Knock Counts Rising with the Engine Off?? 7747

[Drnhandy]

Hey guys, I'm new here and been working on trying to get a gm crate 350 that had a carburetor to run right on my 7747 ECM.

Currently I have knock counts rising in the Knock Accumulator under Running Sample Count even with the engine off and key on, knock sensor plugged in or not. This sounds to me like a software issue. The code I am getting is 43 for the knock and 51 for the prom since I have added a Moates socket booster to be able to burn chips. I am running TunerPro RT ASDU $42 with $42-1227747-V5.9.3.xdf and $42-1227747-V5.2.adx

I will also get a random number of counts when looking at the Most Recent Sample in the Knock Accumulator. It can be anywhere from zero to 17 counts. The number will stay the same for all of the boxes until I restart the truck, which then it will replicate a different number.

David

[NomakeWan]

Are you sure the knock sensor wire from the ECM harness is actually plugged into the knock sensor and not something else? Additionally, Code 43 means that the correct resistance value is not being detected (not whether or not the sensor is working), and even if you're using a socket booster I don't think you should be getting 51 unless something is really not right about your setup. Could be wrong, never worked with $42 personally, but as far as I'm aware using something like Ostrich or Romulator never throws 51 unless you have something hooked up wrong or something is loose.

[Drnhandy]

Ok so I am following what you're saying and I removed and replaced the ecm with a stock aktr 1227747 from a different truck. I'm still getting the random knocks at every rpm and map value like before. I ohmed out the knock sensor with the connector removed and it is at 100K Ohms. I this the correct knock sensor reading for my current ecm set up? What about the ICM module? Not the one on the distributor, but the one next to the egr solenoid.

Edit: Also the 51 code went away so I must have a connection issue or something with the socket booster setup.

[NomakeWan]

And you confirmed continuity between the pin on the ECM harness that is used for knock sensor input and the knock sensor pin itself? I just want to be absolutely sure that you have the wiring going into the ECM correct. If the wrong thing is plugged into the ECM's knock sensor pin, that could potentially cause all kinds of weird stuff to happen.

I am not familiar with your ECM so I don't know what resistance it expects. The expected resistance value for the knock sensor will be found in the FSM for the car the ECM came from. For example, in the FSM for the '95 Corvette, in the section for diagnosing a Code 43, the expected resistance value of a single knock sensor is between 3300 and 4500 ohms. You will need the FSM for whatever vehicle the 7747 came from to verify what resistance value it expects.

That said, the Code 43 is not actually determining whether knock is occurring or not, and has nothing to do with excessive knock. All it is doing is checking to make sure that the correct amount of resistance is found in the circuit (using a voltage divider inside the ECM to check the circuit voltage). This resistance is not inherent to the function of the knock feedback system; it is provided by resistors that GM added to the knock sensors to allow this function to work. These resistors GM added are fragile and often will fail well before the sensor itself fails. While you can of course replace the sensor to get rid of the code, disabling the code works just as well (and doesn't affect the knock feedback system one iota).

However, in this case I highly suspect that there is a wiring fault, because you are getting a Code 43 in addition to strange behavior on the knock feedback system.

[Terminal_Crazy]

Check the continuity & condition of the cable. check that it isn't frayed on body work & shorthing out anywhere. (Just because it might check out fine everytime you pull the PCM.
Mitch

[PlayingWithTBI]

What about the ICM module? Not the one on the distributor, but the one next to the egr solenoid.

That is the ESC (Electronic Spark Control) module. The KS is wired (Dark Blue) to it. The picture is for my 88 C1500 which s/b good for your 7747. Here's a link to some Service Manuals - may help
https://www.gmt400.com/threads/88-95...manuals.43575/

Another thought - maybe your KCs are coming from a loose exhaust pipe, motor mount, or something else hitting the frame?

[NomakeWan]

Oh, this model uses an external ESC?

Then you need to check continuity between the sensor and the sensor pin on the ESC, and continuity between the ESC output pin on the ESC and the ESC input pin on the ECM. Make sure everything is going to the right place. If you've wired the knock sensor directly into the ESC input pin on the ECM and your ECM is expecting only filtered input from an external ESC, that could explain things.

[Drnhandy]

Thank you all to have contributed so far, I knew you were going to help! I feel like I'm finally going to get this problem solved. I plan on checking the continuities between the KS and ESC and also ESC to ECM and related grounds. If that checks out I will be inspecting wiring for possible inductance or damage. I will probably just temp in a new wire at that point. If all of those things pass I will be replacing the ESC Module.

[Drnhandy]

I failed to mention in the OP that this truck used to have a 4.3L engine. It was a 1991 C1500 4.3 2WD with a 16144288 ECM. From what I have found so far most everything has been the same aside from the injectors and distributor. I even ohmed out a knock sensor on a 1990 C1500 and it read 100k Ohms like the one I have installed.

After Testing the wiring so far this is what I have come up with: the KS to ESC Module circuit 496 wire shows continuity and not shorted to ground, ESC to ECM 457 shows continuity and 9v and not shorted to ground, and ground shows continuity and the Ignition wire had 12V.

Code 43 indicates that the ECM is receiving less than 6 volts for a 4 second period from the ESC with the engine running or that the system has failed the functional check. I am getting the proper 9v with key on and running on circuit 457. What am I missing that is throwing the code?

A interesting behavior I noted is that when the engine is running with the ESC plugged in I have sample counts under 20 typically. With the ESC unplugged and running it will show hundreds of knocks. I also read in the comments on tunerpro for knock counts, "there will always be a number. Need to watch cell to cell for increase for real knock. "So if this is true then when I display 12 knocks for instance in each cell depending on rpm and map then the only knock counts are what is different than 12. What would be the best way to view actual knock retard being added to the timing?

THANKS

[PlayingWithTBI]

Maybe start the engine and then start data logging? I know I get KCs from starting but, it's normally around 400 RPM.

[Drnhandy]

Yeah I was wondering more specifically what you use to watch how much timing is being subtracted, whether it is from the monitor or from a table or whatever else could be possible. I have been datalogging this whole time. As far as where I am at now, I repaired the ECM with the Socketbooster setup by replacing the socketbooster since two of the pins were broken. After installing and datalogging I observed 0 knock counts! The running sample count always accumulates, but I assume this is what EagleMark is talking about in his comment.

This still boggles my mind since the stock ECM I hooked up with the correct datalogging file, unless it wasnt the correct ARJT bin, was doing the same this as my broken custom ECM. Anyways, I will keep an eye on it while I am logging for the spreadsheet and see if it comes back. If you guys have any advice what to tune after the fuel spreadsheet I would appreciate advice as I am trying to get MPG out of this DD.