Re: These servo motors worth a darn?

Jon Anderson wrote:

> From: Jon Anderson <janders@...>
>
> Got a question or two. I have a 3 axis cartesian coordinate robot that I
> am reconfiguring as a gantry robot. it's got top notch hardware like NSK
> linear ways and preloaded ball screws. Came with some Tamagawa brush
> servo motors but no amps. The robot was an Apollo, appears to have been
> made in the UK. All I have for specs on the motors is wattage. Two are
> 100 watt with 500ct encoders, one is a 60 watt, and there is a large 200
> watt with integral brake. I can provide model numbers if anyone has ref.
> material.
> I'd like to know if anyone can guess what sort of power output I might
> see out of the 100 watt motors, both peak and average.

Ugh. Not a whole tol. I'm assuming these must be pretty small robots.
Generally, these are continuous ratings, so the motor can deliver 100 Watts
power at the shaft all day, which is not a lot. Typically, all servo motors
can deliver 2 x continuous for some modest percentage of the duty cycle,
perhaps 25 % or so, if the rest of the duty cycle is a bit below 100%.
But, some specific motors for servo applications have extremely conservative
continuous ratings, with extremely large peak power capability.
I was looking at the motors used on Bridgeport's recent 'EZ-Trak'
factory-converted manual milling machine, and the small motors have
something phenomenal like 100V 60 Amps on the nameplate.
Obviously, this motor has WAY more than 2 X peak power.

>
> With 10mm pitch leadscrews, if I use 1000 line quadrature encoders, and
> if I understand what I've read here, I ought to get a potential .0001
> resolution. Is that correct? Due to design, I'd really like to belt
> drive the screws. Can I put the encoders on the end of the screws or
> would that introduce some sort of servo error?

Well, let's work it through. 1000 cycle encoder gives 4000 counts/turn,
assuming direct drive, 10mm /4000 counts = .0025 mm / count.
Putting the encoder directly on the end of the screw with a high
quality helical coupler is the best way. Then, let the motor drive
the screw with a toothed belt.

> Will be back with more questions soon as the Glen-Tek servo amps I got
> on Ebay arrive... (like, will they drive the Tamagawa motors...)

Given some reasonable match of voltage and current between
the amps and the motors, then any brush motor servo amp should
work with the motors. You may need to add tachs, as many older
DC servo amps need a DC tach for velocity feedback. The only
other quirk is some cheap servo amps don't have enough range
of compensation, and so the motor inductance must be within some
range of values. If the motor inductance is too low, you can always
add a series inductor. If it is too high, you have to increase the
loop compensation in the amp.

Jon