Best strategy for diagnosis

Larry M · January 16, 2022

This concerns an early 2015 GMC Sierra 2500 HD we purchased about six months ago. I am trying to diagnose a parasitic battery drain (120 mA after 20 minutes, never gets lower than that). I did not get very far with the fuse-pulling trick. My son has a GM Tech2 and MDI, etc. and thought we should look to see if a module was staying awake. When we started poking around to see what functions the tool had, we ran tests on the CAN bus and learned that the segment that uses pins 3 and 11 reported that pin 3 (CAN-) was stuck at the active CAN-low voltage, 1-volt. Just the high side was switching. There's a scope function on the tool, and its live data showed the same thing: CAN-H was switching but CAN-L was stuck at about 1 volt.

We decided that this was important enough to chase since it needed to be fixed and could easily be related to the drain, too. My thought process was:

If there were a cable short, it would be to +12,+5, or ground. 1 volt isn't in a cable, so it's not a cable short.
If there were a cable open, that line would be swinging all over the place; it's not.

So it's likely a fault within one of the CAN modules.

But how to tell which one? Would you have to unplug all the CAN modules that use the 3-11 pair to find out which one is holding the line? Some of them are easy (we did unplug the EBCM, just because it's easy to reach), but I'm not looking forward to ripping the dash apart.

So what is the best way to go at this problem? Your advice is appreciated.

silveradosid · January 16, 2022

just to confirm, tech2 will not scan the truck properly, you need to use gds2

Larry M · January 16, 2022

My bad--it was GDS2. My son has the hardware and I was overwhelmed with acronyms. GDS2 connected to GM-MDI connected to OBD II port.

Larry M · January 17, 2022

Bump.

Larry M · April 5, 2022

Solved!

After a false start or two, I devised and executed a divide-and-conquer scheme and isolated the parasitic drain in around 30 minutes.

The original symptom was that the current draw never dropped below 120 mA (0.12 Amp). Testing was done by disconnecting the negative battery terminal and connecting a multimeter (on the 10 amp scale) between the negative battery terminal and battery cable and waiting 10 minutes. Waiting additional time never reduced the current further.

I started by exposing that fuse panel that sits atop the battery, taking off the fender brace and removing the cover. Inside are several high-amperage fuses ranging from 60 to 175 amps. A couple of these are bolted down including the ones to the starter and alternator. Three simply plug in using high-amperage connectors.

My plan was to see which one of these cable had the drain and use the schematics in the upfitter manual to see where the draining cable went. I first unplugged all three cables. Reconnecting the ammeter and waiting 10 minutes showed drastically reduced current draw, below 0.01 Amp.

Of the three cables, one has a single plug and the other two share a two-contact plug. I first disconnected the ammeter, plugged in the single cable (because it was easiest), reconnected the ammeter and waited. Current was still below 0.01 Amp.

Now I had to distinguish which of the two cables had the draw. The peak current when the ammeter was connected was around 5 Amp for a few seconds, then rapidly dropped to 1 amp. This was low enough that I could use a clip lead between one of the cables and the fuse panel. This caused just the other contact to still be isolated. Reconnecting the ammeter and waiting showed that current was back up to the high value of 0.12 Amp.

I wanted to make sure that the drain was actually on this cable. It could have been caused by a small current on this cable enabling some relay or computer that was draining current somewhere else. So I disconnected the ammeter (that allowed current to flow to the negative battery terminal), and then replaced the clip lead with a second ammeter, then reconnected the first ammeter and waited. Sure enough, the 120 mA was flowing through this cable.

Several minutes study of the upfitter manual showed that the cable in question powered only the fuse panel at the right end of the instrument panel.

Time to start popping fuses and checking the ammeters, one fuse at a time. I thought this would be tedious, but the second fuse I pulled dropped the current back to the low value of 0.01 A.

Back to the upfitter manual again. That fuse, F7DR, powers a signal that goes through a relay in the Body Control Module (BCM) and powers one low-beam headlight and a parking light on the other side. These were passive loads, on the far side of the BCM. Those loads couldn't have been responsible for the relatively small leakage. The problem HAD to have been in the BCM. I replaced the BCM and sure enough, the current dropped to the low value of 0.01 A.

That fixed everything but the nightmare problem of getting the vehicle to accept the new BCM. That will be the subject of another post.