2006 Pontiac GTO PCM communication issue

Guy Educator Illinois Posted   Latest   Edited  
Resolved
Electrical
Network Communications
2006 Pontiac GTO 6.0L (U LS2) 6-spd (T56) — 6G2VX12U16L556260
U2105
U2108
No Communication With Ecm
Dtcs
Dash Lights
Gauges Drop Out

I have a 2006 GTO that I am mentally struggling with confirming my diagnosis of a faulty PCM. Car was running fine after doing some work to it, and after about 10 min of idling in the shop, battery light comes on, fuel gauge drops out, and ABS light comes on. Hook up scan tool, and can't communicate with ECM, but can communicate with ABS module and all other modules. Then sometimes fuel gauge comes back on, battery light goes off, and can communicate with ECM. I can make the fuel gauge, battery light, and ECM stop communicating by moving the shifter (manual trans) or stepping on the brake pedal which is weird to me, but may be useful data. Engine will run normally with the ECM not communicating, so that leads me to believe that this is related to the CAN part of the PCM and not powers, grounds, sensors, etc.

In ABS module, has U2105 (lost comm with ECM). In the Powertrain Interface Module, has U2105 and U2108 (lost comm with ABS), but I have never not been able to communicate with the ABS. Checked wiring between ECM and ABS for opens, shorts, and resistance, and all was good (diagram below).

Put Pico on CAN H and L at the PCM and at the DLC, and had first (4 channel) pattern. Channel A and B is at DLC, and C and D is at PCM. It is puzzling to me that it communicates “somewhat normally”, then drops, but still has a signal. The second pic is of the CAN at the DLC with the PCM unplugged, so it seems the PCM is causing the issue, but I am trying to understand what the PCM is doing internally that can cause a pattern like this.

I have tried unplugging a bunch of sensors to make sure nothing was bringing down the bus, have checked powers and grounds at the modules, and can't find reasonable explanation aside from internal PCM issues, but trying to understand a bit more of what may be happening inside the PCM that causes an issue like this. Any ideas from those more knowledgeable? Anyone object to replacing the PCM at this point based on the waveforms?

Also forgot to mention, I have 62.2 ohms on the CAN wires.

0
Helpful
Interesting
John Technician
Ontario
John
 

Strange that the shifter and brakes affects the car.. need to see how that is playing with wiring powers or grounds

0
Ð Awarded
Guy Educator
Illinois
Guy
 

I just checked that and seems that they both send inputs to ECM and EBTC. I just removed all fuses that feed the brake switch and the shiftlock control and problem still exists. With the brake switch fuse removed, the brakes no longer do it, but the shifter still does. I have now removed all connectors from trans (reverse lights, vss, shift lock, skip shift) just to eliminate any sort of input…

0
Ð Awarded
Laird Owner/Technician
Nova Scotia
Laird
 

It no looks like the pcm is still communicating and something is dropping the lines to ground. I was wondering if unplugging the pcm, taking it out of the bus and see if the brake and shifter will still cause the pattern to drop?

0
Ð Awarded
Guy Educator
Illinois
Guy
   

Good idea. I just tried that, and the pattern does not drop when moving the shifter or brake pedal, however the fuel gauge drops still, but the fuel gauge comes from the PCM, so I am not sure what the IPC is doing with that. To be fair, the PCM will stop communicating even when not touching the brake or moving the shifter, but if it is communicating, moving those will usually cause it to…

0
Ð Awarded
Geoff Diagnostician
Hawaii
Geoff
 

What if you jump the PIM, so that the UART network is not involved. Does that stop the issue? Wanna rule everything out.

0
Ð Awarded
Guy Educator
Illinois
Guy
 

That sounds like a great idea. Any guidance or suggestions on how to do that without disrupting the integrity of the communication at the DLC? Do I just unplug it and jumper the CAN H in to CAN H out and the CAN L in to the CAN L out?

0
Ð Awarded
Geoff Diagnostician
Hawaii
Geoff
 

Yes, just go "around it" using two jumpers at the unplugged connector. Ins to Outs.

0
Ð Awarded
Guy Educator
Illinois
Guy
 

I will give this a try tomorrow, just left the garage a little bit ago. Thank you for the suggestion and confirming my process.

0
Ð Awarded
Guy Educator
Illinois
Guy
   

UPDATE: I have jumped the PIM to isolate just the high speed CAN (and remove UART) and the signal appears to be about the same. Only thing on this bus is the ECM and the EBCM. None of the diagrams I have access to show where the terminating resistors are, but I am guessing they are in the ECM and the PIM. If a terminating resistor was in the PIM and I took it off the BUS, wouldn't that change…

0
Ð Awarded
Caleb Diagnostician
Indiana
Caleb
   

Yes by all means you can check for the terminating resistor with ohms meter on the module. Done it many times. Honestly every GM I have ever worked on had one resistor in the ECM an the second location varied. Your waveform does not look like a missing resistor tho. Looks more like the bias is getting pulled down. Here is a case study I had a couple years ago that shows a waveform with a missing…

0
Ð Awarded
Dmitriy Analyst
Ontario
Dmitriy
 

For the waveforms, what is used as the channels ground? If it’s not really a good ground (scope it relative to the battery negative while stepping on the brake pedal), that would explain CAN lines dropping below 0 (which does not make sense normally).

0
Ð Awarded
Guy Educator
Illinois
Guy
   

I used the Chassis ground (pin4) in my BOB for the scope ground, but good point, I should ensure that the ground in there is of good quality. UPDATE: Recaptured CAN signals using battery negative as the ground, and still had same waveform.

0
Ð Awarded
Tim Owner/Technician
Virginia
Tim
 

I had one of these and fought with it for a week- the pin connection at pcm was poor and intermittent because of swelling foam seal at connectors. Remove foam and no problem, maybe this will help.

0
Ð Awarded
Guy Educator
Illinois
Guy
 

Good thought, I have already replaced the CAN terminals as they were a bit looser than I liked. Problem still remained.

0
Ð Awarded
Stephen Technician
Tennessee
Stephen
 

This waveform is a puzzle to me. The signature and symmetry of the beginning and ending of the fault is interesting. The blank, transition, times are equal. The waveform of the beginning of the hi matches the ending of the low and vice versa. The bias voltage goes to 0v( I don’t see any going below 0v) but the pulses maintain at 1 volt. I don’t think it can be a short to voltage or ground…

0
Ð Awarded
Guy Educator
Illinois
Guy
 

This was my thought too. The waveform doesn't look like a waveform of something that is shorted, open, or just failing to communicate. It is still making a signal, just a signal that does not seem to make sense to me. I am glad I am not the only one that thinks it is weird. I will start isolating modules and see if taking one out of the network makes any improvement.

0
Ð Awarded
Wayne Owner/Technician
Australia
Wayne
 

this is an Australian built vehicle - you may be able to find some information about your issues common on some Australian websites - Not sure what you model might cross over to but probably a VZ Monaro( two door Commodore) - though I think we only has 5.7 litre version ( may be corrected ) visit such as injectronics​.​com​.​au may be able to help . Common issue with Fuel…

0
Ð Awarded
Guy Educator
Illinois
Guy
 

When the ECM stops communicating, the vehicle still runs flawlessly, but the gauges drop out and the ABS and Battery/Charging light come on.

0
Ð Awarded
Stephen Technician
Tennessee
Stephen
   

Since the PCM is suspected, I wonder if isolating it from the bus by bypassing the CAN lines and terminating the PCM can terminals with a 60 ohm resistor, then monitor the PCM and the rest of the bus with different channels to see which has the drop out. Only one side of each CAN would need to monitored since the fault effects both sides. It occurred to me that the PCM likely has a 120 resistor…

0
Ð Awarded
Caleb Diagnostician
Indiana
Caleb
   

I would agree with replacing the PCM. If your not losing 5v ref out of PCM or any powers and grounds into the PCM when the fault happens then I'd say it's bad. I personally would not get hung up on the waveform. I had a 14 Mini the other day with a no comm ECM and the waveform out of it looked good on scope but the other modules and scanner didn't like it. A new ECM fixed it. This is a good…

0
Ð Awarded
Guy Educator
Illinois
Guy
 

I agree and also suspect the pcm, but want to do my due diligence since PCMs for this car are almost unobtainable and cost almost $900. I just to need to be sure that nothing else is interfering with it.

0
Ð Awarded
Glen Technician
South Carolina
Glen
   

After you unplugged the PCM, then reconnected, did it temporarily correct itself? I have a story from the “seen-it” files. 6.0L GTO. It begins with it twaz the heat of the summer… had time to observe it over multiple events… ends with getting a PCM from FlagShip One, and fixed. Found it caveman style.

0
Ð Awarded
Guy Educator
Illinois
Guy
 

I have had it disconnected and reconnected multiple times at different stages of testing and the problem remains. Flagship One looks like my only option to find one and it is $850, so I want to make sure I am not firing the expensive parts cannon.

0
Ð Awarded
Guy Educator
Illinois
Guy Update
   

So as of tonight, I have jumped the VIM and the CAN signal still drops below zero (just like the 4 channel waveform from Sunday), I have jumped the EBCM and it looks just like the 4 channel signal/VIM removed signal, and still when I remove the ECM from any of these scenarios, it looks like a perfect signal and everything goes back to looking like a CAN signal should. I also unplugged the BCM…

0
Update Ð Awarded
Caleb Diagnostician
Indiana
Caleb
 

You need a PCM brother. You have diagnosed it correctly. The reason the packets get larger amplitude is because when you disconnect the PCM you probably lose a terminating resistor. Check out my other reply with the link to the case study I posted a couple years ago. I show a waveform with one resistor removed. Both waveforms will increase slightly in amplitude when a resistor is missing.

0
Ð Awarded
Guy Educator
Illinois
Guy Update
 

Status update: I got a used PCM for this since no one has one in stock aside from Flagship. After 2 days of messing around with TLC and tis2web, I finally got it programmed and linked to PIM. At KOEO, dash gauges seem to work, there are no communication codes in any modules, but my CAN signal at the DLC still looks the same as with the old ECM in place. I am unable to start it yet because it…

0
Update Ð Awarded
Guy Educator
Illinois
Guy Resolution
   

Resolved: I finally had some time today to do some more investigating on this vehicle. The new PCM is in, and communicates all the time now, so while the PCM may not have been “the problem” it seems it was “a problem” since I was able to communicate with the new one and was not with the old one. The CAN signal still looked the same, and I was getting a few other weird communication codes and…

0
Resolution Ð Awarded
Dmitriy Analyst
Ontario
Dmitriy
 

Good job on finding bad ground! I‘d say your waveform is good now. Little waviness is not going to hurt anything.

0
Ð Awarded