Re: [Emc-users] Two questions.

[Peter C. Wallace]

On Wed, 11 Jan 2017, Gene Heskett wrote:

> Date: Wed, 11 Jan 2017 23:11:51 -0500
> From: Gene Heskett 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Two questions.
> 
> On Wednesday 11 January 2017 22:37:46 Peter C. Wallace wrote:
>
>> On Wed, 11 Jan 2017, Gene Heskett wrote:
>>> Date: Wed, 11 Jan 2017 21:50:19 -0500
>>> From: Gene Heskett 
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>> 
>>> To: emc-users@lists.sourceforge.net
>>> Subject: Re: [Emc-users] Two questions.
>>>
>>> On Wednesday 11 January 2017 13:34:50 Peter C. Wallace wrote:
 On Wed, 11 Jan 2017, Jon Elson wrote:
> Date: Wed, 11 Jan 2017 10:36:18 -0600
> From: Jon Elson 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)"
>  Subject: Re: [Emc-users] Two
> questions.
>
> On 01/10/2017 10:45 PM, Gene Heskett wrote:
>> On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
>>> On 01/10/2017 08:39 PM, Gene Heskett wrote:
 But if I hook up the halscope, the apparent quadrature error is
 worthless due to the 1KHz servo-loop timing limiting the
 bandwidth.
>>>
>>> Well, that's why we use hardware encoder counters, as
>>> sampling the quadrature at 1 KHz just isn't fast enough.
>>>
>>> Jon
>>
>> Plz elucidate Jon. Something has got to be better than this. See
>> the pix I posted with the last reply to Peter & the list.  The
>> top, white trace in the halscope screen is the quadrature noise,
>> easily 10x the timing wibbles I see on the hitachi's screen.
>
> If you are looking at the quadrature signals at a 1 KHz
> sampling rate, then unless the encoder is moving quite
> slowly, the sampling will be too coarse.  That screen shot
> is kind of small (you can take just a single window as the
> screenshot, by the way) but it looks like the duty cycle of
> the encoders is not 50%.  That may well be aliasing, as the
> quadrature signals seem to be moving much too close to 1000
> Hz.  You'd get much better results below 100 Hz.  Or, with
> Mesa hardware, you can sample these with the base thread.
>
> I can't comment on the velocity output from the Mesa, as I
> don't know their gear (or driver).  But, the scale is 2/div,
> meaning the scaled velocity output is jumping around 2 user
> units up and down at a KHz rate.  Clearly, not good.
>
> Jon

 The velocity output is calculated by using the number of counts
 divided by the time between the counts (time measured via a 1 MHz
 timestamp). If you have less than 2 counts per servo period the
 velocity estimation accuracy depends heavily the quadrature
 phase/symmetry since the edge spacing entirely determines the
 estimated velocity at this speed range.

 I suspect what Genes plot shows is that he has significant
 quadrature phase/symmetry errors at the FPGA inputs.


 Again here is a velocity plot of a accurate 100 Hz quadrature
 signal:

 http://freeby.mesanet.com/100_Hz_quadrature_velocity.png
>>>
>>> I've not played with the stepgen type 2 yet, but I have adjusted the
>>> quadrature a few thousandths and reduced it by about 1/3rd. Thats
>>> helpfull as I can filter to smooth a tach dial once I have defeated
>>> the noise.
>>>
>>> But I have another concern. I have had to do a full powerdown reset
>>> to clear a watchdog bite. The original config was I think intended
>>> to clear it by re-enabling the machine, but I could nolinuxcnct, so
>>> I disconnected
>>
>> that
>>
>>> reset in the hal file I started with as the logic was self feeding,
>>> locking out the ability to reset the disabled machine.  And I
>>> couldn't even do it from the halconfig as it claimed another signal
>>> was driving it when I tried to setp that pin.
>>>
>>> Do you, Peter, have a logic diagram I can re-construct for that,
>>> that actually works?
>>
>> You need to restart linuxcnc to clear a watchdog bite
>>
 Peter Wallace
 Mesa Electronics

>
> That didn't work, I have since had to do full power downs with at least
> 15 seconds off time. 5 or 6 times now.

That probably means the  watchdog bite is not the issue but a symptom
of some kind of system crash


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


 ---
 ---  Developer Access Program for Intel Xeon Phi Processors
 Access to Intel Xeon Phi processor-based developer platforms. With
 one year of Intel Parallel Studio XE.
 

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 22:37:46 Peter C. Wallace wrote:

> On Wed, 11 Jan 2017, Gene Heskett wrote:
> > Date: Wed, 11 Jan 2017 21:50:19 -0500
> > From: Gene Heskett 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] Two questions.
> >
> > On Wednesday 11 January 2017 13:34:50 Peter C. Wallace wrote:
> >> On Wed, 11 Jan 2017, Jon Elson wrote:
> >>> Date: Wed, 11 Jan 2017 10:36:18 -0600
> >>> From: Jon Elson 
> >>> Reply-To: "Enhanced Machine Controller (EMC)"
> >>> 
> >>> To: "Enhanced Machine Controller (EMC)"
> >>>  Subject: Re: [Emc-users] Two
> >>> questions.
> >>>
> >>> On 01/10/2017 10:45 PM, Gene Heskett wrote:
>  On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
> > On 01/10/2017 08:39 PM, Gene Heskett wrote:
> >> But if I hook up the halscope, the apparent quadrature error is
> >> worthless due to the 1KHz servo-loop timing limiting the
> >> bandwidth.
> >
> > Well, that's why we use hardware encoder counters, as
> > sampling the quadrature at 1 KHz just isn't fast enough.
> >
> > Jon
> 
>  Plz elucidate Jon. Something has got to be better than this. See
>  the pix I posted with the last reply to Peter & the list.  The
>  top, white trace in the halscope screen is the quadrature noise,
>  easily 10x the timing wibbles I see on the hitachi's screen.
> >>>
> >>> If you are looking at the quadrature signals at a 1 KHz
> >>> sampling rate, then unless the encoder is moving quite
> >>> slowly, the sampling will be too coarse.  That screen shot
> >>> is kind of small (you can take just a single window as the
> >>> screenshot, by the way) but it looks like the duty cycle of
> >>> the encoders is not 50%.  That may well be aliasing, as the
> >>> quadrature signals seem to be moving much too close to 1000
> >>> Hz.  You'd get much better results below 100 Hz.  Or, with
> >>> Mesa hardware, you can sample these with the base thread.
> >>>
> >>> I can't comment on the velocity output from the Mesa, as I
> >>> don't know their gear (or driver).  But, the scale is 2/div,
> >>> meaning the scaled velocity output is jumping around 2 user
> >>> units up and down at a KHz rate.  Clearly, not good.
> >>>
> >>> Jon
> >>
> >> The velocity output is calculated by using the number of counts
> >> divided by the time between the counts (time measured via a 1 MHz
> >> timestamp). If you have less than 2 counts per servo period the
> >> velocity estimation accuracy depends heavily the quadrature
> >> phase/symmetry since the edge spacing entirely determines the
> >> estimated velocity at this speed range.
> >>
> >> I suspect what Genes plot shows is that he has significant
> >> quadrature phase/symmetry errors at the FPGA inputs.
> >>
> >>
> >> Again here is a velocity plot of a accurate 100 Hz quadrature
> >> signal:
> >>
> >> http://freeby.mesanet.com/100_Hz_quadrature_velocity.png
> >
> > I've not played with the stepgen type 2 yet, but I have adjusted the
> > quadrature a few thousandths and reduced it by about 1/3rd. Thats
> > helpfull as I can filter to smooth a tach dial once I have defeated
> > the noise.
> >
> > But I have another concern. I have had to do a full powerdown reset
> > to clear a watchdog bite. The original config was I think intended
> > to clear it by re-enabling the machine, but I could nolinuxcnct, so
> > I disconnected
>
> that
>
> > reset in the hal file I started with as the logic was self feeding,
> > locking out the ability to reset the disabled machine.  And I
> > couldn't even do it from the halconfig as it claimed another signal
> > was driving it when I tried to setp that pin.
> >
> > Do you, Peter, have a logic diagram I can re-construct for that,
> > that actually works?
>
> You need to restart linuxcnc to clear a watchdog bite
>
> >> Peter Wallace
> >> Mesa Electronics
> >>

That didn't work, I have since had to do full power downs with at least 
15 seconds off time. 5 or 6 times now.

> >> (\__/)
> >> (='.'=) This is Bunny. Copy and paste bunny into your
> >> (")_(") signature to help him gain world domination.
> >>
> >>
> >> ---
> >>---  Developer Access Program for Intel Xeon Phi Processors
> >> Access to Intel Xeon Phi processor-based developer platforms. With
> >> one year of Intel Parallel Studio XE.
> >> Training and support from Colfax.
> >> Order your platform today. http://sdm.link/xeonphi
> >> ___
> >> 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 

Re: [Emc-users] Two questions.

[Peter C. Wallace]

On Wed, 11 Jan 2017, Gene Heskett wrote:

> Date: Wed, 11 Jan 2017 21:50:19 -0500
> From: Gene Heskett 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Two questions.
> 
> On Wednesday 11 January 2017 13:34:50 Peter C. Wallace wrote:
>
>> On Wed, 11 Jan 2017, Jon Elson wrote:
>>> Date: Wed, 11 Jan 2017 10:36:18 -0600
>>> From: Jon Elson 
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>> 
>>> To: "Enhanced Machine Controller (EMC)"
>>>  Subject: Re: [Emc-users] Two
>>> questions.
>>>
>>> On 01/10/2017 10:45 PM, Gene Heskett wrote:
 On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
> On 01/10/2017 08:39 PM, Gene Heskett wrote:
>> But if I hook up the halscope, the apparent quadrature error is
>> worthless due to the 1KHz servo-loop timing limiting the
>> bandwidth.
>
> Well, that's why we use hardware encoder counters, as
> sampling the quadrature at 1 KHz just isn't fast enough.
>
> Jon

 Plz elucidate Jon. Something has got to be better than this. See
 the pix I posted with the last reply to Peter & the list.  The top,
 white trace in the halscope screen is the quadrature noise, easily
 10x the timing wibbles I see on the hitachi's screen.
>>>
>>> If you are looking at the quadrature signals at a 1 KHz
>>> sampling rate, then unless the encoder is moving quite
>>> slowly, the sampling will be too coarse.  That screen shot
>>> is kind of small (you can take just a single window as the
>>> screenshot, by the way) but it looks like the duty cycle of
>>> the encoders is not 50%.  That may well be aliasing, as the
>>> quadrature signals seem to be moving much too close to 1000
>>> Hz.  You'd get much better results below 100 Hz.  Or, with
>>> Mesa hardware, you can sample these with the base thread.
>>>
>>> I can't comment on the velocity output from the Mesa, as I
>>> don't know their gear (or driver).  But, the scale is 2/div,
>>> meaning the scaled velocity output is jumping around 2 user
>>> units up and down at a KHz rate.  Clearly, not good.
>>>
>>> Jon
>>
>> The velocity output is calculated by using the number of counts
>> divided by the time between the counts (time measured via a 1 MHz
>> timestamp). If you have less than 2 counts per servo period the
>> velocity estimation accuracy depends heavily the quadrature
>> phase/symmetry since the edge spacing entirely determines the
>> estimated velocity at this speed range.
>>
>> I suspect what Genes plot shows is that he has significant quadrature
>> phase/symmetry errors at the FPGA inputs.
>>
>>
>> Again here is a velocity plot of a accurate 100 Hz quadrature signal:
>>
>> http://freeby.mesanet.com/100_Hz_quadrature_velocity.png
>
> I've not played with the stepgen type 2 yet, but I have adjusted the
> quadrature a few thousandths and reduced it by about 1/3rd. Thats
> helpfull as I can filter to smooth a tach dial once I have defeated the
> noise.
>
> But I have another concern. I have had to do a full powerdown reset to
> clear a watchdog bite. The original config was I think intended to clear
> it by re-enabling the machine, but I could nolinuxcnct, so I disconnected 
that
> reset in the hal file I started with as the logic was self feeding,
> locking out the ability to reset the disabled machine.  And I couldn't
> even do it from the halconfig as it claimed another signal was driving
> it when I tried to setp that pin.
>
> Do you, Peter, have a logic diagram I can re-construct for that, that
> actually works?




You need to restart linuxcnc to clear a watchdog bite

>>
>>
>>
>> Peter Wallace
>> Mesa Electronics
>>
>> (\__/)
>> (='.'=) This is Bunny. Copy and paste bunny into your
>> (")_(") signature to help him gain world domination.
>>
>>
>> --
>>  Developer Access Program for Intel Xeon Phi Processors
>> Access to Intel Xeon Phi processor-based developer platforms.
>> With one year of Intel Parallel Studio XE.
>> Training and support from Colfax.
>> Order your platform today. http://sdm.link/xeonphi
>> ___
>> 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 
>
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your 

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 13:34:50 Peter C. Wallace wrote:

> On Wed, 11 Jan 2017, Jon Elson wrote:
> > Date: Wed, 11 Jan 2017 10:36:18 -0600
> > From: Jon Elson 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: "Enhanced Machine Controller (EMC)"
> >  Subject: Re: [Emc-users] Two
> > questions.
> >
> > On 01/10/2017 10:45 PM, Gene Heskett wrote:
> >> On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
> >>> On 01/10/2017 08:39 PM, Gene Heskett wrote:
>  But if I hook up the halscope, the apparent quadrature error is
>  worthless due to the 1KHz servo-loop timing limiting the
>  bandwidth.
> >>>
> >>> Well, that's why we use hardware encoder counters, as
> >>> sampling the quadrature at 1 KHz just isn't fast enough.
> >>>
> >>> Jon
> >>
> >> Plz elucidate Jon. Something has got to be better than this. See
> >> the pix I posted with the last reply to Peter & the list.  The top,
> >> white trace in the halscope screen is the quadrature noise, easily
> >> 10x the timing wibbles I see on the hitachi's screen.
> >
> > If you are looking at the quadrature signals at a 1 KHz
> > sampling rate, then unless the encoder is moving quite
> > slowly, the sampling will be too coarse.  That screen shot
> > is kind of small (you can take just a single window as the
> > screenshot, by the way) but it looks like the duty cycle of
> > the encoders is not 50%.  That may well be aliasing, as the
> > quadrature signals seem to be moving much too close to 1000
> > Hz.  You'd get much better results below 100 Hz.  Or, with
> > Mesa hardware, you can sample these with the base thread.
> >
> > I can't comment on the velocity output from the Mesa, as I
> > don't know their gear (or driver).  But, the scale is 2/div,
> > meaning the scaled velocity output is jumping around 2 user
> > units up and down at a KHz rate.  Clearly, not good.
> >
> > Jon
>
> The velocity output is calculated by using the number of counts
> divided by the time between the counts (time measured via a 1 MHz
> timestamp). If you have less than 2 counts per servo period the
> velocity estimation accuracy depends heavily the quadrature
> phase/symmetry since the edge spacing entirely determines the
> estimated velocity at this speed range.
>
> I suspect what Genes plot shows is that he has significant quadrature
> phase/symmetry errors at the FPGA inputs.
>
>
> Again here is a velocity plot of a accurate 100 Hz quadrature signal:
>
> http://freeby.mesanet.com/100_Hz_quadrature_velocity.png

I've not played with the stepgen type 2 yet, but I have adjusted the 
quadrature a few thousandths and reduced it by about 1/3rd. Thats 
helpfull as I can filter to smooth a tach dial once I have defeated the  
noise.

But I have another concern. I have had to do a full powerdown reset to 
clear a watchdog bite. The original config was I think intended to clear 
it by re-enabling the machine, but I could not, so I disconnected that 
reset in the hal file I started with as the logic was self feeding, 
locking out the ability to reset the disabled machine.  And I couldn't 
even do it from the halconfig as it claimed another signal was driving 
it when I tried to setp that pin.

Do you, Peter, have a logic diagram I can re-construct for that, that 
actually works?
>
>
>
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your platform today. http://sdm.link/xeonphi
> ___
> 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 

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[W. Martinjak]

On 2017-01-11 18:18, Nicklas Karlsson wrote:
> Yes it is way to slow, hardware counters are cheap and work well, do you have 
> them in you PI board?

These are hardware counters sitting in the Mesa card.

-- 
"In der Wissenschaft siegt nie eine neue Theorie,
nur ihre Gegner sterben nach und nach"

Max Planck


--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 20:34:58 Gene Heskett wrote:

> On Wednesday 11 January 2017 12:18:54 Nicklas Karlsson wrote:
> > Yes it is way to slow, hardware counters are cheap and work well, do
> > you have them in you PI board?
>
> This is my first pi project,its a raspi 3b, and I am learning as I go
> while I bring this machine to life.
>
> ATM its sitting there, with the back gear engaged, belts in lowest
> gear, turning at about 1/2 an rps as the vfd is sitting at about 20.5
> HZ. More to put time on a minmax looking for some huge bursts of speed
> from .5 rps, to 25k revs either way.  I hope I have that fixed, and
> that all I have to do with the x/z steppers when I finally bring those
> up, is more of the same that seems to have been the magic twinky to
> shut the vfd up.
>
> First time I've ever had to deal with an spi bus thats running at
> 32Megabaud when the noise from the switching psu, or the driver itself
> is at 32megabaud, and the noise bouncing around is at 50 to 150
> megahertz. 3 volt p-p buss, noise is faster than the buss and 6 volts
> or more p-p...
>
> It certainly enforces the idea of a single point ground, with BIG,
> Wide conductors.  So far , thats working.

And two hours later its still lazing along at about a third of a turn per 
second, and no noise bursts have been caught by the minmax module.  
Thats encouraging.

So tommorrow, do the same thing with each stepper driver and exercise it 
for odd motions.  Then I can start nailing down home & limit switches 
one axis at a time.  Then grab the taper rig out of the back of the GMC, 
and use its edge of the bed anchor to hold a hinge so I can close a 
swarf cover over the chuck to contain the swarf, make me a tool tip 
height gage, round up some more tools and tool holders, and put it to 
work.

I note there are some (4 or so) 1/2" holes in the chip pan but haven't 
looked for suitable drainage fittings to recover coolant, which will be 
needed for serious work.  One thing at a time though. They currently 
would drain into the left and right drawers in the pedestal, which would 
not be a GOOD THING.


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 

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 12:18:54 Nicklas Karlsson wrote:

> Yes it is way to slow, hardware counters are cheap and work well, do
> you have them in you PI board?
>
This is my first pi project,its a raspi 3b, and I am learning as I go 
while I bring this machine to life.

ATM its sitting there, with the back gear engaged, belts in lowest gear, 
turning at about 1/2 an rps as the vfd is sitting at about 20.5 HZ. More 
to put time on a minmax looking for some huge bursts of speed from .5 
rps, to 25k revs either way.  I hope I have that fixed, and that all I 
have to do with the x/z steppers when I finally bring those up, is more 
of the same that seems to have been the magic twinky to shut the vfd up.

First time I've ever had to deal with an spi bus thats running at 
32Megabaud when the noise from the switching psu, or the driver itself 
is at 32megabaud, and the noise bouncing around is at 50 to 150 
megahertz. 3 volt p-p buss, noise is faster than the buss and 6 volts or 
more p-p...

It certainly enforces the idea of a single point ground, with BIG, Wide 
conductors.  So far , thats working.

> On Tue, 10 Jan 2017 22:02:28 -0600
>
> Jon Elson  wrote:
> > On 01/10/2017 08:39 PM, Gene Heskett wrote:
> > > But if I hook up the halscope, the apparent quadrature error is
> > > worthless due to the 1KHz servo-loop timing limiting the
> > > bandwidth.
> >
> > Well, that's why we use hardware encoder counters, as
> > sampling the quadrature at 1 KHz just isn't fast enough.
> >
> > Jon
> >
> > 
> >-- Developer Access Program for Intel Xeon Phi Processors
> > Access to Intel Xeon Phi processor-based developer platforms.
> > With one year of Intel Parallel Studio XE.
> > Training and support from Colfax.
> > Order your platform today. http://sdm.link/xeonphi
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your platform today. http://sdm.link/xeonphi
> ___
> 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 

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[samco]

http://electronicsam.com/images/matsuura/oscope.jpg

Part of what worked for us was we found a gear that had a profile close to the 
test gear from the alero docs.

http://electronicsam.com/images/matsuura/sensors.jpg

Been working great



On Wed, 11 Jan 2017 16:03:10 -0500
 Gene Heskett  wrote:
> On Wednesday 11 January 2017 11:36:18 Jon Elson wrote:
> 
> > On 01/10/2017 10:45 PM, Gene Heskett wrote:
> > > On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
> > >> On 01/10/2017 08:39 PM, Gene Heskett wrote:
> > >>> But if I hook up the halscope, the apparent quadrature error is
> > >>> worthless due to the 1KHz servo-loop timing limiting the
> > >>> bandwidth.
> > >>
> > >> Well, that's why we use hardware encoder counters, as
> > >> sampling the quadrature at 1 KHz just isn't fast enough.
> > >>
> > >> Jon
> > >
> > > Plz elucidate Jon. Something has got to be better than this. See the
> > > pix I posted with the last reply to Peter & the list.  The top,
> > > white trace in the halscope screen is the quadrature noise, easily
> > > 10x the timing wibbles I see on the hitachi's screen.
> >
> > If you are looking at the quadrature signals at a 1 KHz
> > sampling rate, then unless the encoder is moving quite
> > slowly, the sampling will be too coarse.  That screen shot
> > is kind of small (you can take just a single window as the
> > screenshot, by the way) but it looks like the duty cycle of
> > the encoders is not 50%.  That may well be aliasing, as the
> > quadrature signals seem to be moving much too close to 1000
> > Hz.  You'd get much better results below 100 Hz.  Or, with
> > Mesa hardware, you can sample these with the base thread.
> >
> > I can't comment on the velocity output from the Mesa, as I
> > don't know their gear (or driver).  But, the scale is 2/div,
> > meaning the scaled velocity output is jumping around 2 user
> > units up and down at a KHz rate.  Clearly, not good.
> >
> > Jon
> >
> I have found some quadrature error, so I pulled the ats667 mount off and 
> carved about 5 thou off the divider between the A & B units, jammed a 
> brass shim into the side of the B pocket to push it toward theA unit, 
> carved about 5 thou off the side of the divider between them, washed the 
> oil off several times, glued the A unit in place with a brass shim 
> pushing it snugly toward the B, jammed a piece of rubber beside the B 
> unit to push it toward the A unit, perhaps 10 thou change in the 
> separation, and I'm headed back out to test. Breaking the B unit loose, 
> and filing about 3 or 4 thou off that side of the B  unit so the rubber 
> shim would shove it that much closer to together raised the minmax about 
> 10 points on the min, and lowered the max about 25, which have be an 
> indicator as to the mechanical directions needed.
> 
> Oh, that ground on the vfd control cable I thought might be loose, was.
> 
> the diff in duty cycle, aka time up vs time down? Its about 60/40. If it 
> was optical, I'd put a pot in series with the leds and adjust for 50% :) 
> But I haven't a clue how to do that with these ats667's.  Darn it.
> 
> 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 
> 
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your platform today. http://sdm.link/xeonphi
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 11:36:18 Jon Elson wrote:

> On 01/10/2017 10:45 PM, Gene Heskett wrote:
> > On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
> >> On 01/10/2017 08:39 PM, Gene Heskett wrote:
> >>> But if I hook up the halscope, the apparent quadrature error is
> >>> worthless due to the 1KHz servo-loop timing limiting the
> >>> bandwidth.
> >>
> >> Well, that's why we use hardware encoder counters, as
> >> sampling the quadrature at 1 KHz just isn't fast enough.
> >>
> >> Jon
> >
> > Plz elucidate Jon. Something has got to be better than this. See the
> > pix I posted with the last reply to Peter & the list.  The top,
> > white trace in the halscope screen is the quadrature noise, easily
> > 10x the timing wibbles I see on the hitachi's screen.
>
> If you are looking at the quadrature signals at a 1 KHz
> sampling rate, then unless the encoder is moving quite
> slowly, the sampling will be too coarse.  That screen shot
> is kind of small (you can take just a single window as the
> screenshot, by the way) but it looks like the duty cycle of
> the encoders is not 50%.  That may well be aliasing, as the
> quadrature signals seem to be moving much too close to 1000
> Hz.  You'd get much better results below 100 Hz.  Or, with
> Mesa hardware, you can sample these with the base thread.
>
> I can't comment on the velocity output from the Mesa, as I
> don't know their gear (or driver).  But, the scale is 2/div,
> meaning the scaled velocity output is jumping around 2 user
> units up and down at a KHz rate.  Clearly, not good.
>
> Jon
>
I have found some quadrature error, so I pulled the ats667 mount off and 
carved about 5 thou off the divider between the A & B units, jammed a 
brass shim into the side of the B pocket to push it toward theA unit, 
carved about 5 thou off the side of the divider between them, washed the 
oil off several times, glued the A unit in place with a brass shim 
pushing it snugly toward the B, jammed a piece of rubber beside the B 
unit to push it toward the A unit, perhaps 10 thou change in the 
separation, and I'm headed back out to test. Breaking the B unit loose, 
and filing about 3 or 4 thou off that side of the B  unit so the rubber 
shim would shove it that much closer to together raised the minmax about 
10 points on the min, and lowered the max about 25, which have be an 
indicator as to the mechanical directions needed.

Oh, that ground on the vfd control cable I thought might be loose, was.

the diff in duty cycle, aka time up vs time down? Its about 60/40. If it 
was optical, I'd put a pot in series with the leds and adjust for 50% :) 
But I haven't a clue how to do that with these ats667's.  Darn it.

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 

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Peter C. Wallace]

On Wed, 11 Jan 2017, Jon Elson wrote:

> Date: Wed, 11 Jan 2017 10:36:18 -0600
> From: Jon Elson 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: Re: [Emc-users] Two questions.
> 
> On 01/10/2017 10:45 PM, Gene Heskett wrote:
>> On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
>>
>>> On 01/10/2017 08:39 PM, Gene Heskett wrote:
 But if I hook up the halscope, the apparent quadrature error is
 worthless due to the 1KHz servo-loop timing limiting the bandwidth.
>>> Well, that's why we use hardware encoder counters, as
>>> sampling the quadrature at 1 KHz just isn't fast enough.
>>>
>>> Jon
>> Plz elucidate Jon. Something has got to be better than this. See the pix
>> I posted with the last reply to Peter & the list.  The top, white trace
>> in the halscope screen is the quadrature noise, easily 10x the timing
>> wibbles I see on the hitachi's screen.
>>
>>
> If you are looking at the quadrature signals at a 1 KHz
> sampling rate, then unless the encoder is moving quite
> slowly, the sampling will be too coarse.  That screen shot
> is kind of small (you can take just a single window as the
> screenshot, by the way) but it looks like the duty cycle of
> the encoders is not 50%.  That may well be aliasing, as the
> quadrature signals seem to be moving much too close to 1000
> Hz.  You'd get much better results below 100 Hz.  Or, with
> Mesa hardware, you can sample these with the base thread.
>
> I can't comment on the velocity output from the Mesa, as I
> don't know their gear (or driver).  But, the scale is 2/div,
> meaning the scaled velocity output is jumping around 2 user
> units up and down at a KHz rate.  Clearly, not good.
>
> Jon


The velocity output is calculated by using the number of counts divided by the 
time between the counts (time measured via a 1 MHz timestamp). If you have 
less than 2 counts per servo period the velocity estimation accuracy depends 
heavily the quadrature phase/symmetry since the edge spacing entirely 
determines the estimated velocity at this speed range.

I suspect what Genes plot shows is that he has significant quadrature 
phase/symmetry errors at the FPGA inputs.


Again here is a velocity plot of a accurate 100 Hz quadrature signal:

http://freeby.mesanet.com/100_Hz_quadrature_velocity.png



>

Peter Wallace
Mesa Electronics

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


--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] EDM machining, user interface, increments

[Nicklas Karlsson]

I could confirm it works perfect, it was a really simple fix.

Then I look at code to add a widget does not seem to hard.


> I pushed a fix for hal_dial to 2.7 - not sure what you are using.
> 
> eventually it will cycle thru to master (and later releases)
> 
> 
> 
> The fix is simple (two lines) if you wish to just add it for now.
> 
> 
> in gladevcp/hal_dial.py
> 
> 
> 
> # handle the scroll wheel of the mouse
> def _scroll(self, widget, event):
> if event.direction == gtk.gdk.SCROLL_UP:
> self._count += 1
> self._minute_offset +=1
> self._delta_scaled += self.scale #<< this one
> if event.direction == gtk.gdk.SCROLL_DOWN:
> self._count -= 1
> self._minute_offset -=1
> self._delta_scaled -= self.scale #<<< and this one
> self._last_offset =  self._minute_offset
> self.redraw_canvas()
> self.emit("count_changed", self._count,self.scale,self._delta_scaled)
> 
> I tested on my machine it seemed to work fine after that.
> Chris M
> 
> From: Chris Morley 
> Sent: January 10, 2017 9:51 PM
> To: EMC
> Subject: Re: [Emc-users] EDM machining, user interface, increments
> 
> 
> That sounds like a bug...wheel or mouse should move it the same.
> 
> - Reply message -
> From: "Nicklas Karlsson" 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: [Emc-users] EDM machining, user interface, increments
> Date: Tue, Jan 10, 2017 1:35 PM
> 
> 
> 
> In sort of the delta scale is a little bit slippery. It slip while using the 
> mouse wheel and only graphics is moving but no update of value. Then left 
> mouse button is held down and dial is moved by moving mouse pointer both 
> graphics and value is updated.
> 
> 
> On Tue, 10 Jan 2017 20:45:24 +
> Chris Morley  wrote:
> 
> >
> > Hal dial has a pin called delta scale I think. Then changing the scale (by 
> > clicking) only changes the scale of future dial moments, which is what I 
> > think you want.
> > Chris M
> >
> > - Reply message -
> > From: "Nicklas Karlsson" 
> > To: "Enhanced Machine Controller (EMC)" 
> > Subject: [Emc-users] EDM machining, user interface, increments
> > Date: Tue, Jan 10, 2017 12:34 PM
> >
> >
> >
> > I discovered the limitations of the hal inputs, what is needed is:
> >   min
> >   max
> >   scale
> >   increment value
> >   increment value change from GUI
> >   A choice to between actual or scaled value
> >
> > None of the do it all.The hal dial may change the scale but what is really 
> > needed is possibility to change increment value for fine tuning.
> >
> > Despite these small limitations I still think glade user interface designer 
> > is really good. Since source code is available it is also possible change, 
> > if anybody could give me a hint maybe I could look into it but I better 
> > start with the unfinished g-code database, I think where will be  few hours 
> > for it this weekend.
> >
> >
> >
> > On Tue, 20 Dec 2016 12:04:05 +
> > Sarah Armstrong  wrote:
> >
> > > Ni⁣cklas,
> > > Could you give me details etc, and I can get up to date, and join the 
> > > discussion
> > >
> > > Sarah
> > >
> > > Sent from BlueMail
> > >
> > > On 19 Dec 2016, 20:16, at 20:16, Nicklas Karlsson 
> > >  wrote:
> > > >Finally I have all the necessary hardware parts, some are still on
> > > >desktop not connected to machine but they are tested to work: servo
> > > >drives, digital IO, spark generator.
> > > >
> > > >Right now I start to think about user interface and as I remember
> > > >someone started work on this earlier.
> > > >
> > > >
> > > >If anyone is interested I could share hardware drawings or others. I
> > > >also changed a few rows in drivers so that positions are sent instead
> > > >of speed and put control loop local.
> > > >
> > > >
> > > >
> > > >Regards Nicklas Karlsson
> > > >
> > > >--
> > > >Developer Access Program for Intel Xeon Phi Processors
> > > >Access to Intel Xeon Phi processor-based developer platforms.
> > > >With one year of Intel Parallel Studio XE.
> > > >Training and support from Colfax.
> > > >Order your platform today.http://sdm.link/intel
> > > >___
> > > >Emc-users mailing list
> > > >Emc-users@lists.sourceforge.net
> > > >https://lists.sourceforge.net/lists/listinfo/emc-users
> > > --
> > > Developer Access Program for Intel Xeon Phi Processors
> > > Access to Intel Xeon Phi processor-based developer platforms.
> > > With one year of Intel Parallel Studio XE.
> > > Training and support from 

Re: [Emc-users] Two questions.

[Nicklas Karlsson]

Yes it is way to slow, hardware counters are cheap and work well, do you have 
them in you PI board?

On Tue, 10 Jan 2017 22:02:28 -0600
Jon Elson  wrote:

> On 01/10/2017 08:39 PM, Gene Heskett wrote:
> > But if I hook up the halscope, the apparent quadrature error is worthless
> > due to the 1KHz servo-loop timing limiting the bandwidth.
> Well, that's why we use hardware encoder counters, as 
> sampling the quadrature at 1 KHz just isn't fast enough.
> 
> Jon
> 
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your platform today. http://sdm.link/xeonphi
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Jon Elson]

On 01/10/2017 10:45 PM, Gene Heskett wrote:
> On Tuesday 10 January 2017 23:02:28 Jon Elson wrote:
>
>> On 01/10/2017 08:39 PM, Gene Heskett wrote:
>>> But if I hook up the halscope, the apparent quadrature error is
>>> worthless due to the 1KHz servo-loop timing limiting the bandwidth.
>> Well, that's why we use hardware encoder counters, as
>> sampling the quadrature at 1 KHz just isn't fast enough.
>>
>> Jon
> Plz elucidate Jon. Something has got to be better than this. See the pix
> I posted with the last reply to Peter & the list.  The top, white trace
> in the halscope screen is the quadrature noise, easily 10x the timing
> wibbles I see on the hitachi's screen.
>
>
If you are looking at the quadrature signals at a 1 KHz 
sampling rate, then unless the encoder is moving quite 
slowly, the sampling will be too coarse.  That screen shot 
is kind of small (you can take just a single window as the 
screenshot, by the way) but it looks like the duty cycle of 
the encoders is not 50%.  That may well be aliasing, as the 
quadrature signals seem to be moving much too close to 1000 
Hz.  You'd get much better results below 100 Hz.  Or, with 
Mesa hardware, you can sample these with the base thread.

I can't comment on the velocity output from the Mesa, as I 
don't know their gear (or driver).  But, the scale is 2/div, 
meaning the scaled velocity output is jumping around 2 user 
units up and down at a KHz rate.  Clearly, not good.

Jon

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Two questions.

[Gene Heskett]

On Wednesday 11 January 2017 00:29:01 Peter C. Wallace wrote:

> >> AFAIK there are no significant issues with HM2 encoder velocity
> >> estimation
> >>
> >> Heres a 100 Hz quadrature velocity trace:
> >>
> >> http://freeby.mesanet.com/100_Hz_quadrature_velocity.png
> >>
> >> Notice theres no discernible ripple in the velocity even though
> >> theres considerable sample jitter (green trace)
> >
> > Show me that same trace out of a 7i90 running on the spi bus driven
> > by the rpi 3b.
>
> I expect it would be the same since all hm2 encoder counters are the
> same regardless of interface, and AFAIK the encoder driver code above
> the data transfer level is the same for all interface methods.
>
> If you wish to determine if your error is driver/SPI hardware _or_
> simply bad quadrature you can use a hardware stepgen in velocity mode
> looped back to the encoder counter since it generates basically
> perfect quadrature.
>
I'll see if I can rig my hal file for that tomorrow (later today 
actually). Some sort of a set of mux2's so I can dynamically switch it 
back and forth.

The encoder is a triplet of ats667's watching a 6" diameter, 60 tooth 
gear, so thats 240 transitions a second*rps.  The data sheets say the 
normal duty cycle is between 48 to 49 % as a result of the internal gain 
controls that hold the switchpoint pretty constant in the face of and 
eccentricity in the target. I have dialed that gear, and its 
eccentricity is less than a thou. It doesn't say what effect any 
residual magnetism of the steel of the gear might do but I tried to 
neutralize that by bringing a mag tool base such that it alternated as 
the gear turned, close enough to effect that, then slowly removed the 
tool base so the mag field faded leaving the gear as demagnetized as it 
was going to get.  No effect that I could detect. The lack of a 50% 
symmetry in the wave form is unavoidable when using the 667, and when I 
remachined the mount, I tried to get good spacings, and my scope 
confirms at most a 2 thou error in a 90 degree quadrature.  Thats not 
much but I'll confirm that on the analog scope tomorrow before I start 
mucking around in the hal file.

Does the stepgen in that mode also generate an index?  And should I 
switch that too? I have to assume no, its not mentioned in the manpage. 
>From the manpage I gather I should set a base thread since a step change 
at 1 kilohertz would be /4, or 250 teeth a second /60 
teeth=4.167 rps. Thats about as slow as it will run. ;-)

I also think that latency in the generated waveforms would be a self 
compensated error, and invisible to this test.

It sounds like to do this right, I'd need a dual channel Arbitrary 
Waveform Generator.  The last time I priced one of those, its was about 
10x my paygrade. But 10 years later a dual channel DDS version that can 
run at 6MHz is now around $55 USD, but of course 3 weeks away, from 
China.

I have some hal code that will do a /4 on the noise, but it would also be 
2 to 4 counts behind at a G33.1 reversal. I'll need to establish about a 
10 thou backlash in the z, somewhat more than it measures.

> If this generates noisy velocity signals then the driver/SPI interface
> needs to be looked into.
>
> Peter Wallace
> Mesa Electronics
>
>
>
> --
> Developer Access Program for Intel Xeon Phi Processors
> Access to Intel Xeon Phi processor-based developer platforms.
> With one year of Intel Parallel Studio XE.
> Training and support from Colfax.
> Order your platform today. http://sdm.link/xeonphi
> ___
> 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 

--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users