On Tuesday 06 February 2018 10:20:49 [email protected] wrote:
> As I replied to Andy, spindle-at-speed is just coming from
> vfd-at-speed in custom.hal. net gs2-at-speed motion.spindle-at-speed
> <= spindle-vfd.at <http://spindle-vfd.at/>-speed
>
Nothing at the link. And my vfd on the sheldon has an output, but no clue
what its telling me. If the motor is working within its load, the speed
should be within 5% of the indicated frequency, factored by the number
of motor poles. IOW, 60 hz should get you at least 1750 revs from a 4
pole motor. 120 hz = 3550 etc. Mine, without a lot of low speed boost,
can run at 9 hz for long enough to get the job done w/o serious motor
heating. Conversely, while the motor will run at 240 hz on the table,
phase slippage from low winding current won't let me go above 120 hz,
when pulling the lathe, as spindle bearings are bronze and drag too
much, using up all its limited torque. Warms them up too.
But a near or wcomp, near preferred, really should be setup to tell
motion the spindle is at speed, independent of any signals from the vfd.
10% tolerance is generally fine. I have an led rigged in postgui to tally
that. It only controls whether or not the cycle starts, but has no
effect once its underway, and with a tap bigger than say 5mm, its not
uncommon to see the led turn red once the loading of the tap hits. The
near is much easier to calibrate if the two signals, input to the
pid.command, and return from the encoder are at the same scale factor.
So scale them either to rps, or to rpm's.
You have several outputs from motion:
==========
motion.spindle-speed-out OUT FLOAT
Desired spindle speed in rotations per minute
(assuming its a signed value, the man page doesn't state)
motion.spindle-speed-out-abs OUT FLOAT
Desired spindle speed in rotations per minute, always
positive regardless of spindle direction.
motion.spindle-speed-out-rps OUT float
Desired spindle speed in rotations per second
motion.spindle-speed-out-rps-abs OUT float
Desired spindle speed in rotations per second, always
positive regardless of spindle direction.
=========
And generally the encoder will be scaled to rps, so the above rps outputs
should about match at the near inputs. Because I am using the OEM 1hp at
90 volts motor on my mill and a similar escapee from a treadmill on TLM,
both are being driven by Pico's PWM-SERVO amps, so I've some additional
hal stuff to keep the PWM from exceeding 98% duty. Hit 100% for more
than a few milliseconds and the Pico goes into a self protection
shutdown.
Separately, if you run the encoder output thru an abs before hitting the
gui's tach indicator, then it runs showing a fwd reading in either
direction. If using the tach dial meter, I've not used the sliding bar
version enough to recall if it runs to the right in reverse, but the
number displayed s/b signed.
> I am not using wcomp or near component as suggested in the manual:
> http://linuxcnc.org/docs/html/examples/spindle.html#_spindle_at_speed
>
Thats not ideal Tom. Do something like shown in that link above.
> -Tom
>
> > On Feb 4, 2018, at 9:20 PM, Gene Heskett <[email protected]>
wrote:
> >>> On Sunday 04 February 2018 13:52:33 andy pugh wrote:
> >>>> On 30 January 2018 at 22:42, <[email protected]> wrote:
> >>>>> We read that G33 will wait for spindle-at-speed before looking
> >>>>> for the index pulse.
True.
> >>>>> But our spindle-at-speed signal seems to
> >>>>> be high during the full cycle once the spindle speeds up the
> >>>>> first time.
This shouldn't be. The starting position in radii s/b the right end of
the white line, under the g76 section and the spindle-at-speed needs to
be false the instant the move to that "parking while waiting" position
starts, and which should slow the spindle, and remains false until the
css drive has actually reached the indicated speed. It might even be
wise to cut a bottom of stroke clearance groove if the screw going in
the hole could touch the exit ramp shown in that diagram. But I've never
considered that despite my abuse of G73 to cut tapered threads in the
past.
The biggest argument against using css while thread cutting, is that the
change in the instant cuts diameter while its waiting at the right end
of the white line for at speed and index, also changes the spindle speed
on a per stroke basis AND that changes the movements phase lengthwise so
you will be cutting a wider thread with the trailing edge of the tooth.
If you really had the machine characterized, it might be possible to
change the Q argument a few degrees plus to compensate. Worth the
effort in time saved? Probably not unless you have several thousands of
that part to make.
--
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
