Overvolting a controller

DavidDymaxion said:

First the more generic questions: What are the limits on overvolting a controller? What breaks first?

I can see how switching would get dicey at too high a voltage. What if the controller was at 100% duty cycle, could you then switch in a higher than spec voltage?

Click to expand...

I bet the first thing to go would be the capacitors, but if not as soon as you had a spike that exceeded what the IGBT/MOSFET can handle they would start a cascade failure. However if it's smart enough it wouldn't even turn on and just give you an over voltage warning.

Sizzle Sizzle Pop Pop!

Don't even try, sell your controller and buy a better one. It will be cheaper in the end.

DavidDymaxion said:

First the more generic questions: What are the limits on overvolting a controller? What breaks first?

Click to expand...

Not answerable generically.

DavidDymaxion said:

I can see how switching would get dicey at too high a voltage. What if the controller was at 100% duty cycle, could you then switch in a higher than spec voltage?

Click to expand...

While the switch will only see it's own voltage drop when turned on, the freewheeling diode is reverse-biased whenever the switch (IGBT or MOSFET) is on. The input filter capacitor might also take exception to this abuse.

The very highest voltage rating that electrolytic capacitors come is 450VDC and that is what the limiting factor is in, e.g., the Zillas. Film capacitors of the kind typically used in motor controllers start at a 450VDC rating and go up.

There should always be some buffer between the maximum operating voltage and the minimum rated voltage of the components to accommodate the inevitable dI/dt overshoot and/or ringing (though one should strive to minimize both).

FWIW, the Soliton1 will shut itself down and open its contactors if it sees the pack voltage go over 342V at any time, so not much point in trying to pull this kind of stunt with it.

You can only let the smoke out once...

Many thanks for the responses. I have a sepex motor. It would be nice to buy a high power sepex controller but such a thing doesn't exist. Kelly now has a 144 kW nominal controller -- that's better than the 72 kW of yesteryear, but it would be nice to overvolt for more (one can always dream).

The Kellly has a 180 V limit -- how close would the experts crowd that limit?

My Kelly KDH14100D (180v limit) give me an overvolt signal (flashing red light) last time when I test it to find the max permisible voltage.

For mine, 180.9v seem the maximum. I performed the test with 56 Headway cells (around 3.23v each.) Under 180v the controller seem proprely work, but I don't tried to push 1000A at 180v.

I think it will be possible to run a 54s pack (173v nominal) with this type of controller because cells drop at 3.34-3.35v after a charge. That give 162Kw out of controller if you have a really strong battery pack...

But the voltage probably will be too high with regenerative breaking!!

Yabert said:

But the voltage probably will be too high with regenerative breaking!!

Click to expand...

Exactly. Fresh off the charger, if you go down the road and regen heavily, that voltage will climb fairly fast. If you overvolt, the controller will likely shut down, so your regenerative braking stops, your motor is no longer powered and you have to try to pull over quickly.

Yabert: Great real life data point, many thanks! Good point, that regen thing would make the voltage rise. For a race I could turn off regen and put in some more cells.

Frodus: Good point. Kelly says they limit voltage, I assume they would ramp down current, but that's a good point it might just be a full shut down. That would be worth checking.

Here's another thought: Split the brush and field leads into pairs, and run 2 controllers. That would be 288 kW, I think I could be happy with that.

Yabert said:

My Kelly KDH14100D (180v limit) give me an overvolt signal (flashing red light) last time when I test it to find the max permisible voltage.

For mine, 180.9v seem the maximum. I performed the test with 56 Headway cells (around 3.23v each.) Under 180v the controller seem proprely work, but I don't tried to push 1000A at 180v.

I think it will be possible to run a 54s pack (173v nominal) with this type of controller because cells drop at 3.34-3.35v after a charge. That give 162Kw out of controller if you have a really strong battery pack...

But the voltage probably will be too high with regenerative breaking!!

Click to expand...