Re: [Emc-users] Pncconf VFD Spindle scale bug and fix

[Jason Burton]

Hi Gene,

To your first question, I do not know that this bug is hitting you.  What
it looked like for me was this:  adjustment of spindle scale parameters
(...7i77.0.1.analogout5-scalemax, ...analogout5-minlim, &
...analogout5-maxlim) only worked within a tiny range.

For scale from N=1 through 10 the VFD output was linear for commanded
spindle speeds from s1 though sN as you would expect.

An example with scalemax=10, minlim=0, & maxlim=10:
 1) Issuing m3 s10 gives full speed spindle. As do m3 s11, s1000, s1, &
s10.
 2) Issuing m3 s5 gives 1/2 of full speed.
 3) Issuing m3 s1 gives 1/10 of full speed.
  -All correct, just not scaled to actual RPM.

Here is what was broken:
 -For anything above scalemax = 10 (and appropriate minlim/maxlim) the
spindle would turn on, but at ~60 rpm.  For all commanded speeds.

There are two lines that prohibited scalemax/minlim/maxlim from doing their
job.  They are corrected easily.  See the bug report if it looks like it
applies.

With these two lines fixed, the scale parameters do in fact scale as they
should. Commanding m3 s2000 gives 2000 rpm.

I am no guru but for symptoms you describe, I'd be tempted to examine first
P and then the feedforward terms.  Perhaps drill holes in scrap to tune the
spindle under load. I found John Thornton's tuning tutorial helpful.

To your second question, I have only one gear ratio on mine and have not
needed to examine gearchanges.  Sorry mate.

Sincerely,
Jason Burton

Good luck with the rebuild.




On Fri, Aug 28, 2015 at 9:56 PM, Gene Heskett  wrote:

> On Friday 28 August 2015 22:02:51 Jason Burton wrote:
>
> > Just an heads up to the list, and for future Google searchers.
> >
> > First, let me thank all you developers for building pncconf in the
> > first place. Great tool for quick setup and learning.
> >
> > I have been working on getting my VFD mill spindle to work with proper
> > scaling for some time now.
> >
> > Turns out Pncconf was generating a couple lines incorrectly.  Easy fix
> > only if you know what is wrong.  Thanks for the patient help Peter
> > Wallace!
> >
> > Hardware: big vertical machining center + Mesa 5i25/7i77 + 0-10 VDC
> > controlled VFD spindle.
> >
> > Software: all recent linuxcnc/pncconf versions dating from current
> > live cd Debian with master branch to at least as far back as the
> > Ubuntu 10 live cd.
> >
> > Both the problem and the fix are described here:
> >
> > https://sourceforge.net/p/emc/bugs/433/
> >
> Interesting. pncconf generated a spindle scale factor of 1.0 here, and
> its relatively non-linear but does cover the range of speeds I can get
> with my hardware, the following.
>
> Grizzly GO704 mill, OEM 1 hp motor, PM DC brushed.
>
> Locally built 115 volt dc power supply good for at least 1 KW.
>
> One of the Pico Systems pwm servo amplifiers,
> currently fed at 40 KHz by the pwmgen in a Mesa 5i25 by bypassing the
> optos for that line on my BoB.  Still 1 more opto though, one to
> translate between the BoB's 5 volt swing and the 12 volt swing the pwm
> servo expects.
>
> Overall its working, just not dead accurate, an s2100 actually gets me
> about 2700, but an s80 gets me about 40 revs.  Huge amounts of Pgain
> will stiffen it some but I get the impression I could make Pgain 1500,
> up from the current 150 and it would still be sagging 40% while driving
> a 4-40 tap in 1/2 thick alu using the G33.1 rigid tapping cycle.  This
> with the heads backgear in high range.
>
> So, how might I apply this to my situation to both stiffen the speed
> regulation and tighten up the commanded speed to actual speed obtained?
>
> Also related, do we have a canned halui (or gladevcp) button to change
> the gear scale according to the heads gearshift?
>
> I have 3 machines with this hi-lo gear choice, but only 2 have encoders
> for speed feedback, this new to me mill being one of them.
>
> > Best,
> > Jason Burton
> > (LatheBuilder)
>
> Currently rebuilding mine again. I need a 2x2x5" block of steel for the
> next step, but I expect some gasolene will locate that.
>
> 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 
>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Pncconf VFD Spindle scale bug and fix

[Gene Heskett]

On Friday 28 August 2015 22:02:51 Jason Burton wrote:

> Just an heads up to the list, and for future Google searchers.
>
> First, let me thank all you developers for building pncconf in the
> first place. Great tool for quick setup and learning.
>
> I have been working on getting my VFD mill spindle to work with proper
> scaling for some time now.
>
> Turns out Pncconf was generating a couple lines incorrectly.  Easy fix
> only if you know what is wrong.  Thanks for the patient help Peter
> Wallace!
>
> Hardware: big vertical machining center + Mesa 5i25/7i77 + 0-10 VDC
> controlled VFD spindle.
>
> Software: all recent linuxcnc/pncconf versions dating from current
> live cd Debian with master branch to at least as far back as the
> Ubuntu 10 live cd.
>
> Both the problem and the fix are described here:
>
> https://sourceforge.net/p/emc/bugs/433/
>
Interesting. pncconf generated a spindle scale factor of 1.0 here, and 
its relatively non-linear but does cover the range of speeds I can get 
with my hardware, the following.

Grizzly GO704 mill, OEM 1 hp motor, PM DC brushed.

Locally built 115 volt dc power supply good for at least 1 KW.

One of the Pico Systems pwm servo amplifiers,
currently fed at 40 KHz by the pwmgen in a Mesa 5i25 by bypassing the 
optos for that line on my BoB.  Still 1 more opto though, one to 
translate between the BoB's 5 volt swing and the 12 volt swing the pwm 
servo expects.

Overall its working, just not dead accurate, an s2100 actually gets me 
about 2700, but an s80 gets me about 40 revs.  Huge amounts of Pgain 
will stiffen it some but I get the impression I could make Pgain 1500, 
up from the current 150 and it would still be sagging 40% while driving 
a 4-40 tap in 1/2 thick alu using the G33.1 rigid tapping cycle.  This 
with the heads backgear in high range.

So, how might I apply this to my situation to both stiffen the speed 
regulation and tighten up the commanded speed to actual speed obtained?

Also related, do we have a canned halui (or gladevcp) button to change 
the gear scale according to the heads gearshift?

I have 3 machines with this hi-lo gear choice, but only 2 have encoders 
for speed feedback, this new to me mill being one of them.

> Best,
> Jason Burton
> (LatheBuilder)

Currently rebuilding mine again. I need a 2x2x5" block of steel for the 
next step, but I expect some gasolene will locate that.

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Pncconf VFD Spindle scale bug and fix

[Jason Burton]

Just an heads up to the list, and for future Google searchers.

First, let me thank all you developers for building pncconf in the first
place. Great tool for quick setup and learning.

I have been working on getting my VFD mill spindle to work with proper
scaling for some time now.

Turns out Pncconf was generating a couple lines incorrectly.  Easy fix only
if you know what is wrong.  Thanks for the patient help Peter Wallace!

Hardware: big vertical machining center + Mesa 5i25/7i77 + 0-10 VDC
controlled VFD spindle.

Software: all recent linuxcnc/pncconf versions dating from current live cd
Debian with master branch to at least as far back as the Ubuntu 10 live cd.

Both the problem and the fix are described here:

https://sourceforge.net/p/emc/bugs/433/


Best,
Jason Burton
(LatheBuilder)
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users