Re: [CAD_CAM_EDM_DRO] 5-Phase steppers

On Monday 17 January 2005 05:04 pm, Stephen Wille Padnos wrote:

> Well - I did a little Googling and found this paper:
> http://ee.tamu.edu/~empelab/power_tran98.pdf
>
> It's an electrical/mathematical analysis of a 5-phase stepper, so all
> the information is in there. I didn't want to slog through the math,
> but I did look at the drive waveforms (on page 2, with a drive
> pseudo-circuit). Basically, there should be 5 wires coming from the
> motor. The drive system they analyze has 10 steps per electrical
> cycle. Each phase can be connected to +V or -V at any time - note that
> "disconnected" isn't an option. Essentially, the 5 phases (A-E) are
> each connectd to +V for 5 periods, then to -V for 5 periods. The phases
> are staggered by two periods each.
>
> So, given 10 "steps" per electrical cycle, the phases are energized like
> this:
> Phase A B C D E
> period
> 1 + - - + +
> 2 + - - - +
> 3 + + - - +
> 4 + + - - -
> 5 + + + - -
> 6 - + + - -
> 7 - + + + -
> 8 - - + + -
> 9 - - + + +
> 10 - - - + +
>
> I think the 10-step mode is actually a "half-stepping" mode, so this
> might give 1000 steps per revolution instead of 500.

Looks like you could do this real simply in hardware with a ring counter...

> If someone sends me a circuit diagram, I can lay out a board and write
> some PIC or AVR code for it. (I'm an electrical engineer, but I've
> always been more comfortable with the digital design/programming end,
> not the power end of things :)

Take a look here:

http://caladan.nanosoft.ca/index.php

under "circuit corner", one of the entries under there shows some H-bridge
stuff. You'll need half an H-bridge for each of the five wires, parts
depending on how much voltage and current have to be handled.