I am getting constant knock retard. I tried unpluggin gthe sensor and still get knock retard from 6 to 20 degrees. 20 on tip-in and when try to hard accel.
350, miniram, fastburn, medium cam, Super_8dm2.xdf, 1227727 ecm, moates G1 adapter, 27C256 chip.

Been researching for days...

Here is a log file:

5.dcf

 

If this is 1990 or above, it could be the knock sensor board in the memcal going bad.

 

it is a 1990 corvette, AXCN or ANHX memcal.

Problem is, where do I get a replacement memcal?

 

Would an auto memcal from 90 or 91 work?

There is a 1990 corvette manual prom on ebay, but I do not think it is actually a 1990 manual memcal: https://www.ebay.com/itm/19601317396...Bk9SR8LNuPPMZA

 

Tunedperformance often has lots of these Memcals.

If you're on Corvetteforum, he goes by Tunedport90.

Are you able to burn your own chips and do your own reprogramming?

 

Yes, I am buring my own chips with moates hardware. TunerPro and TunerCats softwares.

 

ok, that opens up other options for you besides just Corvette memcals.

Any TPI 350 from a Camaro will also work. Since the 27S256 chip on the G1 adapter is what the ECM is actually running on, the only thing needed is the resistor board in the memcal.... And they're the same across the 350 Camaros and Corvettes I believe.

BTW, what happens if you disable the knock retard entirely within the calibration? Does it still behave the same?

 

All correct except I have a 29C256 chip. So, now I have huge options for the memcals.

I am going to try diasbling the knock retard table tonight and see what happens. If the knock still present, then would that prove the memcal board is bad?

 

obviously I can't say for 100% certain... but I'm thinking if the knock retard subroutine is engaging with the knock sensor disconnected, then I think the only other piece of hardware that can cause the ECM to react is that knock filter board in the memcal.

If you can, maybe pop the plastic lid off the memcal and if you have any magnification devices you can inspect filter board and see if you can detect and damaged resistors or anything like that. It could be that with the ECM in the engine compartment and subjected to vibrations and stuff, that somehow there's some intermittent connection on the knock filter board that's getting exacerbated.

Do you get any difference if you ease into the throttle vs more of a stab on the throttle?

Maybe one of the other knowledgable folks on this board can chime in too. The memcals aren't cheap, so I'd hate to have oyu drop a couple hundred bucks on one for nothing.

Question: Is this a recent development? Was it running correctly at one point and now it's not?

 

It has never really ran correctly. I thought I was failing at any tye of tuning when I tried, but now this just caught my attention. There were a series of mechanical issues occuring over the years that also were shielding this issue (like a broken piston one time, damaged cylinder wal from broken piston rings another time, broken valve spring another time, oil leaks I chased for 6 months, ... the kind of crap that drives you nuts with a hobby vehicle). For a while I though the spark retard was from bad SA tables. I tried stock AXCN, I tried LT4 b/c the fastburn heads are close to LT4 ports and ombustion chamber design, I also tried some other peoples suggestions and all were failing. This seems like a probable cause.

Thank you so much for your input @ULTM8Z. I am going to try buring a test calibration tonight and zero out all the knock retard, but without the sensor plugged in, I cannot see why the bin file would be setting any knock retard from my revious testing.

I should have a new-used memcal laster this week for further testing.

 

If you can log in Tunerpro, post the xdl file here

 

There's another possibility that your g1 adapter is bad. The knock filter board makes it's connection to the ECM through the adapter too.

If you ran your current memcal without the adapter....that would tell you.

 

UPDATE:

1. The G1 adapter is good
2. set knock enable temp to max temp then ran car and got NO knock retard, but was still getting knock counts eventhough I have the knock sensor disconnected. Car actually ran very strong.

So I gues the knock retard circuit is bad on the memcal. I have a new memcal coming this week and will make another update once I get that changed out.

 

Fingers crossed!

 

Well, changed the Memcal, same results. Switched ECM's, same results. Could be that the new Memcal I got is bad too. I got the DTC 43 code which is a knock sensor fault. Ran throught the fiagnostic tree routine and ended up with a bad Memcal answer. Do I have 2 bad memcals? maybe, I mean, they are like 40 years old at this point...

Again, when I disable the knock retard enable tables, then the engine runs great.

So... ?? should I try buying another Memcal? They are not easy to find nor are they cheap.

 

Does the ks ohm 3.9k or 100k ?

 

Did you consider the possibility that knock was in fact being reported by the knock sensor? Knock retard with no knock counts detected is normal, as I believe knock counts that are under the threshold value needed to be detected "as knock" by the sensor, but enough through a steady stream of knock counts under that detection value are still being stored elsewhere in the code. These undetected knock counts are still knock counts and are being stored and used as a fail safe to retard timing when the ALDL monitors a steady stream of them. I believe this was discussed years back. So I wouldn't blame the ECM or its hardware for malfunction, the system is picking up a steady stream of knock enough for the ALDL to retard timing. Pretty sure disconnecting the sensor wouldn't necessarily erase whatever was previously stored in the Bin, and all the replacement parts you tried did not solve the issue. The only thing that worked for you was zeroing out the knock retarding tables, so this would tell me that knock is in fact being sensed. Are your spring retainers kissing the guides? Is your exhaust banking around? How do your plugs look? Or, like Tuned Performance mentioned, maybe you're running the wrong knock sensor...

- Rob

 

The issue is that somehow he's getting knock retard kicking in without the knock sensor even hooked up.

 

If I remember correctly because its been awhile, I believe it is based on 1 knock count that is reported by the knock sensor for every 255 knock counts that the knock sensor detects. Knock counts that do not reach that 255 threshold are not reported, but are still being detected by the knock sensor, and they are still being stored, and reported via the ALDL which is what he is seeing when he datalogs. I know that Rauscher fixed this in his EBL code to have all knock counts being reported as actual and not have some being used as a hypothetical, so the former was a normal occurrence. Not sure what the OP is doing in his situation, don't know if he removed the knock sensor but didn't disconnect the battery when he tested it, but he is in fact able to keep the timing from being pulled by zeroing the retarding tables out, and he did seem to replace most of the areas that were being suggested, including the ECM...

- Rob

 

Yeah, with tip in and heavy change to the throttle, I get 20 degrees knock retard. At idle it is constant 6

 

17k ohm with one memcal and 18k ohms with a second

 

sounds normal for the knock filter but was asking about the ks itself should be 3.9k

 

Are you thinking Forced Knock Test related...?

- Rob

 

Yes I was thinking maybe it was and has a 100k tbi or early tpi ks installed.

 

Can test that theory by disabling it...

I think if you set this to a very high temperature and see if the problem goes away.

Knock Enable, Min Temp to Enable Err 43B Forced Knock Test

Hex location 0x29C

Note that 84Elky did a write up 8D knock retard logic

https://www.thirdgen.org/forums/diy-...ck-retard.html

Might want to peruse through here and see if any of it applies

 

It was 4100

 

Yes, I disabled it and the car runs great. BUT, it would be nice to have a functioning knock retard system