Re: [CAD_CAM_EDM_DRO] CNC Servo motor gear ratio: overheating. EMC step rate. Table speed

Kim Lux wrote:

>Hi people.
>
>We are retrofitting a small mill. The force required to move the table
>is low, ie 25 pounds moves it with the screw removed.
>
>We are using EMC as the controller. I've read it has a maximum step
>generation rate of 5 KHz.
>

No. It is closer to 25 KHz with a reasonable Pentium CPU. 333 MHz Pentium
II should be fine at this rate. If you need more steps/sec, you can try
a faster
CPU, or my Universal Stepper Controller, hardware that generates the step
pulses for the CPU.

>
>We've purchased 620 oz in servo motors from Dan Mauch (Camtronics) with
>500 step per rev encoders.
>
>We are planning to direct drive a 4 or 5 TPI ball screw.
>
>My calculations show that 600 oz direct driving a 5 TPI ball screw will
>generate a force of 1059 lbs.
>
>The reason for wanting to direct drive the screw is speed. EMC
>supposedly has a maximum step generation frequency of 5 KHz. 5 KHz/ 500
>steps per rev = 10 revs per second or 600 RPM. 600 RPM x 0.2 = 120
>inches per minute, which is slower than we want. We were thinking of
>using 4 TPI ball screw to get the speed up to 150 IPM.
>
>Question: will the motors suffer from heating problems ? At a feed rate
>of 5 inches per minute, the motors will be turning 25 RPM.
>

It is hard to say. Is there a continuous torque or continuous current spec
on these motors? If so, you can make an attempt to see whether friction
and cutting forces (if any) will exceed this spec. Always throw in a fudge
factor for unanticipated drag. Anyway, if 25 Lbs linear force is required,
add a generous 20% fudge for the ballscrew and you get 25 * 1.2 = 30
30/1059 = 0.028, or 2.8% of peak torque. (I'm assuming the 620 Oz-In
is the peak torque, otherwise, this is a HUGE motor). it seems pretty
safe to assume that running at 2.8% of peak torque, it is not going to
overheat.

>
>
>

With the Universal stepper controller, I've run some 200 Oz-In size 34
steppers
at 4800 full steps/second (48,000 steps/sec into a Gecko201A). That ran a
mini-mill to 90 IPM with 16 TPI screws. 48000 was the limit for these
motors, the USC can go to 300,000 steps/sec.

Are those 500 cycle/rev, or 125 cycle/rev encoders? (You get 4 counts from
every black/white cycle on the encoder disc.) If they are 500
Cycle/rev, you'll
actually get 2000 counts/rev.

(you can get more info on my USC at
http://jelinux.pico-systems.com/univstep.html )

Jon