Re: [CAD_CAM_EDM_DRO] RE: Stepper questions

Alan Marconett wrote:

>HI Roy,
>
>Ah! Thanks! That's the point I was overlooking. Left to it's own, the
>winding current would take much longer to decay then when DRIVING it in the
>opposite direction.
>
>

Yup, this is sort of the key. Shorting the winding would cause the current
to decay slowly. The unipolar drive switches the current to the opposite
end of the center-tapped winding, and that action attempts to reverse
the flux.
So, it is actually somewhat similar to the way a bipolar drive applies the
reverse voltage to a single winding. But, there is leakage inductance
that is
not coupled fully between the two halves of the phase coil of a unipolar
setup. Mariss may know more about this that I do, but my feeling is this
leakage inductance may be substantial. If you didn't have the freewheel
diode
to contend with, and some very small unipolar drive/motor combos do NOT,
then you actually are doing something quite similar to a bipolar drive.
But, when the motors get up above size 23, for certain, the inductances
start to develop voltage spikes that are hard to control. Then, you have to
put in some kind of arrangement to catch that inductive energy. But, you
want the diode circulating current straight back to the power supply
(really the winding center tap) when the drive is in the chopping mode,
but allow the end of the winding to fly up to twice the supply voltage when
the drive steps. All sorts of schemes have been devised to do this, but
they all
end up being more complex than a bipolar drive.

One other scheme is to use no freewheel diodes, or have them with
substantial
resistance or zener diodes in series, but never chop at the 4 main
transistors.
Instead, you add 2 more transistors where the massive power resistors would
go in an L/R drive, and chop those transistors instead. Freewheel
diodes from
ground to the winding center taps would be provided to circulate the winding
current. This ends up with a 6 transistor drive.

Jon