Re: [CAD_CAM_EDM_DRO] Re: Motor vs. Servo

jtfrimenko wrote:

> Since the motor is connected ridgedly, through the lead screw, to the
> work, how does the inertia of the work affect motor inertia?

There are a number of fine points, mostly having to do with the torsional
springiness of the leadscrew, belt stretch, and things of that sort.
So, it is not completely true that the inertias just add together
(accounting
for one being linear, the other rotational, and the gearing, belts,
leadscrews
etc. in between). But, for practical, relatively low speed servo motion
systems like a Bridgeport mill up to 200 IPM or so, these fine points are
really insignificant in most systems.

> A
> bridgeport table must weight several hundred pounds and is sliding on
> gibs. Driving a high inertial load with a low inertia motor seems to
> be inefficient. If we were doing pick-and-place with a mill table,
> then a fast motor would be necessary.

Yes, this is very true. And, in fact, commercial CNC systems do NOT
use servo motors of particularly low inertia. Perhaps the inertia of these
motors is 1/2 of what might be the most economical, "square" motors,
and this increases cost by a relatively small amount - maybe 25%
at the most. (Motors with approximately "square" armatures, ie.
length = diameter are a little cheaper, as the minimum amount of field
magnet strength is needed. Making a longer armature requires more
field magnet material.)

Note the SEM motors used on Bridgeport's own CNC machines (BOSS 7
and Ez-Trak) are quite long and very small diameter. There are two
reasons for this. One is they use a large belt reduction ratio, something
like 6:1, I think. So, the faster the motor spins, the more rotational
inertia
affects system performance. But, if you are going to use extremely
exotic field magnets (such as Samarium-Cobalt rare earth magnets)
then this may be cheaper than a larger motor with less belt reduction.
Secondly, these long, skinny, sealed motors may develop more internal
cooling of the armature than a shorter, fatter one.

> The next question, given a typical feed rate and screw pitch, what is
> the minimum motor characteristics required. High speed slew is not
> necessary for a home machine.

This gets real complicated. It depends on the screw pitch, belt reduction,
table weight, acceleration required, drag on the ways, spindle horsepower
and cutter type and size, etc. After friction losses, the force applied to
the
table is needed to accelerate, and to resist cutter reaction forces.
What is the worst case cutter reaction force for the work you do? Do you
have any idea at all? You can get a very rough idea from the force required

to turn the handles of a manual mill.

Anyway, after some rough calculating, I figured the worst case side force
on a Bridgeport machine with R-8 spindle was probably around 250 Lbs.
This would be using a very large cutter at low speed, and plowing through
steel at full width. I multiplied by 4 to come up with what force the servo
needed
to supply to overcome friction and acceleration demands.

I then back calculated required motor torque from the leadscrew and belt
drive, and saw that it really wasn't very hard to do. I am using motors
rated at 1/8 Hp continuous, and I go to about 4 x that torque for peak.
It works well. I have these calculations at http://206.19.206.56/motor.htm

Jon