. Something has to give. Either a pushrod bends or the lifter squeezes enough to absorb the bind... If the lifter can squeeze no more, and the pushrod doesn't bend... what takes the hit?
<hr></blockquote>

I haven't played the sound files yet, I don't (can't) download big files at work...

Here's a couple of thoughts on the whole lifter/bind topic:

The plunger has oil behind it, which resists against the pushrod (ultimately against the spring pressure). That oil behind the plunger can be compressed further than it already is. If the bind increases, the oil is compressed more. Eventually either the oil can no longer be compressed (it hits an upper limit), or the plunger bottoms out, whichever occurs first. So it could be possible to break a spring, without killing the lifter, bending a pushrod, or breaking a stud, if the pressure builds high enough but the plunger has not bottomed yet. Whatever the weakest link is, will let go. That pressure build is some sort of curve, and probably has a pretty high value at the top, approaching infinity, but not quite there yet. Just high enough to break something else besides itself.

Damage to the lifter could be from bottoming out the plunger and damaging it, or overpressure on the drainback valve might cause it to fail, allowing too much slack (noise) during normal operation. Replacing the lifter would fix either of these.

Correcting the original coil bind problem may be needed too, if that's what it is, unless you just leave the dampeners out, and are sure there is not any remaining bind issue.

Did you use custom length pushrods, and how much did they preload (offset) the lifter plunger in it's travel?