Completely died while driving. 12v good, but no power

Sounds like the main fuse is blown. Rodents are the most common cause of that malady.

No warning means that the "car" thought you turned it off, as in turned the key off. Until it receives the ignition signal, it will not, in fact cannot, restart.

 

Somebody else will probably chime in shortly; I'm more knowledgeable about gen 3, but I think I have read here that it is not uncommon for the coolant pump that cools the inverter electronics to fail after umpty thousand miles, and when it goes it goes in a way that takes a fuse with it. The things that are and aren't downstream of that fuse give the same sort of "some things work, but instruments don't" picture that you're describing. "AM2" might be the fuse.

If you check that fuse and find it blown, don't just pop in a new one before swapping the pump, or the bad pump will just eat the new fuse too.

 

We have a winner!!! lol.

Inverter coolant pump failure blew the AM2 fuse in the engine compartment fuse box. Unplug the coolant pump at the plug shown in the upper left of the photo and then replace the fuse. Then replace the pump, ASAP. It's located directly behind the driverside headlight. Car can be driven at low speeds / short trips even with the bad pump, but should be minimized.

 

The two I had have seen fail did not do that - they just stopped pumping and the inverter got too hot. What causes the AM2 fuse to blow, an internal short in the pump? If that is the case, can it be seen electrically, for instance by measuring the resistance between the two plug contacts?

 

Yes, there are two failure modes, each with very specific symptoms.

1. Inverter cooling water pump fails, but does not blow fuse-many cars are probably driving right now in this condition and the owners don't even know it.. Symptoms only show up if driven long enough/fast enough, hard enough, to cause temp issues with the inverter. You'll also see the AC working intermittently even if temp doesn't get quite high enough for the alarm.
2. Inverter cooling water pump fails and blows fuse-opens the AM2 fuse, which kills the dash board and the car...scary as heck if you're doing 70 down the highway. Sometimes, with this mode, there will be a heavy acrid odor and black melted goo under the pump. The goo is just potting material on the bottom of the pump body that got hot from the winding failure.
3. Honestly, I have never measured from the plug to ground to see what value was present with a good pump. Give me a few minutes and I'll check it out.
4. Ok, I measured from the male pin at the connector in the above photo on my 2007. It was 0.1 ohms to ground, so I'm going to assume that connector is on the ground side of the pump motor.

OK, so below, the top photo is the bottom of the ICWP I removed from the 2005 Driftwood Pearl that I bought last Fall. Although it melted the potting material, it never blew the AM2 fuse. I drove that car home from across town, about 25 miles, iirc. It measures about 200 ohms each way across the pins.

The second photo is a new pump, never installed. Just wanted to show orientation of the pins. I measured 5 pumps I had sitting on the shelf. 3 old ones and two brand new ones.
First measurement is with leads + on left and - on right. Second measurement is leads reversed.
Documented exactly as the meter read, some where negative. NBD, just whatever electronics are in there are playing with the readings.The 200 ohm one is the one melted on bottom.

2005 driftwood P0A93......206 ohms...…..200 ohms
2005 blue P0A93...………-600kohms......700kohms
Not tagged...………………-1.3Mohms.....1.2Mohms
New...……………………...1.3Mohms...….-1.1Mohms
New...……………………...1.3Mohms...….-1.1Mohms

 

Are these all Honda pumps? If so the most likely explanation is that they changed manufacturers and the new one is making not very good pumps. The first replacement I did was with a Dorman pump, and it did not last long at all. The second was with a Honda pump and it is still going. But it has not been all that long, so I cannot say yet that my experience differs from yours.

Has the coolant in that loop ever been changed? I suppose if it was very acid or full of metal bits or something along those lines it might cause the pump bearing to fail rapidly, and that added load would eventually burn out the motor.

TMR-JWAP's measurements are interesting. It looks like new/good pumps are >= 1.1 MOhm and pump failures are associated with lower resistance than that. AM2 is I think a 15A fuse so the pump resistance would need to be below 1 Ohm to blow it all by itself (not taking into account other loads on that circuit.) Of course this is a "cold" measurement, for all we know the resistance might be quite different when the pump is running. At least the resistance measurement presents a relatively easy test. Caveat, the resistance might not be the same in both directions, and one doesn't generally know the polarity of the resistance test on a multimeter.

 

Yeah, I'm pretty sure these are electronically-commutated motors: with a measurement at the terminals, you're not really measuring a wire winding, but the input end of some electronic circuit. Also the failure might not be in the nature of damaged windings, but something more typical of electronic failures. Say, two transistors on the same side of an H-bridge that end up with their switched-on time overlapping for some reason ... oops!

It can be really hard to come up with a simple story for what such an electronic circuit ought to "look like" to a meter at the terminals. Notice that even the one with the lowest resistances shown (200-ish ohms) would be drawing less than a tenth of an amp at 13.8 volts if it were a simple DC circuit, hardly enough to make bad smells and goo and blown fuses. But something that circuit did while it was operating sure did.