Colleagues;

I am in a bind on a 6HP motor controller. I have removed most of my hair, and remembered that I could ask the list (shaa..)

I have a 3phase 6HP brushless DC motor with three hall effects at 60 degrees.

My power section is a 150VDC bus (12 batteries.. this is transportation). I have six mosfets, in a six pack config.

I have six power diodes in the flyback positions as is typical. I have three more between the top and bottom fet to prevent freewheel in brake mode. Fets are APT.

Fets are driven by three IR2110 high voltage gate drivers. Neat parts, they convert logic levels to drive low side fet and high side N channel fet with a boost feature to generate the high VCC.

the commutation part is a Unitrode (now TI?) UCC3626N. This part does commutation, PWM, over current, and a few other goodies. It appears to be doing it's job.

If I run the PWM up open loop (no I or V feedback yet!) you can clearly see the switching waveform. Trouble is, the motor simply refuses to start out on it's own. In other words, it sits there on one phase and doesn't turn. If I run the PWM up a ways, and turn the shaft by hand, it trots along on it's merry way. If I run the PWM low and it stops, it is stuck until I turn it again! I have tried every concievable combination of phasing on the three wires (9).

Before you all call me a dolt, I know it isn't the basic nature of open loop with this particular motor. The customer can hook the same motor up to a commercial of the shelf controller and it starts up like warm butter. In open loop mode!

Although I am unsure if this is related, the design will shoot through if I change the PWM too abruptly. It promptly blows the fuse and I have to unsolder parts. Usually high side FET. Rumor has it the IR2110 will latch the high side if missused.

Oh, and there is a PIC on board. It controls batt voltage, coast, brake, and anti-cog features (future, when the friggin thing actually works).

Desperate in Pittsburgh

Chris Eddy

Pioneer Microsystems, Inc.