Re: [CAD_CAM_EDM_DRO] Correct or Max voltage for Bridgeport Nema 42? + more...

> Not knowing any better, at this point I plan to use Gecko 201's
> because It's the most common drive I've seen discussed for these
> conversions. The Gecko rep says using 80v is out because they max at
> that voltage and with the power company fluctuations, I'd probably
> toast it. So that leaves me with 60 or 70 v to work with.
> You are also supposed to know the max rated voltage for your
> motors...which I have no clue.

My guess is you would probably be fine with 60 volts, but if you know
the rated current for the motors you could figure out the 'steady
state' voltage with the aid of a high accuracy ohm meter, or a hefty
bench supply capable of current regulation. I think the concern is
simply that if the supply voltage is more than something like 10 times
the designed voltage, then the drive can't make the chopping pulses
short enough to avoid killing the magnets with overcurrent - but I'm
by no means an expert.

If you have 6 or 8 wire motors, you could connect them in series,
using both halves of each coil, which would double the voltage
requirement, increasing the safety factor there. Then if the drive
seems to top out at a dissapointing speed, trying half coil connection
for lower resistance,inductance->voltage might be warranted.

Depending on the software you use, you may want the G210 drives with
the step pulse multiplier in order to keep the rapid rates high (the
G201 effectively divides the resolution through its microstepping,
meaning a very high step rate is needed during rapids) though you can
buy the multipliers later if you find you need them. And depending on
what a step translates to in motion of your machine, you may want the
microstepping to keep the resolution high (though microsteps are not
perfectly even).

> The reason I ask is because I can see some limitations in what you
> can do with only step and direction pulses on a single LPT. You'd
> think that some developer would at least consider accessing two LPT
> ports for more options. I've seen relay boards that could be quite
> useful, but would require an extra LPT, not to mention a few DAC
> boards.

Some of the packages do support multiple LPT ports or other types of
I/O, including DAC's for axis and spindle servos. However being able
to address these I/O capabilities is only usefull if the program
architecture is set up to control them in useful ways. Those who want
to experiment with add-ons basically need to work with an open-source
control that is designed to be modified, such as EMC. There is a
learning curve, with that package, but it has been made to talk to all
sorts of I/O devices, controlling both steppers, step-input-servos,
and normal servos. Alas is does not yet support rigid threading.

The geckos are a good investement, and about the best you can do for
driving the existing stepper motors, using any one of the pc-based
control packages. You can easily switch software, and between
parallel port or other types of I/O cards while keeping the same
drivers and motors. The primary alternative would be to go to a servo
system - but then you would have to replace the motors.

Chris Stratton

-------------------------------------------------
Christopher C. Stratton
Engineer, Instrument Maker, and Horn Player
ccs@... 617 628 1062
http://web.mit.edu/~stratton/www/brassbuild.html