1. Attachments are working again! Check out this thread for more details and to report any other bugs.

Dead Inverter Strikes: P3004

Discussion in 'Gen 3 Prius Care, Maintenance & Troubleshooting' started by kbeck, Jul 11, 2018.

  1. kbeck

    kbeck Active Member

    Joined:
    Feb 10, 2010
    420
    274
    0
    Location:
    Metuchen, NJ
    Vehicle:
    2010 Prius
    Model:
    III
    Well, it finally hit. I have a built-in-2009, model year 2010 Prius 3 which now has around 135,000 miles on it. This past Thursday was driving with the wife from New Jersey to Boston, stopped for lunch in Wallingford, Connecticut, about half way. It was a hot day, 92F or so. No problems with the car.
    Came out, tried to start the car. It was all the warning and alarm lights, all the time, plus "Check Hybrid System". Sounded like death.
    As it happened, there's a Toyota dealer about 1 mile from the Panera Bread Parking Lot of Doom. They were willing; and, after a longish wait for the AAA truck, got the car over there. They're good people: Figured out it was the "computer" in the engine, gave us a 2012 Prius as a loaner, and we were on our way, albeit four hours late.
    Got the car back today. Error code P3004 which, as you guys probably know, is the Intelligent Power Module. Solution was to "Replace the IPM Transistor". Ha. That's one of the Big Guys bonded onto that massive heat sink in the basement of the inverter. Interesting. The list of replacement parts included:
    • TRANSISTOR KIT, POWER
    • PLUG KIT, INVERTER D (more characters, but cut off for some reason)
    • SEAL PACKING
    • GREASE X 23 7884 4 (two of these..)
    • GASKET
    Back in 2010 when the original recall and re-burning of the firmware was done, here on Prius Chat there was quite a bit of speculation as to just What Had Gone Wrong. Based upon the documents provided to the NHTSA, my guess at the time was differential expansion between the silicon dies of the transistor and the massive heat sink upon which said dies were mounted. The thought was that during a hard acceleration, in the first few tenths of a second, the transistor(s) would heat up faster than the underlying heat sink. Because the coefficient of expansion/contraction of the silicon and underlying heat sink would be (a) a bit different, being different materials and (b) the difference in temperature being sudden, that could pop the die right off. Followed immediately by Death of Inverter.
    It appeared that the recall did two things:
    1. A bit of diagnostics: Apparently, resistances and voltages can be measured remotely down there. Therefore a partially popped-off transistor could in theory be detected. So, there were reports that people would take their cars down for the recall, then be told that all the parts were on back order (i.e., new parts were required). That didn't happen to me.
    2. A burn of a new firmware, presumably to slow down the onset of All That Power a bit, allowing Not So Much Differential Expansion. This was presumed to result in a slight reduction in initial acceleration, and there's some indication that that was happening. (Once temperatures had equalized, which would happen pretty quickly, full power and acceleration could resume.)
    When the extended warranty showed up there was a little hypothesizing, mainly along the lines that a partially popped transistor might not be detected right off, and, that given time and heating/cooling cycles, such a pop would be in that transistor's future. Well, sure enough, looks like that here.

    Back before the recall went in and immediately afterwards when people were dragging their Priuses to dealers, the initial replacements were of the entire inverter. I heard some mention a few years back that the dealers weren't replacing the entire inverter any more on these types of failures: That "Transistor Kit, Power" is probably a sealed module with all the high-power transistors upon it. I suspect the grease is of the thermally conductive type. Further, I got called on Tuesday around noon to come and get the car, with the parts guessed to be arriving on that same Tuesday. By this, I take it that dealerships (and this dealership in particular) are getting Really Good at getting to those transistors, which are buried in the bottom of the inverter box.

    Final note. I told the dealership when giving them the car in the first place that the 12V battery had been replaced around 45,000 miles and not since then, and to Make Sure that the battery was sound before jumping to conclusions. That got me a mild eye roll, but no markup on the repair order that they had looked. On the other hand, by now, dealerships are getting better at checking the obvious on Priuses, so what the heck.
    KBeck
     
    4est and The Critic like this.
  2. GardenWeasel

    GardenWeasel Member

    Joined:
    Jul 25, 2009
    76
    51
    1
    Location:
    Columbia, SC
    Vehicle:
    2010 Prius
    Model:
    III
    Great report and great information. Thanks!!