Re: [Emc-users] To PCW: retuning for inches

2015-10-12 Thread Gene Heskett 
On Monday 12 October 2015 18:02:33 Peter C. Wallace wrote:

> On Mon, 12 Oct 2015, Gene Heskett wrote:
> > Date: Mon, 12 Oct 2015 17:02:20 -0400
> > From: Gene Heskett 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] To PCW: retuning for inches
> >
> > On Monday 12 October 2015 16:10:47 Peter C. Wallace wrote:
> >> On Mon, 12 Oct 2015, Gene Heskett wrote:
> >>> Date: Mon, 12 Oct 2015 15:56:07 -0400
> >>> From: Gene Heskett 
> >>> Reply-To: "Enhanced Machine Controller (EMC)"
> >>> 
> >>> To: "Enhanced Machine Controller (EMC)"
> >>>  Subject: [Emc-users] To PCW:
> >>> retuning for inches
> >>>
> >>> Greetings Peter;
> >>>
> >>> In getting a decent rapids w/o following error, I wound up with
> >>> all the pid.Pgains at 2000, and while I played with the FF2
> >>> setting, couldn't detect that is was affecting anything I'm smart
> >>> enough to detect.
> >>
> >> PGain should really be 1/period, more _will_ cause
> >> overshoot/oscillations
> >>
> >>> However, after a move of one of the axis's I can hear a faint tone
> >>> at the servo-thread frequency from the motors. Firing up halscope
> >>> and looking at the x,y,x outputs, when its singing I can see this
> >>> tone at all the pid.outputs, at from 15 to 50 milivolts peak to
> >>> peak.  Move an axis a couple thou and it may stop.
> >>
> >> This is most likely the step driver (or excessive Pgain, see above)
> >
> > I'll get following errors fairly quickly at 1000.
>
> Then something is really broken, you might have to post your hal/ini
> files (also FF2 of .0005 is way to large .5 shoud be closer)
>
Done, but it will be tomorrow before I test it now.  Between clearing a 
gutter with a major clog, and making a piece of 1/2" copper pipe about 
an inch long into a bent in fingers to engage the output gear on the 
gauge flipping motor, required to it can slide a bit due to 
miss-alignment, and that was battling the o-ring I put in to pull the 
end play out of the piano hinge. So my x results for a G38.2 were 
randomized by 2 or 3 thou.

I had copied out the needed pieces of the jig finder, removed the G10 L2 
stuff and just (debug, etc) the output to see how consistent it wasn't, 
x in particular was wandering around at least 2 thou, and z was still 
walking down half a thou or so per repeat pass. That could be thermal I 
suppose. So I just stretched the timings on that DM860 driver another .5 
microseconds, which might make that a little more repeatable.  Fiddled 
with the backlash until a .005 move was a .005 move going both ways, on 
sll 3 active axis's then setup a 1" travel indicator & made the scales 
produce within a thou of an inch when jogged an inch both ways.  For 
some UNK reason, my math from a TI-030x for the scale changes was 
spitting out figures that were about 5 thou off. Either way.

The bit of copper is working (I cut with a 1/16" carbide mill, 8 slots 
about .095" wide & .3" deep, which were then crushed into the gaps 
between the gear teeth, sloppily but also slides endways about 1/8" as 
the motor moves, so its not pushing on the o-ring taking up the endplay 
in the piano hinge, but I need to find the next size bigger hose clamp 
to reduce the chance of the hose slipping on the pipe.  Cheap, some sort 
of black plastic hose, I had to setup the hot air wand to about 230C to 
make it soggy enough I could stretch it over the 1/2" copper pipe. 
Cooled off, it seems to have enough grip, but based of experience so 
far, I don't trust it without an aeroseal binding it up 1/4 turn from 
stripped.

Now I'd better get the pix out of the camera an posted on my web page.

Thanks Peter.
>
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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>

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

Re: [Emc-users] To PCW: retuning for inches

2015-10-12 Thread Peter C. Wallace 
On Mon, 12 Oct 2015, Gene Heskett wrote:

> Date: Mon, 12 Oct 2015 17:02:20 -0400
> From: Gene Heskett 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] To PCW: retuning for inches
> 
> On Monday 12 October 2015 16:10:47 Peter C. Wallace wrote:
>
>> On Mon, 12 Oct 2015, Gene Heskett wrote:
>>> Date: Mon, 12 Oct 2015 15:56:07 -0400
>>> From: Gene Heskett 
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>> 
>>> To: "Enhanced Machine Controller (EMC)"
>>>  Subject: [Emc-users] To PCW:
>>> retuning for inches
>>>
>>> Greetings Peter;
>>>
>>> In getting a decent rapids w/o following error, I wound up with all
>>> the pid.Pgains at 2000, and while I played with the FF2 setting,
>>> couldn't detect that is was affecting anything I'm smart enough to
>>> detect.
>>
>> PGain should really be 1/period, more _will_ cause
>> overshoot/oscillations
>>
>>> However, after a move of one of the axis's I can hear a faint tone
>>> at the servo-thread frequency from the motors. Firing up halscope
>>> and looking at the x,y,x outputs, when its singing I can see this
>>> tone at all the pid.outputs, at from 15 to 50 milivolts peak to
>>> peak.  Move an axis a couple thou and it may stop.
>>
>> This is most likely the step driver (or excessive Pgain, see above)
>
> I'll get following errors fairly quickly at 1000.

Then something is really broken, you might have to post your hal/ini files
(also FF2 of .0005 is way to large .5 shoud be closer)




Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


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

Re: [Emc-users] To PCW: retuning for inches

2015-10-12 Thread Gene Heskett 
On Monday 12 October 2015 16:10:47 Peter C. Wallace wrote:

> On Mon, 12 Oct 2015, Gene Heskett wrote:
> > Date: Mon, 12 Oct 2015 15:56:07 -0400
> > From: Gene Heskett 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: "Enhanced Machine Controller (EMC)"
> >  Subject: [Emc-users] To PCW:
> > retuning for inches
> >
> > Greetings Peter;
> >
> > In getting a decent rapids w/o following error, I wound up with all
> > the pid.Pgains at 2000, and while I played with the FF2 setting,
> > couldn't detect that is was affecting anything I'm smart enough to
> > detect.
>
> PGain should really be 1/period, more _will_ cause
> overshoot/oscillations
>
> > However, after a move of one of the axis's I can hear a faint tone
> > at the servo-thread frequency from the motors. Firing up halscope
> > and looking at the x,y,x outputs, when its singing I can see this
> > tone at all the pid.outputs, at from 15 to 50 milivolts peak to
> > peak.  Move an axis a couple thou and it may stop.
>
> This is most likely the step driver (or excessive Pgain, see above)

I'll get following errors fairly quickly at 1000.

> > However, when its doing it, the z, being tasked with supporting the
> > weight of the head, can be seen to move, perhaps as much as 5 thou
> > in a minute or less.  I have twice added half a microsecond to the
> > dirsetup and dirhold settings, and a quarter to the step widths
> > which seems to have helped.  This particular driver is a DM860 since
> > thats a nema34 motor, running on 60 volts.  And 60 is NOT enough for
> > that high of an inductance motor.  120 would be about right, and
> > should put its speeds up there with the 450 oz x&y motors, but no
> > one makes a stepper driver rated for that high a voltage.  The DM860
> > is also noisey as can be as it switches from the microstep chosen to
> > lower values as the speeds requested rises. 3IPM is silent, 30 IPM
> > positively howls, its like a 16 speed road ranger truck
> > transmission, taking the engine thru its optimum power rpm range
> > from the traffic light till its out of sight and hearing.
> >
> > So my guess is that the oscillation may be confusing its internal
> > gearshift.
> >
> > What has been the pulse timing settings a DM860 needs in your
> > experience, and is there something in the pid settings that will
> > dampen that 1 kiloihertz oscillation out, without lowering the
> > Pgain, which seems to be required to get any real rapids speeds
> > above 20 ipm.
>
> If you cannnot get rapids above 20 IPM without a Pgain of 2000 you
> have something badly broken in the PID ( _is_ FF1 1.000? )
>
Yes, to about 4 decimal places. 

> I would set the pulse timimg as long as the maximum step rate will
> allow (with about a 20% margin)

That would then be about 40% of the 1/scale then?  Scale by the time I 
get it moving exactly an inch for an inch requested, is about 5 digits 
either side of the decimal point.

Here is the .ini as it exists right now.

I *think* I have done as suggested.

> > Clues will be gleefully applied at this point.  Thanks Peter.
> >
> > 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>
> >
> > 
> >-- ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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>
# Generated by PNCconf at Tue Jun 30 22:44:28 2015
# If you make changes to this file, they will be
# overwritten when you run PNCconf again

[EMC]
MACHINE = GO704fast
DEBUG = 0x

[DISPLAY]
DISPLAY = axis
PYVCP = pyvcp-panel.xml
#GLADEVCP = -U --catalog=mill-mm feat

Re: [Emc-users] To PCW: retuning for inches

2015-10-12 Thread Peter C. Wallace 
On Mon, 12 Oct 2015, Gene Heskett wrote:

> Date: Mon, 12 Oct 2015 15:56:07 -0400
> From: Gene Heskett 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: [Emc-users] To PCW: retuning for inches
> 
> Greetings Peter;
>
> In getting a decent rapids w/o following error, I wound up with all the
> pid.Pgains at 2000, and while I played with the FF2 setting, couldn't
> detect that is was affecting anything I'm smart enough to detect.

PGain should really be 1/period, more _will_ cause overshoot/oscillations


>
> However, after a move of one of the axis's I can hear a faint tone at the
> servo-thread frequency from the motors. Firing up halscope and looking
> at the x,y,x outputs, when its singing I can see this tone at all the
> pid.outputs, at from 15 to 50 milivolts peak to peak.  Move an axis a
> couple thou and it may stop.

This is most likely the step driver (or excessive Pgain, see above)

>
> However, when its doing it, the z, being tasked with supporting the
> weight of the head, can be seen to move, perhaps as much as 5 thou in a
> minute or less.  I have twice added half a microsecond to the dirsetup
> and dirhold settings, and a quarter to the step widths which seems to
> have helped.  This particular driver is a DM860 since thats a nema34
> motor, running on 60 volts.  And 60 is NOT enough for that high of an
> inductance motor.  120 would be about right, and should put its speeds
> up there with the 450 oz x&y motors, but no one makes a stepper driver
> rated for that high a voltage.  The DM860 is also noisey as can be as it
> switches from the microstep chosen to lower values as the speeds
> requested rises. 3IPM is silent, 30 IPM positively howls, its like a 16
> speed road ranger truck transmission, taking the engine thru its optimum
> power rpm range from the traffic light till its out of sight and
> hearing.
>
> So my guess is that the oscillation may be confusing its internal
> gearshift.
>
> What has been the pulse timing settings a DM860 needs in your experience,
> and is there something in the pid settings that will dampen that 1
> kiloihertz oscillation out, without lowering the Pgain, which seems to
> be required to get any real rapids speeds above 20 ipm.
>

If you cannnot get rapids above 20 IPM without a Pgain of 2000 you have 
something badly broken in the PID ( _is_ FF1 1.000? )




I would set the pulse timimg as long as the maximum step rate will allow
(with about a 20% margin)


> Clues will be gleefully applied at this point.  Thanks Peter.
>
> 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>
>
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


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

[Emc-users] To PCW: retuning for inches

2015-10-12 Thread Gene Heskett 
Greetings Peter;

In getting a decent rapids w/o following error, I wound up with all the 
pid.Pgains at 2000, and while I played with the FF2 setting, couldn't 
detect that is was affecting anything I'm smart enough to detect.

However, after a move of one of the axis's I can hear a faint tone at the 
servo-thread frequency from the motors. Firing up halscope and looking 
at the x,y,x outputs, when its singing I can see this tone at all the 
pid.outputs, at from 15 to 50 milivolts peak to peak.  Move an axis a 
couple thou and it may stop.

However, when its doing it, the z, being tasked with supporting the 
weight of the head, can be seen to move, perhaps as much as 5 thou in a 
minute or less.  I have twice added half a microsecond to the dirsetup 
and dirhold settings, and a quarter to the step widths which seems to 
have helped.  This particular driver is a DM860 since thats a nema34 
motor, running on 60 volts.  And 60 is NOT enough for that high of an 
inductance motor.  120 would be about right, and should put its speeds 
up there with the 450 oz x&y motors, but no one makes a stepper driver 
rated for that high a voltage.  The DM860 is also noisey as can be as it 
switches from the microstep chosen to lower values as the speeds 
requested rises. 3IPM is silent, 30 IPM positively howls, its like a 16 
speed road ranger truck transmission, taking the engine thru its optimum 
power rpm range from the traffic light till its out of sight and 
hearing.

So my guess is that the oscillation may be confusing its internal 
gearshift.

What has been the pulse timing settings a DM860 needs in your experience, 
and is there something in the pid settings that will dampen that 1 
kiloihertz oscillation out, without lowering the Pgain, which seems to 
be required to get any real rapids speeds above 20 ipm.

Clues will be gleefully applied at this point.  Thanks Peter.

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