My Kingdom for a Diode.
Apr. 15th, 2008 10:27 am![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png)
When your xray machine goes *WHOMP* and lets the magic smoke out, don't expect i'll be able to just snap my fingers and fix it. It'll be fixed when i can trace down the short that keeps burning up the current-limit resistor on the charger board. Yes, this repair only took 15 minutes last time; those circumstances were slightly different, despite the similar symptoms. No, swapping the charger board didn't fix it; so the short is elsewhere in the machine. Yes, it reeks of death cooked phenolic in here. Yes, that's the thing that blew up. Excuse me wtf r u doin NO DON'T TOUCH THAT DANGER DANGER! There's 400 volts DC stored in those battery racks, it WILL bite you; and i can just guess who you'd (try to) blame if you got zapped, pigfucker. In fact, just get out. Shoo. Shoo! AWAY WITH THEE, KNAVE.
Three days later, i've sorted out that the problem isn't in the battery racks (by pulling all 30 cells and running each one through the dynamic load-test analyzer, which takes a few hours per cell). Manufacturer tells me that if the batteries are all good, the short might be in one of the 8 FET boards instead. A new FET board is some $1200. Isolate the feed runs to the FETs, ....none of these are shorted? Hmm, the short goes away when i unplug the oem-repaired (@ $450 per) replacement charger board.
Comparison of the oem-repaired board with a known-good board out of a different machine, clearly illustrates why i can't trust the jackasses at $vendor to actually *fix* something. Looks like they just replaced the blown-up resistor, without even bothering to check any of the other components on the board, or testing it in a live system; as diode 10 is clearly shorted. Oh look, D10 is shorted on the board we just blew up, too. Replace D10 on both boards, and the system works fine. OH FRABJOUS DAY.
Three days later, i've sorted out that the problem isn't in the battery racks (by pulling all 30 cells and running each one through the dynamic load-test analyzer, which takes a few hours per cell). Manufacturer tells me that if the batteries are all good, the short might be in one of the 8 FET boards instead. A new FET board is some $1200. Isolate the feed runs to the FETs, ....none of these are shorted? Hmm, the short goes away when i unplug the oem-repaired (@ $450 per) replacement charger board.
Comparison of the oem-repaired board with a known-good board out of a different machine, clearly illustrates why i can't trust the jackasses at $vendor to actually *fix* something. Looks like they just replaced the blown-up resistor, without even bothering to check any of the other components on the board, or testing it in a live system; as diode 10 is clearly shorted. Oh look, D10 is shorted on the board we just blew up, too. Replace D10 on both boards, and the system works fine. OH FRABJOUS DAY.