Re: [CAD_CAM_EDM_DRO] upgrade strategy

In a message dated 19-Aug-02 00:45:03 Central Daylight Time,
keongsan@... writes:

> 2. less than satisfactory pulse train. (deskNC for windows)
>

San: There is a point perhaps related to this I HOPE Mariss will address
here, and that is the "minimum ON-time" of the input STEP pulse to the G210.
I had an early ASM-86 "machine code patch" in my BASIC home-brew CNC
programming on a 10-year old Tandy laptop, resulting in ca. 3.75 µS input
pulses. This basically "worked", but I noticed audible "glitching" in ONE
axis (they are electrically identical) in this plotter I just finished, and I
adjusted the "order of march" in my machine-code patch, getting the ON-time
width up there into the tens of µS. This smoothed things out. The "white
paper" on the G210 specifies 4 µS min. ON-time.

HOWEVER! These G210 bipolar-chopper drives powering KML-061 (Superior
Electric) motors, 5.09 V./1.2? A. will return reliably to the same starting
point ONLY if the STEPPING RATE is "just so". I have found MUCH wider
tolerance of stepping-rates with old "6-wire unipolar" designs!

I must investigate the G201. Are they 200 step/turn drives, or 2000? I have
forgotten, now, and am so busy with "hardware cobbling" that I can't take
time to re-read every "white paper" in the file on the topic! The
"microstepping" seems to make things nicer, but the result SEEMS to be
step-skipping even at only 300 s/sec.

Jan Rowland


[Non-text portions of this message have been removed]