Re: Digest Number 68
Posted by
Andrew Werby
on 1999-06-28 02:57:39 UTC
"Ian W. Wright" <Ian@...>
Subject: Re: Stepper controllers
Hi Andrew,
Sorry I have taken so long to get back to you but I have been away for a
couple of weeks.
I am not trying to do any more than persuade the steppers to run
reliably. Using the Pentium I can get about 150 rpm out of the motors
without losing steps but, using the 486, I can only get reliable
stepping at about 50 rpm and then only on selected motors. The 486 I was
using only had a 486/66SX chip in and so I wonder if the problem was due
to the lack of a maths co-processor (as in the DX chip)? I was running
the computer in Windows '95 DOS mode from the hard disk and using 16Mb
of RAM.
[I don't think this has much to do with floating-point math- as far as I
know (and we're getting to the outer edge of what I do know about this) the
task is simply to spit out precalculated numbers at the correct time. It
does sound like your problem is related to processor speed, though.]
I did use the Timeset program to set the initial parameters for MaxNC
and fiddled with the corresponding settings in the Stepster program.
Using the Pentium the motors just fly but I was hoping to be able to use
the 486 so as not to tie up the Pentium which could be better employed
elsewhere.
Ian
[Well, the MaxNC system works on a 486, but it doesn't go very fast. I
wasn't aware that it would go much faster with a Pentium. Perhaps the raw
speed of the motors isn't what's important, but the speed at which it can
reliably count steps, and the faster processor is better at this. If you're
using Stepster instead of MaxNC's software, you might have to run a
time-setting program specific to that application.]
Andrew
Andrew Werby wrote:
Sculpture, Jewelry, and Other Art Stuff
http://unitedartworks.com
Subject: Re: Stepper controllers
Hi Andrew,
Sorry I have taken so long to get back to you but I have been away for a
couple of weeks.
I am not trying to do any more than persuade the steppers to run
reliably. Using the Pentium I can get about 150 rpm out of the motors
without losing steps but, using the 486, I can only get reliable
stepping at about 50 rpm and then only on selected motors. The 486 I was
using only had a 486/66SX chip in and so I wonder if the problem was due
to the lack of a maths co-processor (as in the DX chip)? I was running
the computer in Windows '95 DOS mode from the hard disk and using 16Mb
of RAM.
[I don't think this has much to do with floating-point math- as far as I
know (and we're getting to the outer edge of what I do know about this) the
task is simply to spit out precalculated numbers at the correct time. It
does sound like your problem is related to processor speed, though.]
I did use the Timeset program to set the initial parameters for MaxNC
and fiddled with the corresponding settings in the Stepster program.
Using the Pentium the motors just fly but I was hoping to be able to use
the 486 so as not to tie up the Pentium which could be better employed
elsewhere.
Ian
[Well, the MaxNC system works on a 486, but it doesn't go very fast. I
wasn't aware that it would go much faster with a Pentium. Perhaps the raw
speed of the motors isn't what's important, but the speed at which it can
reliably count steps, and the faster processor is better at this. If you're
using Stepster instead of MaxNC's software, you might have to run a
time-setting program specific to that application.]
Andrew
Andrew Werby wrote:
> [How fast are you trying to run your steppers? I've run MaxNC's softwareAndrew Werby - United Artworks
> successfully on one of their machines, using an antique 386, and it worked
> fine- no lost steps. Perhaps if you get into higher speed ranges you need a
> faster CPU. But there might be a problem coordinating the speed of the
> motors with the speed of the computer. MaxNC has a program called "Timeset"
> that you need to run when initializing a machine- did you do this?]
>
> Andrew Werby
Sculpture, Jewelry, and Other Art Stuff
http://unitedartworks.com
Discussion Thread
Andrew Werby
1999-06-28 02:57:39 UTC
Re: Digest Number 68
Dan Mauch
1999-06-28 13:27:47 UTC
Re: Digest Number 68