slimdawson

Injuneer, in your opinion, what causes the kr without knock count and vice versa?

As far as the other problems, I would definately check out the opti and wires. Also, EMI could be a problem after hearing about your scan logs. Have you done any wiring to the car?

Injuneer

I don't know what logic the stock PCM uses to determine knock retard. I'm just telling you what I have seen in hundreds of data logs from cars that run well and cars that don't run so well.

I fooled around with my stock PCM for about a year, then wrote a description of how to use scanners for thge 'net. I then got rid of my stock PCM, and that was 5 years ago. So I have never programmed a stocker, have never attempted to crack the stocker, have not followed the LT1_Edit or other tuning forums.

He has a major problem, but he's focusing on the knock retard, and apparently decided to try and cover it up with a band-aid by desensitizing the knock sensor. I seriously doubt he will make much progress with that approach. The data logs are FUBAR. How do you explain a "0" pulse width in the middle of a WOT pass? How do you explain RPM data that suddenly drops by 500rpm, and then immediately regains the RPM and more in the next frame. I think it could be the Opti, could be faulty grounds, or could be an electrical interference problem. Or the scanner cable may be picking up some sort of interference. But he's gone after the knock retard, so I've sort of given up on trying to help.

ktg

Yes, I've started with the knock retard. Ya gotta start somewhere. I asked for opinions and the only one I got was 'it's fubar'.

Injuneer

I think I gave you a bit more info than that in response. Sorry I wasted your time.

ktg

So if knock retard was a bad place to start, where would you recommend?

ktg

I think that will be my next step, if I can ever get ahold of Ion. He's been hard to catch lately.

BlownF1

Perhaps I'm missing something since I've never looked at any log files, other than Diacom. Is there a field on these files called "knock retard" or should it really be called "timing retard"? What I was getting at, is there other reasons for the PCM to pull timing rather than knock? I was wondering if he's chasing a ghost by focusing on "knock retard", versus trying to find the source of the timing reduction?

For example; suppose he had an intermittent temp sensor wire and it radically changed value all the sudden, but didn't result in knock...would the PCM pull timing? That would have nothing to do with the knock sensor or knock tables, etc.

Hopefully that makes more sense, I don't know how else to explain it...see what I'm getting at?

Injuneer

Now I see what you are getting at... does the "retard" field (whatever it is called) also sum retard caused by things other than the KS? If you look at some scans, the field is called "knock retard", and in others it is called "spark retard". The majority seem to call it "knock retard", but I can't identify the software source of the logs once they are sento to me in .csv format.

DataMaster calls it "spark retard", and the manual contains the definition "The degrees of spark retard which have been subtracted form (sic) the nominal spark advance value. Is generally caused by an engine knock condition." This would appear to suggest there may be some other sources, but they don't identify them.

Where I think the logic falls down.... in my MoTeC setup, as coolant temp or inlet air temp goes up, I pull timing. If there was a similar offset table in the stock PCM, and the IAT or CLT was high, you would see a constant value for "spark retard" if those factors were contributing to the calculation of the number in the "spark retard" field. And the only time I have ever seen anything with "constant" knock retard it when you get a KS failure and the PCM defaults to a programmed calculation based on operating variables.

This is an early LT1 paper by Chris Bennight, and he calls it "knock retard" and also says it comes from "knock". If you look at the other tables, he mentions adjustment to timing based on temperatures.

http://www3.baylor.edu/~Christopher_...ht/LT1Edit.htm

My "take" on this is that the "spark retard" field is showing the knock retard component.

But the idea is woth looking closer at.

Jon A

You certainly can get knock retard without the knock count going up. Burst knock is a very common way this happens.

Also, the stock PCM's are set up with a very lazy rate of decay--you get some knock, it pulls a bunch of timing and it takes a long time for that timing to come back even with no additional knock. Where this can fool people is when you get some knock, it pulls a bunch of timing then you change some parameter briefly (like lifting the throttle, etc) where the PCM eliminates the KR. But if that's only brief and you get back to the same parameters where you had the knock, it'll add the KR back in according to the original rate of decay.

BlownF1

Thanks for the explanation and link...good info! Sometimes problems like this shed new light on how things work or at least make us wonder. I'd pick a different screwed up sensor reading and try tracking that down instead of focusing on "knock retard"...whatever ever that may be.

I think I'd try a different PCM as well...

slimdawson

I forgot to mention grounds. The grounds definately need to be checked and ohmed at the pcm connections. And back to my earlier question, have you done any wiring to the car?

ktg

Thanks for your input. No, no wiring has been done.
I will triple check the grounds. One on the block in front of the starter, one by the big red '+' by the battery, and one on the driver's side under the ABS. Any others?
What do you mean 'ohmed at the pcm connections'?
Different computer is first on the list now. In the meantime, I just read another thread entitled 'backfire through the intake'. Some in that thread think their opti may be installed off a bit. The symptoms sounded a lot like mine but my car runs and that one doesn't. Could my opti be off by enough to cause the symptoms but close enought to run?

slimdawson

I don't think you can put the opti on wrong, so I wouldn't assume that.

I am not familiar where the grounds are on a 4th gen. There should be some on the back of the head/s and some on the coil bracket. By ohm, I mean disconnect the pcm connections(battery disconnected) and ohm the grounds. Some of those pins are grounded. Put one probe in the connector and one on the chassis. I can't remember exactly what they should be under. You can get a list of what the connectors are asking around here(or search here). Grounds are critical to our cars. A new pcm will be junk without making sure grounds are there.

Also, check your ignition system in the dark to look for arcing.

A&SAUTOMOTIVE

I have seen this many times and replacing the knock module in the pcm will very possibly take car of it.I have seen cars that you can just bring the rpm's up in nuetral and it will retard timing with no increasing knock sensor counts and then swap a used one or new one and the problem goes away.Sometimes you can unplug the knock sensor and still get retard,its like it gets burned into the knock module and you cant reset it...

BlownF1

Not so...it CAN and DOES happen all the time. It's setup so that you can't install it wrong easily, but people still do. I've seen it.