Fuse an IGBT -> DBR circuit

L

Thread Starter

Leggir

We had a catastrophic failure of a dynamic brake and I've been tasked at coming up with a solution to prevent/restrict future failures.

I work on VFDs daily for artificial lift, primarily pump jacks. We're using Unico (a division of Regal Beloit) branded drives that are built in Wisconsin. They like many others use DBR (Dynamic Braking Resistors) to control regen and speed.

I've seen several units overheat and get burn marks on the enclosure, although generally the thermal switch will shut down the drive and prevent it from restarting until it cools by at least 40 degrees. Recently we had a catastrophic failure of such a brake. It suffered severe arc damage so much it put a hard ball sized hole in the side of the enclosure and a finger sized hole in the end where the terminal block is. There's a ceramic terminal block where the leads from the VFD (110 degree C) are connected to the 200 degree C conductors going into the resistor bank that turned to glass and melted.

In my investigation I've looked at other brands of VFD's including Yaskawa (Weatherford), Toshiba (Lufkin GE), and Allen-Bradley (Rockwell) and most are similar. Everyone has a fused braking controller and none have any fuses in the circuit from the braking IGBT to the resistors. Even the Rockwell automation acknowledges that a short circuit may occur and it's up to the thermal switch to interupt the drive.

Thus my question is why wouldn't there be a fuse directly in the resistor bank circuit (preferably in the VFD cabinet where temperatures are -25 to 40 degrees C)? If it's possible, why isn't everyone doing this? If it's possible should I be sizing the fuse just to protect the wire or simply to interrupt a SC?

Specs:
20-60 HP VFD
460VAC 3Φ
650-700 VDC
Carrier Frequency 3600-7200 Hz
DBR rated at 10% of drive output. (50 HP, 37KW, 3.37 KW @ 17.6 Ohms)
DBR made by PowerOhm resistors.
 
Top