Simple Power Sink Circuit

Question

I've got a test bench for brushless DC motors, which is supplied with a regulated power supply unit. The motor itself is driven by an electronic speed control (ESC), which feeds energy back to the power supply when actively slowing down the motor (regenerative braking). My power supply cannot sink current, which leads to non-transient high voltages when braking or spinning down. This would damage the test bench's measurement equipment, causing it to terinate the test run.

I have (unsuccessfully) tried to sink the power into a MOSFET crowbar circuit, which turned out to be a bad idea. Although rated for a continuous \$I_D\$ of 72A (at \$V_{GS}\$ = 10V, though), the transistor got fried at ~500 mA already, despite a solid connection to a large metal beam as heatsink.

Before frying the next components, I'd like to ask:

1. Is the (a) crowbar circuit applicable in this case? It does not seem such a good idea to me anymore.
2. What would be a better power sink circuit?

Answer 1

1. It could be. But you need to keep the power on the mosfet within limits.
   • Electrical limits.
   • Thermal limits.

Although the datasheet might say it can do 150 Watt, you can't go from 0 to 150 Watt in an instance when it's at room temperature. It will destroy itself from the thermal stress. I'd keep the TO-220 around 40 Watts maximum.

2. A variable frequency drive with brake uses a resistor with chopper. This can be as simple as a comparator enabling the resistor when the DC-Link voltage exceeds the threshold. This is the voltage on the capacitor bank.

(source)

This is basically what you did with the crowbar, except with a resistor as the dissipating element. Resistors can handle more abuse than semiconductors in linear mode*.

*Infineon appnote: Linear Mode Operation and Safe Operating Diagram of Power-MOSFETs (J . Schoiswohl)