Part I: the-problem-with-shorts-part-i-t628344.html
Part II: the-problem-with-shorts-part-ii-t628356.html

So, the officer didn't ticket you, but he did let you know your brake lights were out. You got home and found a blown fuse, but the lights worked fine after you replaced it. Two days later the old lady next you at the traffic light signals to roll down your window. They're out again, and the fuse is blown again. Not overlooking the obvious, you replace the bulbs and brake lamp relay "just because", then cross your fingers.

No love. A day later it's blown again, and all you can think is, "Uh-oh." Intermittent short circuits are the most difficult problem an electron can cause, and you already know it won't be easy. You can of course paper-clip the fuse and see what smokes, which might work if you don't ignite the whole harness (or the whole car). Leaving that aside, how do you find it? Circuit tracing tools and sniffers won't work to find a short that isn't there. Most often it will prove to be something chafed that's contacting the chassis with bumps or vibration, but tearing out whole harnesses for inch-by-inch inspection is obviously undesirable.

There are two places you need to go. You know by now that the first place is the SM for a complete wiring diagram. The second place is the parts store for some inline fuse holders and a fuse assortment. There's only one good way to find a hide-and-seek short, and that's with staged fusing and successive approximation.

https://drive.google.com/file/d/1rJAe4Z ... sp=sharing

That's a digital reproduction of an actual case from about 5 years back, a '10 Armada with a popping brake fuse. The first time it rolled in we tried the obvious, new bulbs and a new relay, along with a cursory inspection of whatever wiring was exposed. We told the customer we couldn't promise anything, and warned him that things would get complicated if that didn't work. Sure enough, it was back in a few days.

Looking over the WD, you can see that the brake circuit is segmented. Like most circuits on a car, it goes multiple places, and in this case it traverses four different harnesses, Engine Room (E), Body (B), rear hatch (D), and Main (M). If you don't see the M, it's because it's on the WD for the shift-lock, indicated by the off-page arrow on the left. It actually goes to the BCM as well. With that many wires running to that many different places, the only good way to locate the problem is to narrow the possibilities. First locate which leg the short-circuit current is flowing into, then inspect that leg, and if necessary break it down into even smaller portions. The X's indicate where we snipped wires and installed staged fuses with successively smaller values, allowing the circuit to tell us where it hurt. Obviously you must use real fuses for this, since a polyfuse will reset itself. The real fuse will serve as a "memory" for where the current went. Since the lamps needed to work normally while we awaited the next blow, we checked the wiring between the fuseblock and relay and determined that the primary supply wires were 16~18AWG, thick enough to support a 15A fuse instead of the factory 10A. That allowed us to place a 10A at the relay supply, which would blow if the short was anywhere between the relay and the SMJ connectors (B40 and B107). We then wired three individual 5A fuses on the body sides of the SMJ's, plus one where the off-page wire to the shifter entered the Main harness. That one powers the shift-lock solenoid and could certainly cause a problem, but it's also a perfect example of why your WD's must be complete. Finally, we placed a 3A fuse on the body side of the hatch connector, in case the problem was in the door harness. The only leg we elected not to fuse was the ABS, since that was an input-only wire and probably the least-likely suspect. We then sent the customer home, with instructions to return when something changed.

It barely took an hour, but the symptom, oddly enough, was that the lamps still worked but the truck wouldn't come out of park. The 5A to the shifter was blown, and knowing where to look, we found a chafed spot under the console. A restraint had failed, allowing the harness to vibrate against the sharp edge of some stamped-metal. The whole exercise, including the customer's "blow it" drive, ended up taking less than 5 hours. We left the extra fuseholders in place, just installed 30A fuses that would never blow and taped them safely back. The customer wasn't happy about the $500 bill, but he didn't complain, realizing from our explanations that it could have been much, much worse.

We could have stripped the interior and engine box chasing ghosts, right? Many shops would have. Maybe we would have stumbled on the problem. But with a little strategy and $15 worth of fuseholders, well... who's afraid of that ghost? If you take the time to do it right, even the "intermittent monster" doesn't need to be scary.

Hope these articles make your life a bit easier the next time electrons clash. Happy motoring!