CAD CAM EDM DRO - Yahoo Group Archive

Re: [CAD_CAM_EDM_DRO] Re: Stepper vs. Servo

Posted by Jon Elson
on 2001-06-25 11:29:58 UTC
Tom Eldredge wrote:

> Mariss,
>
> Thank you for your good post on stepper vs servo. I will add a comment in
> the form of a question, since I my experience is so limited. In my two
> years of working on a controller using a stepper driven BP machine as my
> test bed, I learned that acceleration and deceleration are real touchy with
> steppers. I have found that servos are much more forgiving in this area.
> Is this your observation? Also, when making moves with a stepper wherein
> the x and y axi are very close in frequency, but not exactly, such as when
> one makes a move at 43 degrees, DEPENDING ON HOW the pulses are generated,
> the steppers may have more difficulty in following an irregular pulse rate
> if there is one. This would seem like it would be an issue with controls
> which are windows based.

Yes, it is a real problem. The stap rate must be tightly controlled, with no
discontinuities. You can't do it with a plain windows program, but may be
able to do it with a VXD or DLL that can get control of the CPU when it needs
it. The entire motion control needs to be running under this mode, so that
changes in velocity going from one "waypoint" to the next can be performed
when Windows is busy with something else.

A servo system would not lose position reference if not updated for a while,
but it could still go sailing (oh, so smoothly) right on past a waypoint, and
put a nasty gouge in the work! So, it is really not all that forgiving,
either.

This is why DeskNCrt is stuck in DOS, because that's the only way he can be
sure he has the entire CPU to himself.

If external hardware can command the attention of the CPU (or reprogramming
of the motherboard timer and appropriate software), then you can have
the same effect as a real time OS, with some constraints. You have to be
sure the interrupt controlling motion is the highest priority, is not maskable,

that the motion control routines are not paged out of memory, and similar
things to keep the OS from preventing the motion routines from doing their
job exactly on schedule.

Jon

Discussion Thread

dougwalker13@y... 2001-06-22 21:27:42 UTC Stepper vs. Servo mariss92705@y... 2001-06-23 00:06:13 UTC Re: Stepper vs. Servo M. SHABBIR MOGHUL 2001-06-23 21:08:11 UTC Re: [CAD_CAM_EDM_DRO] Stepper vs. Servo Tom Eldredge 2001-06-25 04:31:03 UTC Re: [CAD_CAM_EDM_DRO] Re: Stepper vs. Servo Jon Elson 2001-06-25 11:29:58 UTC Re: [CAD_CAM_EDM_DRO] Re: Stepper vs. Servo M. SHABBIR MOGHUL 2001-06-25 20:56:22 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot NCS 2001-06-25 22:04:43 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot Jon Elson 2001-06-25 22:58:09 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot Tom Murray 2001-06-26 03:17:34 UTC Re: Stepper vs. Servo ballendo@y... 2001-06-26 05:40:56 UTC Re: Stepper vs. Servo ballendo@y... 2001-06-26 06:27:53 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot ballendo@y... 2001-06-26 06:42:48 UTC Re: Stepper vs. Servo mariss92705@y... 2001-06-26 07:43:00 UTC Re: Stepper vs. Servo Jon Elson 2001-06-26 11:09:45 UTC Re: [CAD_CAM_EDM_DRO] Re: Stepper vs. Servo Jon Elson 2001-06-26 11:11:41 UTC Re: [CAD_CAM_EDM_DRO] Re: Stepper vs. Servo JanRwl@A... 2001-06-26 11:12:56 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot Tom Murray 2001-06-26 19:28:45 UTC Re: Stepper vs. Servo M. SHABBIR MOGHUL 2001-06-27 20:52:45 UTC Re: [CAD_CAM_EDM_DRO]stepper is very hot M. SHABBIR MOGHUL 2001-06-27 20:52:46 UTC required current disabling circuit for stepper