Re: Servo torque requirements

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "siado01" <siadohauc@h...>
wrote:

>
> Greetings all,
>
> Let me start with the disclaimer that I am a complete newbie to the

CNC arena, so if my

> questions are nonsensical, bear with me.
>
> I am planning a Benchtop CNC Mill using THK linear guides and some

Parker SM232 servos

> I picked up for a great deal. I am shopping around for ballscrews

and I guess I'm not really

> sure what I'm after.
>
> The Parkers are rated for 161 oz-in continuous stall torque. I have

seen several larger lead

> ballscrews, along the lines of 16mm/turn or so. I plan on using

direct coupling for the

> servos, so reduction hasn't really been considered at this point.

Would this ballscrew be a

> bit much for standard milling work with these servos?
>
> I know discussing resolution is pointless, since the servos with

1000 line encoders would

> be far more precise than the rest of the machine could be. I don't

think the lead would be

> limiting, but then I am a newb.
>
> Thanks for any advice,
>
> Jim

The first issue I see is that the direct coupling is going to cause
problems. A ballscrew with a 16mm pitch is really course and will
give you amazing rapids but servos perform best when run at moderate
to higher RPM (as opposed to steppers that loose torque with RPM).
Lets say your Servo's are 36 volt 3500 rpm units (for easy math) you
would like to operate them in the 750 to 2000 rpm section of their
range. Extremely low speed operation will display cogging. If you
direct couple the motors at 750 RPM you will be moving at 200
mm/sec. A 3 or 4:1 reduction ratio would bring those speeds down to
about 50mm/sec (approx 120IPM) and increase your available torque.
If you went with a 4mm pitch ballscrew then once again you gain force
and slow things down by 4X. You want a machine that has good smooth
operation at normal cut speeds and decent rapids. Even with all of
the reductions you still would have 58 mm/sec rapids and LOTS more
force at the cut.

One other issue will be the maximum speeds you can expect given the
1000 line encoders. That's 4000 pulses in quadrature so it will take
4000 pulses to turn the motor one rev. That means with typical PC
based step and direction to servo designs you can expect about 10 RPS
(600 rpm max) out of your motors. If you go with a pulse multiplier
like the Gecko 340 then you can up that by a factor of 5 or 10 so you
could operate the motors to their full RPM if needed.

Torque ratings for servos are much more meaningfull than for steppers
but stall torque is not much to work with since you really don't want
to operate a servo at stall.

What size machine are you looking to use?