On 9/15/24 10:28, Sam Sokolik wrote:
So... Without ripping out the current spindle motor and starting over..
It works surprisingly well - the 0-950 acts like a normal dc motor (field
voltage stays constant) The field weakening is a bit more mushy. Dropping
the field voltage from 115v to 28V runs the rpm up from 950 to 2500 a bit
slower..
Currently I have a 2 lincurv components that take RPM and spit out a
voltage to each drive (field and armature) In front of that is a limit2
which limits the slew to 400rpm/sec. If I put a pid loop around that - it
works ok - and maybe that is as good as it gets.
Thoughts?
Have you tried two loops, one to control the whole loop and one inverted
to control the field? Seems to me that might give better midrange
torque response. Rig it for full field up to 950, at which point max
power is available from the first pid, then fold the field down for rev
inputs above 950.
I can't mentally visualize how the main pid would respond to the windup
it might do at the higher speeds. I have nothing like it available to
play with, my DC motors are all 1hp, 90 volts at 9.7 amps. For threading
I bang them quite a bit harder, up to 18 amps @125 volts. A decade+ old,
still running on OEM brushes.
As a first try, given the track record of the new atpid being quite
destructive here, burning up the motor on my BS-1 while attempting an
autotune in about 30 seconds. It was not a huge success, way to slow. A
different setup solved that perfectly for what I wanted to do.
I don't think I'd pid the field control, but just use a lincurve to turn
it down when the requested revs go above 950. Speed limit the change
rate of the input command revs with a limit3.
High end control is going to be mushy and I might want to tally the
motor temp as a motor safety feature to make sure its not cooking before
the job is completed. You could make that out of an si diode clamped to
the motor (or better yet mounted to measure its air's output temp) and
an op-amp. Run a milliamp thru the si diode to measure its fwd drop. Hot
motor=less drop. Done right, it s/b self regulating but will of course
lose some torque when hot.
Good luck Sam.
On Fri, Sep 13, 2024 at 6:35 PM Sam Sokolik <samco...@gmail.com> wrote:
You are right - for just spindle control.. But the 5 year plan is to also
cnc the monarch. (and linuxcnc has all the stuff working and known -
encoder, limit2, pid, lincurve, pwmgen and so on..) I would rather not
re-invent the wheel. (and I am far from comfortable coding on the arduino)
sam
On Fri, Sep 13, 2024 at 6:11 PM andy pugh <bodge...@gmail.com> wrote:
On Fri, 13 Sept 2024 at 23:25, Sam Sokolik <samco...@gmail.com> wrote:
Enter Hal, 2 amc drives and a rpi5... (and a few other things)
Much as I am a fan of LinuxCNC, that's probably a job for an Arduino,
really :-)
--
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912
Cheers, Gene Heskett, CET.
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
