Modeling a motor winding

See the section of a fairly long thread, starting here, where I had a lot of expert advice form Tony Williams:

Newsgroups: sci.electronics.design Subject: Re: Simple 555 PWM - disappointing performance Date: Thu, 26 May 2005 09:43:18 +0100

Thanks. I snagged the model, but only had time to browse the subsequent discussion a bit; will read it more carefully later. I also didn't read the fine print when I dl'ed LTSpice. Is it a demo with limitations? or is it the real deal?

Anyway, it's a stepper, with much simpler physics than a PMDC. I'm more interested in what's happening on the gate side of the MOSFET. Wondering if I can't make it work with an opamp comparator for current sense/limiting instead of the more traditional PWM chopping. Quad opamps are pennies these days. I think I can squeeze a whole second motor onto the PIC if it can off-load the current limiting part. (More for futzing around's sake than any real economy. How many controllers would I need to make to recoup the $1.50 I save on the second PIC? What an upside down world. The protection diodes alone cost more than the power MOSFETS.)

I've seen it done. It was used as a current control loop on a motor controller

Robert

You can apply PIC using pulse skipping technique, you have fixed pulse width and use ADC to monitor output voltage, if the voltage goes above threshold, you skip pulse. It very simple and effective but poor ripples performance. The advantage of fixed pulse width that the current ramp is the same (under discontinous mode).

The LT model is the real thing.

The key thing is the current ramp that feed into transformer (or inductor). Make sure the core does not get into saturation by excessive current ramp, where magnetic flux density exceed the core specification (Bsat). There is simple equation to use in Philip Ferrite Core datasheet, this is good starting point for transformer/inductor development.