Re: [CAD_CAM_EDM_DRO] stepper cuts better circle than servo?

pcfw wrote:

>Hi,
>
>A plus point for stepper? Read somewhere (can't find it now) that a
>stepper system cuts a more perfect circle than a servo system. This,
>it seems, is because of the constant hunting by the servo system.
>
>Waiting for response from the servo champions.
>
>

Poorly designed or installed servos do hunt, producing noise that will
slowly
drive you nuts. Well designed and installed servo systems produce NO
detectable hunting, without sensitive instruments. My X and Y servos
don't hunt at all, they settle to within one encoder count (.00005") and
just sit there, every once in a while bumping over one count. My Z
axis, due to the weight of the quill, does hunt, at a rate of one
encoder count
avery 4 seconds, roughly. That is, the weight causes it to slowly creep
down
.000025" (resolution is twice as high on my Z due to a 10 TPI ballscrew
as opposed to 5 tpi on the X and Y) until it passes one encoder
transition, then
a command is sent to bump it up. If you can be troubled by a deviation
of 25 Micro-Inches, you aren't making parts on a Bridgeport.
I'd like to see you make a 25 uIn move on a typical stepper system.

Similar servo systems are used to point telescopes to milli-arc-second
accuracy, and do other incredibly fine positioning tasks.

So, a good stepper system is admittedly better than the worst servos out
there, but that does NOT prove the point.

Jon