18FET explodes when ignition on

[Skedgy Sky]

One rainy day I parked my bike and turned wheel lock on as usual.
I come back 5 hours later, turn wheel lock off and controller blows.

Today I replaced these FETs and did the usual resistance tests for shorts - everything looked fine as far as I could tell.
I plug power to the controller... everything good.
I flip the ignition switch and all the 6 FETs on the phase that I replaced blew again!

Seems FETs blew both times ONLY when controller was powered AND ignition was on.
With controller powered and wheel lock on, FETs did NOT blow.

Why didn't replacing the FETs work? Why is the controller blowing up when ignition switch is on? It was working great for weeks before the first incident.

Edit: Probing around with my multimeter... One of three of the 0 ohm gate driver resistors is not showing any continuity. Is it related to cause of all this?

Edit 2: I put two good FETs in, one on high side (connected to one of the good gate drivers)and one on low. Power to controller - good. Ignition switch turned on - *boom*. Wasn't as bad as if I had put in 6FETs, I suppose.

 

[dnmun]

don't know anything about wheel lock.

usually the 0 ohm surface mount resistors are there as jumpers to bridge a trace on the pcb. the gate should have 100 ohm.

it does sound like both hiside and loside mosfets are turned on at the same time and so it is shorting the battery positive to ground through the mosfets. a dead short through the mosfets is more current than most of them can handle.

 

[Punx0r]

Is it this controller?

http://endless-sphere.com/forums/viewtopic.php?f=30&t=55877&p=858377#p858377

 

[Skedgy Sky]

Is it this controller?

http://endless-sphere.com/forums/viewtopic.php?f=30&t=55877&p=858377#p858377

*pause* .. ...... yes. That would be this same controller, but it worked for a while!
I don't think the power I was running through it was related to it randomly blowing up on ignition after leaning against a pole for 5 hours... untouched.

 

[dnmun]

did you even test to see if the mosfets were both turned on at the same time?

 

[Skedgy Sky]

did you even test to see if the mosfets were both turned on at the same time?

No. How would I do this?
Take out FETs on the phase causing the problems and looks for voltage using DMM on drain or source contacts on controller board when ignition is on?

 

[dnmun]

i cannot understand your question. to test to see if the mosfets are turned on you would measure the gate voltage on the mosfets.

obviously you have to disconnect the drain of the hiside mosfet from the battery in order to turn the controller on without shorting them out but since they are shorted already all you are looking for is to determine if the processor has turned on both mosfets at the same time because of a failure of the hall sensors or whatever could cause it.

but if your mosfets are blowing up it is because they are turned on and conducting full pack voltage to ground, or because one of them is shorted when you turn on the other one and that shorts the battery B+ to ground. so you have to test the mosfet to see if it is shorted too.