What did I do wrong?

[Steve Trumann]

I've ran my battery bank down pretty low 2 times in a row. Decided now would be a good time to charge them up all the way and do a equalizing charge. Did this during the day for the past 2 days so I could watch the wind turbine and watch the meters. Between the wind and battery charger, got the volts up to 60 volts. I've got a 48 volt system. Everything still OK at this time. Decided to let them charge for another 2 hours for equalizing. Batteries didn't gas a lot or get hot. So I put the battery desulphator on. The volts went down, current went up. I just figured it was doing it's job. By the way the specific gravity was good. Came back in a couple of hours, noticed my dump resistors were warm. Something is wrong.....
     I had set the dump voltage above the charging voltage. Unhooked the wire going to the gate of my FET's. Thinking this would provide protection for the dump circuit. By this time the voltage is up to 62 volts, I unplug the charger. Within 5 minutes the voltage drops back to 52 volts. Unhook B+ to resistor bank. Tried to set dump voltage control back to a lower voltage, no led indication saying so. Must of fried the controller. Checked resistance and voltage on the FET's. They're fried.

So what did I do wrong?
Steve Trumann

[ghurd]

"Unhooked the wire going to the gate of my FET's"

Floating Gate.
The Gate needs to be at 0V or 12V-ish.  If it is not, then it sort of 'leaks' and gets hot.

If disconnecting an N-fet Gate, it should be connected to 0V (battery negative, or Source).

Do the fets have a high Ohm resistor from the Gate to Source?  I was thinking yours did.
It should have been enough to keep the fet solidly shut off.
G-

[Steve Trumann]

Hi Glen, glad to hear from you.
I have 1 meg resistor between gate and source. Everything was OK at 60 volts, but that 62 volt seemed to be the magic point. Every time I think I got FETs figured out, I find there is something else to learn.
  Do you think the pulsing of the desulfator had any thing to do with it?
SteveTrumann

[TomW]

I always think odf FETs as "picky" and when I was learning solid state electronics we called them Fire Emitting Transistors. because they could fail spectacularly from minor transgressions on specifications.

When I learned this stuff my instructor was learning, too. 

No help, I know but there you have it nonetheless.

Tom

[Steve Trumann]

That's a good way to put it. No fire yet, but I sure can make them smoke! I was always told ( if you're not burning stuff up, you're not learning ).
Steve Trumann

[luv2weld]

( if you're not burning stuff up, you're not learning ).

Exactly!!!
But my problem is just like yours right now---- it's burned up, but I have no idea what the heck caused it, so I learned nothing.
Maybe after burning it 3 or 4 times I start to get a GBO! (Glimpse of the Blindingly Obvious)


Ralph

[Steve Trumann]

Thanks to Ghurd and others, I keep learning.javascript:void(0);

Steve Trumann

[ghurd]

The 1M resistor on the fet should have saved it.  I think.

No idea what is going on with desulfators in the big picture.  I suppose it could be possible the desulfator caused the fet failure.
G-

[Simen]

I''m starting to get a vague idea of the fragilities of fets after burning a few fets and more than a few fet drivers in my pwm dumpload circuit, and one of the thing are peaks/transients when the fet are switched off. Such peaks can easily get big enough to go through your 1meg resistor to the gate and exceed the gate's max voltage... (those peaks can also mess with your driver for the gate...)

You should protect your gate with a fast diode instead of your pulldown resistor. The diode will work as a pulldown resistor, but at the same time protect against peaks from the source.

[ghurd]

Such peaks can easily get big enough to go through your 1meg resistor to the gate and exceed the gate's max voltage... (those peaks can also mess with your driver for the gate...)

The 1M resistor was a safety.
The primary pulldown resistor is much smaller.
G-

[Steve Trumann]

Thanks for all the input guys. I think when I rebuild my FET board, I'll put a switch in line with my gates to shut them down to ground when I'm doing an equalization charge. Leave 60 volts as max on batteries while charging.
Steve Trumann