Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Mark]

On 03/04/2018 08:22 PM, Chris Albertson wrote:

On Sun, Feb 18, 2018 at 2:10 PM, Gene Heskett  wrote:

Seem like you machine runs into some form of resonance at 7 to 20 ipm.
   I'd like to see a controllers that can be reconfigured in real time.
for low speeds you might use 1/8 steps them move to /14/ and full
steps at higher RPM.  It could be built.   But simply going to a
digital drive might be enough.


Gecko drives are designed to do just that.



Mark



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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[andy pugh]

On 5 March 2018 at 01:23, Chris Albertson  wrote:
> I'm not contributing to EMC development so I have no say in how to do
> it.  But I'd suggest a move to Github.

The code moved to Github some time ago. ( https://github.com/LinuxCNC )

Moving a mailing list is more tricky as people will have the existing
address in their address lists, and will have the list whitelisted
etc.

-- 
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, 1916

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Sunday 04 March 2018 22:58:09 John Kasunich wrote:

> On Sun, Mar 4, 2018, at 10:40 PM, Gene Heskett wrote:
> > On Sunday 04 March 2018 20:23:01 Chris Albertson wrote:
> > > I'm not contributing to EMC development so I have no say in how to
> > > do it.  But I'd suggest a move to Github.  Not just because they
> > > are stable but because Git offers a better way to work
> > >
> > > On Sun, Mar 4, 2018 at 10:36 AM, andy pugh  
wrote:
> > > > On 4 March 2018 at 18:00, Peter Blodow  wrote:
> > > >> This mail has been sent to sourceforge this morning and was
> > > >> returned without reason.
> > > >
> > > > Yes, Sourceforge is having problems.
> >
> > This, I'm afraid, is symptomatic of a site thats well on the way of
> > going broke. We should get used to the idea of doing it ourselves,
> > before the whole code base disappears. They have already lost the
> > Nitros9 code base and can't seem to find it, its an hg repo, and its
> > been months since I've been able to refresh my copy with an hg pull.
> >
> > Contingency plans are in order.
>
> Just to clarify a few things:
>
> The code base is absolutely not at risk.
>
> The LinuxCNC code was moved off of Sourceforge years ago.  It was on a
> privately hosted git server for a while, but was moved to github
> several months ago.
>
> The LinuxCNC website is also NOT on Sourceforge.  I'm not exactly sure
> where it is hosted now, but it hasn't been at SF for years.
>
> The only thing we use Sourceforge for now is the mailing lists.

That is very reassuring to hear John K.  Thanks for the clarification.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[John Kasunich]

On Sun, Mar 4, 2018, at 10:40 PM, Gene Heskett wrote:
> On Sunday 04 March 2018 20:23:01 Chris Albertson wrote:
> 
> > I'm not contributing to EMC development so I have no say in how to do
> > it.  But I'd suggest a move to Github.  Not just because they are
> > stable but because Git offers a better way to work
> >
> > On Sun, Mar 4, 2018 at 10:36 AM, andy pugh  wrote:
> > > On 4 March 2018 at 18:00, Peter Blodow  wrote:
> > >> This mail has been sent to sourceforge this morning and was
> > >> returned without reason.
> > >
> > > Yes, Sourceforge is having problems.
> > >
> This, I'm afraid, is symptomatic of a site thats well on the way of going 
> broke. We should get used to the idea of doing it ourselves, before the 
> whole code base disappears. They have already lost the Nitros9 code base 
> and can't seem to find it, its an hg repo, and its been months since 
> I've been able to refresh my copy with an hg pull.
> 
> Contingency plans are in order.
> 

Just to clarify a few things:

The code base is absolutely not at risk.

The LinuxCNC code was moved off of Sourceforge years ago.  It was on a 
privately hosted git server for a while, but was moved to github several months 
ago.

The LinuxCNC website is also NOT on Sourceforge.  I'm not exactly sure where it 
is hosted now, but it hasn't been at SF for years.

The only thing we use Sourceforge for now is the mailing lists.

-- 
  John Kasunich
  jmkasun...@fastmail.fm

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[MC Cason via Emc-users]

If you are considering GitHub, then how they fended off a massive DDOS 
attack may be of interest:

https://www.wired.com/story/github-ddos-memcached/

BTW, SF was fending off it's own DDOS attack during a migration:
https://meta.slashdot.org/story/18/03/04/0428237/slashdot-outage-update


On 03/04/2018 07:23 PM, Chris Albertson wrote:

I'm not contributing to EMC development so I have no say in how to do
it.  But I'd suggest a move to Github.  Not just because they are
stable but because Git offers a better way to work

On Sun, Mar 4, 2018 at 10:36 AM, andy pugh  wrote:

On 4 March 2018 at 18:00, Peter Blodow  wrote:

This mail has been sent to sourceforge this morning and was returned without
reason.

Yes, Sourceforge is having problems.


--
MC Cason
Eagle3D - github.com/mcason/Eagle3D
Lathe Electronic Edge Finder - raccoonelectronics.com


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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Sunday 04 March 2018 20:23:01 Chris Albertson wrote:

> I'm not contributing to EMC development so I have no say in how to do
> it.  But I'd suggest a move to Github.  Not just because they are
> stable but because Git offers a better way to work
>
> On Sun, Mar 4, 2018 at 10:36 AM, andy pugh  wrote:
> > On 4 March 2018 at 18:00, Peter Blodow  wrote:
> >> This mail has been sent to sourceforge this morning and was
> >> returned without reason.
> >
> > Yes, Sourceforge is having problems.
> >
This, I'm afraid, is symptomatic of a site thats well on the way of going 
broke. We should get used to the idea of doing it ourselves, before the 
whole code base disappears. They have already lost the Nitros9 code base 
and can't seem to find it, its an hg repo, and its been months since 
I've been able to refresh my copy with an hg pull.

Contingency plans are in order.

> > --
> > 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, 1916
> >
> > 
> >-- 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
> > 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Sunday 04 March 2018 20:22:47 Chris Albertson wrote:

> On Sun, Feb 18, 2018 at 2:10 PM, Gene Heskett  
wrote:
> > On Friday 16 February 2018 06:16:00 Chris Albertson wrote:
> >
> >
> > That I have figured out finally Chris, its moving close to 100 ipm
> > with a 32 to 42 pulley set in the z on my Sheldon but its noisy as
> > can be in the 7 to 20 ipm box. Its noise telegraphs up the keyboard
> > shelf mounted on its mounting bracket. I'd had to put a 1/4" high
> > fence around the edge to keep other stuff from vibrating off
> > including the keyboard itself. Its also running an a few more volts
> > than it was on the mill with its heavy head. 7 more volts IIRC. X is
> > slow, but if I ever get off my duff and make a thin vibration
> > damper, I think that would bring it up to 60 ipm. Its a nema 24, 8
> > wire, wired parallel and getting all a 2m542 can give it at around
> > 43 volts. Neither motor is best match, but what I had, or could get
> > in the crowded area behind the new apron.
>
> Seem like you machine runs into some form of resonance at 7 to 20 ipm.
>   I'd like to see a controllers that can be reconfigured in real time.

So would I, but that would also involve some magic at the transition 
point else it would slip at a fractional step without first placing the 
motor in its 0 phase position. And that would need to be done 
synchronously in lcnc at the same zero phase point.

> for low speeds you might use 1/8 steps them move to /14/ and full
> steps at higher RPM.  It could be built.   But simply going to a
> digital drive might be enough.
>
> I also have an M542.   One of the things I want to do is compare it
> with an A/B test to a DM542.  The newer one with e "D" is digital and
> the manufacturer claims a significant reduction in noise and
> vibration.   They cost about the same.   (I'm making a test jig for
> another project where I measure the strength of 3D printed plastic
> gears.  It will use the same motor size.)

Most, but not all, are 2M542's now a decade old. The only diff I've noted 
is the dip switch positions to get a common speed. /8 or /16 but /16 is 
generally too fast for the opto's.

Buying more as the need arises, I probably have at least 3 versions in 
service now. And once programmed, they Just Work(TM).

>
>   I also want to try one of their new "closed loop" motors which
> people on you tube have demoed.   Those are even better running but
> cost about 60% more.   I can do this because I need to buy more
> motors, quite a few more as the thing I'm using the mill and lathe and
> printer for also uses motors.
>
> I did some more calculations again, The static load of a milling head
> on one vertical dovetail means that power is required even when moving
> downward slowly.  In theory if there were no friction we should be
> able to generate power by letting the head fall down which would spin
> an electric motor  With a vertical acme screw more then half the motor
> power goes into overcoming friction.
>
> In a few months I expect to have real data on vertical ball screw
> performance and also onplatic gear strength and the three types of
> stepper motor drivers.  Maybe longer because client are calling be
> more now, work seems to be picking. Darn.



-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

On Sun, Feb 18, 2018 at 2:10 PM, Gene Heskett  wrote:
> On Friday 16 February 2018 06:16:00 Chris Albertson wrote:
>

> That I have figured out finally Chris, its moving close to 100 ipm with a
> 32 to 42 pulley set in the z on my Sheldon but its noisy as can be in
> the 7 to 20 ipm box. Its noise telegraphs up the keyboard shelf mounted
> on its mounting bracket. I'd had to put a 1/4" high fence around the
> edge to keep other stuff from vibrating off including the keyboard
> itself. Its also running an a few more volts than it was on the mill
> with its heavy head. 7 more volts IIRC. X is slow, but if I ever get off
> my duff and make a thin vibration damper, I think that would bring it up
> to 60 ipm. Its a nema 24, 8 wire, wired parallel and getting all a 2m542
> can give it at around 43 volts. Neither motor is best match, but what I
> had, or could get in the crowded area behind the new apron.

Seem like you machine runs into some form of resonance at 7 to 20 ipm.
  I'd like to see a controllers that can be reconfigured in real time.
for low speeds you might use 1/8 steps them move to /14/ and full
steps at higher RPM.  It could be built.   But simply going to a
digital drive might be enough.

I also have an M542.   One of the things I want to do is compare it
with an A/B test to a DM542.  The newer one with e "D" is digital and
the manufacturer claims a significant reduction in noise and
vibration.   They cost about the same.   (I'm making a test jig for
another project where I measure the strength of 3D printed plastic
gears.  It will use the same motor size.)


  I also want to try one of their new "closed loop" motors which
people on you tube have demoed.   Those are even better running but
cost about 60% more.   I can do this because I need to buy more
motors, quite a few more as the thing I'm using the mill and lathe and
printer for also uses motors.

I did some more calculations again, The static load of a milling head
on one vertical dovetail means that power is required even when moving
downward slowly.  In theory if there were no friction we should be
able to generate power by letting the head fall down which would spin
an electric motor  With a vertical acme screw more then half the motor
power goes into overcoming friction.

In a few months I expect to have real data on vertical ball screw
performance and also onplatic gear strength and the three types of
stepper motor drivers.  Maybe longer because client are calling be
more now, work seems to be picking. Darn.

-- 

Chris Albertson
Redondo Beach, California

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

I'm not contributing to EMC development so I have no say in how to do
it.  But I'd suggest a move to Github.  Not just because they are
stable but because Git offers a better way to work

On Sun, Mar 4, 2018 at 10:36 AM, andy pugh  wrote:
> On 4 March 2018 at 18:00, Peter Blodow  wrote:
>> This mail has been sent to sourceforge this morning and was returned without
>> reason.
>
> Yes, Sourceforge is having problems.
>
>
> --
> 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, 1916
>
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



-- 

Chris Albertson
Redondo Beach, California

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[andy pugh]

On 4 March 2018 at 18:00, Peter Blodow  wrote:
> This mail has been sent to sourceforge this morning and was returned without
> reason.

Yes, Sourceforge is having problems.


-- 
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, 1916

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Peter Blodow]

This mail has been sent to sourceforge this morning and was returned 
without reason. Therefore, I forward this to your own mail address. P.


Andy, better keep fingers off that sort of thread (Panzergewinde, 
Sta-Pa-Rohrgewinde). It was widely used, but only for unimportant 
purposes (like protecting cables dangling form the ceiling to a 
machine). The cutters had square outer outline and nobody had holders 
for them, so they used monkey wrenches and pliers to turn them! Besides, 
the norm says that the thread is cut only part way into the pipe wall so 
that they are flat outside, having no sharp tips and little holding 
power. The threads may also be hard to catch at mounting time, standing 
on a ladder below the ceiling. And the pipes were rolled and seam 
welded, leaving a sharp burr inside that might cut cable insulation. 
This whole stuff is only for dirty purpose and no good craftsmanship. I 
used up my last pipes to make handles for alu pouring equipment because 
they will fit plastic bicycle handles.


Peter


Am 20.02.2018 um 18:38 schrieb Andy Pugh:



On 17 Feb 2018, at 15:26, Peter Blodow  wrote:



Am 16.02.2018 um 06:32 schrieb Chris Albert

I think this thread size os used for those very thin rings used for
electrical  jacks and toggle switches.  The ring nuts are only 2mm thick so
they need an ultra fine pitch.

...which usually is 0.75

Another way is to use a very large thread angle.
This one is metric on diameter, 80 degree thread angle and TPI pitch.

https://en.m.wikipedia.org/wiki/Panzergewinde


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




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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Wednesday 28 February 2018 10:04:38 Gene Heskett wrote:

> On Wednesday 28 February 2018 07:03:42 andy pugh wrote:
> > On 28 February 2018 at 01:15, Gene Heskett 
>
> wrote:
> > > So I am supposed to get a fitbit, and let it yelp
> > > at me every so often. So I did, then discovered it only runs on
> > > winderz or Macs.  Its all packed up, never powered up, get my
> > > money back until its private, and runs on linux.
> >
> > https://linuxaria.com/article/how-to-sync-your-fitbit-under-linux
>
> Following those instructions, I get to the:
> gene@coyote:~/src/pyusb-master$ sudo pip install galileo
> Downloading/unpacking galileo
>   Cannot fetch index base URL http://pypi.python.org/simple/
>   Could not find any downloads that satisfy the requirement galileo
> No distributions at all found for galileo
> Storing complete log in /root/.pip/pip.log
>
> from the pip log in root:
> root@coyote:~/.pip# cat pip.log
> 
> /usr/bin/pip run on Wed Feb 28 09:57:11 2018
> Downloading/unpacking galileo
>
>   Getting page http://pypi.python.org/simple/galileo
>   Could not fetch URL http://pypi.python.org/simple/galileo: HTTP
> Error 403: SSL is required
>   Will skip URL http://pypi.python.org/simple/galileo when looking for
> download links for galileo
>   Getting page http://pypi.python.org/simple/
>   Could not fetch URL http://pypi.python.org/simple/: HTTP Error 403:
> SSL is required
>   Will skip URL http://pypi.python.org/simple/ when looking for
> download links for galileo
>   Cannot fetch index base URL http://pypi.python.org/simple/
>
> And from that it seems SSL is required. A considerable number of
> openssl files are installed. So whats next?

It turns out that galileo is not until stretch.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Wednesday 28 February 2018 07:03:42 andy pugh wrote:

> On 28 February 2018 at 01:15, Gene Heskett  
wrote:
> > So I am supposed to get a fitbit, and let it yelp
> > at me every so often. So I did, then discovered it only runs on
> > winderz or Macs.  Its all packed up, never powered up, get my money
> > back until its private, and runs on linux.
>
> https://linuxaria.com/article/how-to-sync-your-fitbit-under-linux

Following those instructions, I get to the:
gene@coyote:~/src/pyusb-master$ sudo pip install galileo
Downloading/unpacking galileo
  Cannot fetch index base URL http://pypi.python.org/simple/
  Could not find any downloads that satisfy the requirement galileo
No distributions at all found for galileo
Storing complete log in /root/.pip/pip.log

from the pip log in root:
root@coyote:~/.pip# cat pip.log

/usr/bin/pip run on Wed Feb 28 09:57:11 2018
Downloading/unpacking galileo

  Getting page http://pypi.python.org/simple/galileo
  Could not fetch URL http://pypi.python.org/simple/galileo: HTTP Error 
403: SSL is required
  Will skip URL http://pypi.python.org/simple/galileo when looking for 
download links for galileo
  Getting page http://pypi.python.org/simple/
  Could not fetch URL http://pypi.python.org/simple/: HTTP Error 403: SSL 
is required
  Will skip URL http://pypi.python.org/simple/ when looking for download 
links for galileo
  Cannot fetch index base URL http://pypi.python.org/simple/

And from that it seems SSL is required. A considerable number of openssl 
files are installed. So whats next?


-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[andy pugh]

On 28 February 2018 at 01:15, Gene Heskett  wrote:

> So I am supposed to get a fitbit, and let it yelp
> at me every so often. So I did, then discovered it only runs on winderz
> or Macs.  Its all packed up, never powered up, get my money back until
> its private, and runs on linux.

https://linuxaria.com/article/how-to-sync-your-fitbit-under-linux


-- 
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, 1916

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Andy Pugh]

> On 17 Feb 2018, at 15:26, Peter Blodow  wrote:
> 
> 
>> Am 16.02.2018 um 06:32 schrieb Chris Albert
>> 
>> I think this thread size os used for those very thin rings used for
>> electrical  jacks and toggle switches.  The ring nuts are only 2mm thick so
>> they need an ultra fine pitch.
> ...which usually is 0.75

Another way is to use a very large thread angle. 
This one is metric on diameter, 80 degree thread angle and TPI pitch. 

https://en.m.wikipedia.org/wiki/Panzergewinde


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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Friday 16 February 2018 06:16:00 Chris Albertson wrote:

> If you are bending a screw the speed does not matter it is the
> acceleration.  That is measured in meters per second squared (or
> whatever the Imperial equivalent is?  feet per second per second?)
>
> I think a typical low performance mill like most of us have might have
> go a 2m/s^2  A very high performance one might do 20m/s^2
>
> If the mill can do about 10m/S^2 then the force on the ball screw is
> equal to the weight of the table and whatever is bolted down to it.  
> Remember what Newton said "f=ma".
>
> Which happens first the ball nut breaks or the screw bends is
> determined by the length of the screw from motor to nut.
>
> Gene when you used the motor in a design where there was a very large
> static load, where the motor had to hold the weight of large vertical
> moving head  you'd expect poor movement compared to a horizontal axis.

That I have figured out finally Chris, its moving close to 100 ipm with a 
32 to 42 pulley set in the z on my Sheldon but its noisy as can be in 
the 7 to 20 ipm box. Its noise telegraphs up the keyboard shelf mounted 
on its mounting bracket. I'd had to put a 1/4" high fence around the 
edge to keep other stuff from vibrating off including the keyboard 
itself. Its also running an a few more volts than it was on the mill 
with its heavy head. 7 more volts IIRC. X is slow, but if I ever get off 
my duff and make a thin vibration damper, I think that would bring it up 
to 60 ipm. Its a nema 24, 8 wire, wired parallel and getting all a 2m542 
can give it at around 43 volts. Neither motor is best match, but what I 
had, or could get in the crowded area behind the new apron.

> On Fri, Feb 16, 2018 at 12:54 AM, Gene Heskett  
wrote:
> > On Friday 16 February 2018 02:20:40 jeremy youngs wrote:
> > > Chris, I think you are spot on . Nema 34 1600 oz in. And
> > > automation direct 110v drives.
> >
> > I have one of those motors. On the Z of my mill, with a 60 volt psu,
> > very poor performance. 27 ipm or stall lifting the head on the
> > G0704. Its now on the sheldons Z and while noisy, is also about 4x
> > faster.
> >
> > > I had y at 300 ipm 9×42 table. It was a very cheap
> > > ballscrew and I had it cobbled together now that the Machine is 3
> > > axis and spindle functional I will be able to make some belt and
> > > pulley mounts and resolve some issues.
> > > --
> > >  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
> > > 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 
> >
> > 
> > --
> > 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
> > 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Tuesday 20 February 2018 12:53:08 John Alexander Stewart wrote:

> Hi Gene;
>
> Regarding the DM1182 - Ok - I see what it is you are doing. Very
> interesting - I've been out of the loop for a year or two and getting
> back into it.
>
> John.

And I'm just now catching up, been in the shop for a few days while they 
filtered a gallon or so of heperin thru my kidneys. The sore knee turned 
into the whole leg, and the ultrasound yelped about a nice sized blood 
clot fairly close to the groin. Still sore, which I'm told will be a 
week or more getting back to normal.  And they blamed it on too long 
sitting in this chair! So I am supposed to get a fitbit, and let it yelp 
at me every so often. So I did, then discovered it only runs on winderz 
or Macs.  Its all packed up, never powered up, get my money back until 
its private, and runs on linux.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[John Alexander Stewart]

Hi Gene;

Regarding the DM1182 - Ok - I see what it is you are doing. Very
interesting - I've been out of the loop for a year or two and getting back
into it.

John.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gregg Eshelman via Emc-users]

Screw whip and bending can be mitigated by putting it under some tension, with 
angular contact ball or tapered roller bearings at the ends. 

On Sunday, February 18, 2018, 10:47:47 AM MST, Chris Albertson 
 wrote:  
 
 If you are bending a screw the speed does not matter it is the
acceleration.  That is measured in meters per second squared (or whatever
the Imperial equivalent is?  feet per second per second?)

I think a typical low performance mill like most of us have might have go a
2m/s^2  A very high performance one might do 20m/s^2

If the mill can do about 10m/S^2 then the force on the ball screw is equal
to the weight of the table and whatever is bolted down to it.  Remember
what Newton said "f=ma".

Which happens first the ball nut breaks or the screw bends is determined by
the length of the screw from motor to nut.

Gene when you used the motor in a design where there was a very large
static load, where the motor had to hold the weight of large vertical
moving head  you'd expect poor movement compared to a horizontal axis.  
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Tuesday 20 February 2018 08:25:03 John Alexander Stewart wrote:

> Regarding the g0704 z axis and the automation direct 120v system,
> could someone post a link to the product?
>
> All in this thread an interesting discussion, thanks - John.

I'm currently laid up with a game knee, so I can't go get the model 
number off it, but it ought to be google-able. Except the google links 
to AD aren't loading for some reason. But it finally did load, but AD 
doesn't make this, so it may be a leadshine product. Further googling 
has not located a decent pix of it. I paid around $175 for the one I 
have.

And I believe this is the culprit I bought:

line powered, up to 8.2 amps. Buy it now for $157.50 USD + $19.90 ship 
from China.
Good drive, zero problems.
> --
> 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
> 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[John Alexander Stewart]

Regarding the g0704 z axis and the automation direct 120v system, could
someone post a link to the product?

All in this thread an interesting discussion, thanks - John.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Friday 16 February 2018 11:19:09 jeremy youngs wrote:

> I have one of those motors. On the Z of my mill, with a 60 volt psu,
> very poor performance. 27 ipm or stall lifting the head on the G0704.
>
>
> With the 860 drive and 48 volts I could only get about 60 ipm . The
> automation direct ( kelling) drives at 120 volt really make the
> difference, far less motor heating as well.

Thats my experience too. Change the motor for an 8 wire 960inch, and I 
can now move that g0704's head at 140 ipm, either direction. And do it 
20 db quieter than the 860 drive could do it. That A. D. line powered 
drive is amazing. But it sticks out of the front of the box around 2", 
precluding ever putting the lid back on 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

I have one of those motors. On the Z of my mill, with a 60 volt psu, very
poor performance. 27 ipm or stall lifting the head on the G0704.


With the 860 drive and 48 volts I could only get about 60 ipm . The
automation direct ( kelling) drives at 120 volt really make the difference,
far less motor heating as well.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

Petter,

*After* you point it out, I remember and yes, 0.75 mm.



>> ...which usually is 0.75 mm.
>
> Peter
>


-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Peter Blodow]

Am 16.02.2018 um 06:32 schrieb Chris Albertson:
/snip/

/snip


I think this thread size os used for those very thin rings used for
electrical  jacks and toggle switches.  The ring nuts are only 2mm thick so
they need an ultra fine pitch.

...which usually is 0.75 mm.

Peter

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

If you are bending a screw the speed does not matter it is the
acceleration.  That is measured in meters per second squared (or whatever
the Imperial equivalent is?  feet per second per second?)

I think a typical low performance mill like most of us have might have go a
2m/s^2  A very high performance one might do 20m/s^2

If the mill can do about 10m/S^2 then the force on the ball screw is equal
to the weight of the table and whatever is bolted down to it.   Remember
what Newton said "f=ma".

Which happens first the ball nut breaks or the screw bends is determined by
the length of the screw from motor to nut.

Gene when you used the motor in a design where there was a very large
static load, where the motor had to hold the weight of large vertical
moving head  you'd expect poor movement compared to a horizontal axis.

On Fri, Feb 16, 2018 at 12:54 AM, Gene Heskett  wrote:

> On Friday 16 February 2018 02:20:40 jeremy youngs wrote:
>
> > Chris, I think you are spot on . Nema 34 1600 oz in. And automation
> > direct 110v drives.
>
> I have one of those motors. On the Z of my mill, with a 60 volt psu, very
> poor performance. 27 ipm or stall lifting the head on the G0704. Its now
> on the sheldons Z and while noisy, is also about 4x faster.
>
> > I had y at 300 ipm 9×42 table. It was a very cheap
> > ballscrew and I had it cobbled together now that the Machine is 3 axis
> > and spindle functional I will be able to make some belt and pulley
> > mounts and resolve some issues.
> > --
> > 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
> > 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 
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Friday 16 February 2018 02:20:40 jeremy youngs wrote:

> Chris, I think you are spot on . Nema 34 1600 oz in. And automation
> direct 110v drives.

I have one of those motors. On the Z of my mill, with a 60 volt psu, very 
poor performance. 27 ipm or stall lifting the head on the G0704. Its now 
on the sheldons Z and while noisy, is also about 4x faster.

> I had y at 300 ipm 9×42 table. It was a very cheap 
> ballscrew and I had it cobbled together now that the Machine is 3 axis
> and spindle functional I will be able to make some belt and pulley
> mounts and resolve some issues.
> --
> 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
> 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Friday 16 February 2018 00:32:32 Chris Albertson wrote:

> I ordered a sampling of the sensors from Chinese vendors on eBay.  (At
> $1.90 each, ordering a handful don't break the bank)   I measured the
> treads and 1mm pitch seems to be the standard.   I looked and found it
> is a metric standard but not very common.  I found a very low M12x1
> cost plug tap.
>
> I can print 1mm pitch threads but they don't come out very clean or
> sharp. The tap cleans them up really well.
>
> Actually it makes doing mental math easy.  I measure the taget to
> senior distance and if I want to move it say 0.25mm it is pretty easy
> to figure out I need to rotate the sensor 90 degrees.
>
> I think this thread size os used for those very thin rings used for
> electrical  jacks and toggle switches.  The ring nuts are only 2mm
> thick so they need an ultra fine pitch.
>
> With the sensor in a threaded holder I can screw it in until it makes
> contact with the target then back it off until the signal goes low
> then screw it back until it goes high.  It is self measuring at 1mm
> per turn.

That does certainly simplify the setup. Similar to my useing 50 tpi to 
make the plugs that bind the loose, not even a mount flange, nuts in the 
xy axis's of my small HF micromill. Somewhat difficult to get to because 
I have a limit switch to detect when its about to unscrew itself off the 
screw, and that precludes dropping a pin drive wrench over the screw to 
adjust the nuts.

50 tpi? Seemed like a good deal at the time. If I had to do it over I 
think I'd use 80 tpi as it would have given me a little more room.

> On Thu, Feb 15, 2018 at 8:11 PM, Gene Heskett  
wrote:
> > On Thursday 15 February 2018 22:23:34 Chris Albertson wrote:
> > > I'm slowing getting set up to test one of those sensors.  I just
> > > 3d printed a holder.   The darn thing uses M12x1 threads so I had
> > > to order a 12x1 tap 6 weeks from China and the tap arrived.
> >
> > 12mmx1mm?. Thats pretty fine, way finer than a 1/2 SAE. And close to
> > being made of un-obtainium I'd expect.
> >
> > > Printing
> > > 12x1 almost works but you need to chase out the printed threads.
> > >
> > > None of the local stores carry anything that is M12x1.Soon
> > > enough I will know if it can "see" things like hex bolt heads and
> > > aluminum foil.
> > >
> > > On Tue, Feb 13, 2018 at 5:04 PM, jeremy youngs
> > > 
> >
> > wrote:
> > > > I had to service this motor today, as a previous technician had
> > > > improperly installed the brushes and it would are off at speed.
> > > > In doing so I removed the tach cover and found plenty of room
> > > > for an omron encoder. So I will not likely try these for spindle
> > > > encoder. But will be putting them in as home switches soon.
> > > >
> > > > Gene, you are absolutely right about the optos, the strongest
> > > > and best these have performed today after I replaced the bob to
> > > > a straight through board. However, I discovered my y axis
> > > > ballscrews is bent. It's been in there 4 years now and was kinda
> > > > jury rigged anyhow. So I guess it's time to jump that hurdle,
> > > > may as well put one in x as well.
> > > > Now where's that box with pulleys and belt in it from China that
> > > > has been setting in Chicago for 2.5 weeks...
> > > > 
> > > > --
> > > > 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
> > > > 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 
> >
> > 
> > --
> > 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
> > 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Friday 16 February 2018 00:09:14 Chris Albertson wrote:

> On Thu, Feb 15, 2018 at 8:06 PM, Gene Heskett  
wrote:
> > On Tuesday 13 February 2018 20:04:27 jeremy youngs wrote:
> > > I had to service this motor today, as a previous technician had
> > > improperly installed the brushes and it would are off at speed. In
> > > doing so I removed the tach cover and found plenty of room for an
> > > omron encoder. So I will not likely try these for spindle encoder.
> > > But will be putting them in as home switches soon.
> > >
> > > Gene, you are absolutely right about the optos, the strongest and
> > > best these have performed today after I replaced the bob to a
> > > straight through board. However, I discovered my y axis ballscrews
> > > is bent.
> >
> > Ouch. Time to do it right. But how the heck did it get bent?  I'd
> > find out why, and fix it.
>
> If you have a massive machine and a powerful motor the motor can
> accelerate and produce a large force.
> The ball screw is no different from any other steel column and based
> on the ratio to length over diameter will buckle if enough force is
> applied. In other words nothing has to be broken to bend a ball screw.
>  OK the CNC controller has the maximum acceleration set to high.  To
> protect the ball screw the acceleration has to be limited based how
> far down the screw the nut is. From a mechanical engineering point
> of view a ball screw is simply an unsupported column.
>
> The force on the screw might be much more then the weight of the
> machine.

I'd submit that the screw was originally too small for the job. A Y screw 
in the context of a milling machine, usually has only the table and its 
mounted mass to deal with and the weight is handled by the ways. If it 
was under a half ton of knee, I could see it being stressed, but in that 
case I'd expect to see nut damage before it bent.

But pix of the machine I haven't seen, so all this is idle assumptions on 
my part as I may not have a good image in my mind.

Thanks Chris.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Marcus Bowman]

On 16 Feb 2018, at 04:11, Gene Heskett wrote:

> On Thursday 15 February 2018 22:23:34 Chris Albertson wrote:
> 
>> I'm slowing getting set up to test one of those sensors.  I just 3d
>> printed a holder.   The darn thing uses M12x1 threads so I had to
>> order a 12x1 tap 6 weeks from China and the tap arrived. 
> 
> 12mmx1mm?. Thats pretty fine, way finer than a 1/2 SAE. And close to 
> being made of un-obtainium I'd expect.

The x1 thread series is not infrequently used. Although M6 x 1 is probably the 
most common, M8, M10, M12 and M14 x 1 are not uncommon. M8 x 1 and M10 x 1 
bolts and nuts can both obtained without too much trouble, but not in your 
average ironmonger or DIY store.
M12 x 1 and M14 x 1 threads are used on some lathe accessories, although I 
think that may be to try to dissuade workshop owners from making matching parts 
instead of buying them. There is, for example, a set of commercially available 
small ball bearing tailstock centres with replaceable/interchangeable nose 
shapes, and that has an M14 x 1 thread. The quality of manufacture is not good, 
though, but the M14 x 1 probably dissuades most folks from making the nose 
pieces (or the whole thing) in the workshop.
The Unimat PC has an M14 x 1 mandrel nose thread, and the Unimat SL has an M12 
x 1 nose thread. Both have been widely sold for quite a long time, and there 
are lots of matching accessories like chucks. 
Personally, I use the x1 thread series quite often, especially M10 and M12 x 1 
which I use to get a fine thread for adjustable fittings.
I have a pair of Dormer E105 M12 x 1 geometric taps which are a joy to use. I 
bought them many years ago for a regular batch job, and they were very, very 
expensive because, at that time, we were still in the old imperial age and 
anything other than metric coarse was jolly difficult to get. This morning, I 
see M12 x 1 'standard' hand taps in the MSC catalogue at £28.89 + VAT each, 
which is way cheaper than they used to be. Maybe that's the influence of the 
Far Eastern low cost versions. Geometric taps in the x1 series are still 
difficult to get.
There's an M20 x 1 die in the drawer which I use for one job. Chinese 
manufacture, because I could not find another source anywhere.


Marcus


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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

Chris, I think you are spot on . Nema 34 1600 oz in. And automation direct
110v drives. I had y at 300 ipm 9×42 table. It was a very cheap ballscrew
and I had it cobbled together now that the Machine is 3 axis and spindle
functional I will be able to make some belt and pulley mounts and resolve
some issues.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

On Thu, Feb 15, 2018 at 8:06 PM, Gene Heskett  wrote:

> On Tuesday 13 February 2018 20:04:27 jeremy youngs wrote:
>
> > I had to service this motor today, as a previous technician had
> > improperly installed the brushes and it would are off at speed. In
> > doing so I removed the tach cover and found plenty of room for an
> > omron encoder. So I will not likely try these for spindle encoder. But
> > will be putting them in as home switches soon.
> >
> > Gene, you are absolutely right about the optos, the strongest and best
> > these have performed today after I replaced the bob to a straight
> > through board. However, I discovered my y axis ballscrews is bent.
>
> Ouch. Time to do it right. But how the heck did it get bent?  I'd find
> out why, and fix it.
>

If you have a massive machine and a powerful motor the motor can accelerate
and produce a large force.
The ball screw is no different from any other steel column and based on the
ratio to length over diameter will buckle if enough force is applied.
In other words nothing has to be broken to bend a ball screw.  OK the CNC
controller has the maximum acceleration set to high.  To protect the ball
screw the acceleration has to be limited based how far down the screw the
nut is. From a mechanical engineering point of view a ball screw is
simply an unsupported column.

The force on the screw might be much more then the weight of the machine.





-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

I ordered a sampling of the sensors from Chinese vendors on eBay.  (At
$1.90 each, ordering a handful don't break the bank)   I measured the
treads and 1mm pitch seems to be the standard.   I looked and found it is a
metric standard but not very common.  I found a very low M12x1 cost plug
tap.

I can print 1mm pitch threads but they don't come out very clean or sharp.
The tap cleans them up really well.

Actually it makes doing mental math easy.  I measure the taget to senior
distance and if I want to move it say 0.25mm it is pretty easy to figure
out I need to rotate the sensor 90 degrees.

I think this thread size os used for those very thin rings used for
electrical  jacks and toggle switches.  The ring nuts are only 2mm thick so
they need an ultra fine pitch.

With the sensor in a threaded holder I can screw it in until it makes
contact with the target then back it off until the signal goes low then
screw it back until it goes high.  It is self measuring at 1mm per turn.




On Thu, Feb 15, 2018 at 8:11 PM, Gene Heskett  wrote:

> On Thursday 15 February 2018 22:23:34 Chris Albertson wrote:
>
> > I'm slowing getting set up to test one of those sensors.  I just 3d
> > printed a holder.   The darn thing uses M12x1 threads so I had to
> > order a 12x1 tap 6 weeks from China and the tap arrived.
>
> 12mmx1mm?. Thats pretty fine, way finer than a 1/2 SAE. And close to
> being made of un-obtainium I'd expect.
> > Printing
> > 12x1 almost works but you need to chase out the printed threads.
> >
> > None of the local stores carry anything that is M12x1.Soon enough
> > I will know if it can "see" things like hex bolt heads and aluminum
> > foil.
> >
> > On Tue, Feb 13, 2018 at 5:04 PM, jeremy youngs 
> wrote:
> > > I had to service this motor today, as a previous technician had
> > > improperly installed the brushes and it would are off at speed. In
> > > doing so I removed the tach cover and found plenty of room for an
> > > omron encoder. So I will not likely try these for spindle encoder.
> > > But will be putting them in as home switches soon.
> > >
> > > Gene, you are absolutely right about the optos, the strongest and
> > > best these have performed today after I replaced the bob to a
> > > straight through board. However, I discovered my y axis ballscrews
> > > is bent. It's been in there 4 years now and was kinda jury rigged
> > > anyhow. So I guess it's time to jump that hurdle, may as well put
> > > one in x as well.
> > > Now where's that box with pulleys and belt in it from China that has
> > > been setting in Chicago for 2.5 weeks...
> > > 
> > > --
> > > 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
> > > 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 
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Thursday 15 February 2018 22:23:34 Chris Albertson wrote:

> I'm slowing getting set up to test one of those sensors.  I just 3d
> printed a holder.   The darn thing uses M12x1 threads so I had to
> order a 12x1 tap 6 weeks from China and the tap arrived. 

12mmx1mm?. Thats pretty fine, way finer than a 1/2 SAE. And close to 
being made of un-obtainium I'd expect.
> Printing 
> 12x1 almost works but you need to chase out the printed threads.
> 
> None of the local stores carry anything that is M12x1.Soon enough
> I will know if it can "see" things like hex bolt heads and aluminum
> foil.
>
> On Tue, Feb 13, 2018 at 5:04 PM, jeremy youngs  
wrote:
> > I had to service this motor today, as a previous technician had
> > improperly installed the brushes and it would are off at speed. In
> > doing so I removed the tach cover and found plenty of room for an
> > omron encoder. So I will not likely try these for spindle encoder.
> > But will be putting them in as home switches soon.
> >
> > Gene, you are absolutely right about the optos, the strongest and
> > best these have performed today after I replaced the bob to a
> > straight through board. However, I discovered my y axis ballscrews
> > is bent. It's been in there 4 years now and was kinda jury rigged
> > anyhow. So I guess it's time to jump that hurdle, may as well put
> > one in x as well.
> > Now where's that box with pulleys and belt in it from China that has
> > been setting in Chicago for 2.5 weeks...
> > 
> > --
> > 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
> > 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Tuesday 13 February 2018 20:04:27 jeremy youngs wrote:

> I had to service this motor today, as a previous technician had
> improperly installed the brushes and it would are off at speed. In
> doing so I removed the tach cover and found plenty of room for an
> omron encoder. So I will not likely try these for spindle encoder. But
> will be putting them in as home switches soon.
>
> Gene, you are absolutely right about the optos, the strongest and best
> these have performed today after I replaced the bob to a straight
> through board. However, I discovered my y axis ballscrews is bent.

Ouch. Time to do it right. But how the heck did it get bent?  I'd find 
out why, and fix it.

> It's been in there 4 years now and was kinda jury rigged anyhow. So I
> guess it's time to jump that hurdle, may as well put one in x as well.

> Now where's that box with pulleys and belt in it from China that has
> been setting in Chicago for 2.5 weeks...

It will probably arrive, in what the folks in Windy would call due 
time. :)

> --
> 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
> 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

I'm slowing getting set up to test one of those sensors.  I just 3d printed
a holder.   The darn thing uses M12x1 threads so I had to order a 12x1 tap
6 weeks from China and the tap arrived.  Printing 12x1 almost works but
you need to chase out the printed threads.   None of the local stores carry
anything that is M12x1.Soon enough I will know if it can "see" things
like hex bolt heads and aluminum foil.

On Tue, Feb 13, 2018 at 5:04 PM, jeremy youngs  wrote:

> I had to service this motor today, as a previous technician had improperly
> installed the brushes and it would are off at speed. In doing so I removed
> the tach cover and found plenty of room for an omron encoder. So I will not
> likely try these for spindle encoder. But will be putting them in as home
> switches soon.
>
> Gene, you are absolutely right about the optos, the strongest and best
> these have performed today after I replaced the bob to a straight through
> board. However, I discovered my y axis ballscrews is bent. It's been in
> there 4 years now and was kinda jury rigged anyhow. So I guess it's time to
> jump that hurdle, may as well put one in x as well.
> Now where's that box with pulleys and belt in it from China that has been
> setting in Chicago for 2.5 weeks...
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

I had to service this motor today, as a previous technician had improperly
installed the brushes and it would are off at speed. In doing so I removed
the tach cover and found plenty of room for an omron encoder. So I will not
likely try these for spindle encoder. But will be putting them in as home
switches soon.

Gene, you are absolutely right about the optos, the strongest and best
these have performed today after I replaced the bob to a straight through
board. However, I discovered my y axis ballscrews is bent. It's been in
there 4 years now and was kinda jury rigged anyhow. So I guess it's time to
jump that hurdle, may as well put one in x as well.
Now where's that box with pulleys and belt in it from China that has been
setting in Chicago for 2.5 weeks...
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[theman whosoldtheworld]

From what I know and from what I've seen and from what I've tried until
now, that kind of photosensors are the same ones used to make the classic
500ppr encoders  so that I would not have many problems about  In
addition, as I said the series E has a time of 5μs so that this dilemma
should become a good story.



2018-02-13 15:04 GMT+01:00 Gene Heskett :

> On Tuesday 13 February 2018 08:14:28 theman whosoldtheworld wrote:
>
> > PM series response time: Under light received condition: 20 μs or less
> > (as write on my first reply, first row) . not fast enought??
>
> I would say it is. But whats its recovery time if the part is moving
> away? IOW, if sensing gear teeth, how fast can it repeat? The ats-667's
> I used 3 of in the Sheldon seem plenty fast, but they are only looking
> at a 60 tooth gear thats hard put to get to 1600 revs, under 500 in 99%
> of the cases, so I'm am not asking them for much compared to the omron
> on the G0704's motor. The ats667 seems to be about ideal for a gear
> whose teeth are nominally 1/8" thick. How it would perform on a gear
> whose teeth were only 1/16th inch thick is TBD. Perhaps they make one
> for higher pitched gears too. IDK. IIRC the ats is around a 4
> microsecond response time, but it also costs 6x more, which makes this a
> more attractive solution.  Certainly worth playing with to see.
>
> > regards
> > bkt
> >
> > 2018-02-11 23:47 GMT+01:00 jeremy youngs :
> > > How fast are these? Could they be utilized as a spindle encoder?
> > > 
> > > --
> > > 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
> > > Emc-users@lists.sourceforge.net
> > > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> > --
> > 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
> > 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 
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Tuesday 13 February 2018 08:14:28 theman whosoldtheworld wrote:

> PM series response time: Under light received condition: 20 μs or less
> (as write on my first reply, first row) . not fast enought??

I would say it is. But whats its recovery time if the part is moving 
away? IOW, if sensing gear teeth, how fast can it repeat? The ats-667's 
I used 3 of in the Sheldon seem plenty fast, but they are only looking 
at a 60 tooth gear thats hard put to get to 1600 revs, under 500 in 99% 
of the cases, so I'm am not asking them for much compared to the omron 
on the G0704's motor. The ats667 seems to be about ideal for a gear 
whose teeth are nominally 1/8" thick. How it would perform on a gear 
whose teeth were only 1/16th inch thick is TBD. Perhaps they make one 
for higher pitched gears too. IDK. IIRC the ats is around a 4 
microsecond response time, but it also costs 6x more, which makes this a 
more attractive solution.  Certainly worth playing with to see.

> regards
> bkt
>
> 2018-02-11 23:47 GMT+01:00 jeremy youngs :
> > How fast are these? Could they be utilized as a spindle encoder?
> > 
> > --
> > 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
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> --
> 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
> 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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[theman whosoldtheworld]

PM series response time: Under light received condition: 20 μs or less (as
write on my first reply, first row) . not fast enought??

regards
bkt

2018-02-11 23:47 GMT+01:00 jeremy youngs :

> How fast are these? Could they be utilized as a spindle encoder?
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Monday 12 February 2018 00:12:21 jeremy youngs wrote:

> It's a brown boveri 2.5 kw servo motor with a 10 v /krpm tachometer.

That will function to tell you how fast its spinning alright, but there 
is no spindle position feedback available from such a device, so its not 
usable to cut threads where synchronous motion is needed.

The omron with its ABZ style pulses, can tell linuxcnc exactly where in 
its rotation it is, updating the info in my case as often as 350,000 
times a second or nearly 15000 times a spindle turn in low gear.

Thats way overkill but sure precise. Had I known the gear ratio between 
between the motor and spindle was as high as it was, I would have bought 
a 250 or 256 line/turn encoder, it would have been plenty. 1000 line was 
overkill, but the .hal file & a couple mux4's, and sum2's took care of 
that rather precisely. In my case I still use the Z (index) from my old 
setup that used opto-interrupters, and the AB quadrature signals come 
from the omron, passing thru a pair of rs485 gismo's, dollar each from 
fleabay to make good TTL level signals then that submitted to the opto 
stripped inputs of the bob. Such a lashup works well with a 5i25 
interface, and would likely work just as well with a 7i90HD if you 
needed the plethora of gpio (72 minus whatever the firmware loaded uses) 
the 7i90 has. I've added lots of "gingerbread" stuff on the bigger 
lathe, and still have around 22 uncommitted i/o's I could use to run a 
coffeepot, the lights, lube pumps, coolant/mist pumps, yadda yadda. I 
already have all the motor power controlled by the  motion enable button 
on linuxcnc, so its all off for the night with a poke at the f2 key. 
Having i/o to throw away? Priceless.

Where the 5i25 is protected from high voltages by the bob, the 7i90 is 
not, its basically its own bob, so figure on using a 7i42TA on each 50 
pin i/o the 7i90 has 3 of for protection against i/o noise that exceeds 
the 7i90's 3.3 volt ratings. The 7i42TA also gives you nice screw 
terminals to hook everything up to. I stacked mine above the 7i90 and 
already had a bunch of old scsi-ii cables that I cut down and 
reassembled for the 2 to 4" 50 pin cables that took.

> I'm running a selema 1224 servo amplifier , at 100 dc . I still have
> some room for more voltage but my transformer cabbaging skills aren't
> as solid as my ability to find drives and motors.

Old but solid state PA amplifiers in the kilowatt of audio range make 
good sources of such critters. And an old ex(now long retired like  me) 
employee has a nose for power toroids that I've taken advantage of on a 
couple occasions. My psu for TLM's spindle is the remains of a Phase 
Linear 750 watter, and I've 4 of a 350 watt toroid stacked in series and 
paralleled for the G0704's 90 volt 1 hp. So I'm banging it with around 
125 DC. Pico Systems PWM-Servo drivers are serving as the controllers 
for both. Just tell Jon you need the spindle motor version as he makes 2 
slightly different versions of it.

> I may well try one 
> of these for an index pulse, thanks for the heads up about low speed
> instability and the positive feedback on the omron. I need to figure
> out how to mount one of them in this contortionist box called mill
> head. Time to turn in , good night

A pix might get some suggestions.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

It's a brown boveri 2.5 kw servo motor with a 10 v /krpm tachometer. I'm
running a selema 1224 servo amplifier , at 100 dc . I still have some room
for more voltage but my transformer cabbaging skills aren't as solid as my
ability to find drives and motors. I may well try one of these for an index
pulse, thanks for the heads up about low speed instability and the positive
feedback on the omron. I need to figure out how to mount one of them in
this contortionist box called mill head. Time to turn in , good night
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Sunday 11 February 2018 20:12:12 jeremy youngs wrote:

> http://ebay.com/itm/LJ12A3-4-Z-BX-Inductive-Proximity-Sensor
> These were the winner in the opening comment. They have a five pack
> for less than$20 which would leave 2 left over . Their mounting would
> work well with a tone wheel on my spindle . My
> Drive runs a tach and this would be a good solution for spindle index
> as I have no great place for an encoder.

I didn't either, so I drilled and tapped the rear of the motors shaft, 
which was sitting flush with the top of the rear bearing, and made a 5mm 
extension about 10mm long that fit the omron encoder's 5mm coupling. 
Made a square hole in the top cover and mounted a 1/8" alu plate about 
an inch above the top of the cover to mount the encoder itself on. 
Sounds rather like apple tree mechanics, but its working quite well.

2 of those, if they are fast enough would work well with a tone wheel, 
but the low speed performance would suck. The problem there would be the 
low tooth count of the tone wheel.

My combo of using the old interrupter index against an encoder on the 
motor is that the scale count per spindle rev is quite high, over 7000 
in high gear, and over 14000 in low gear, but some tally switches on the 
shift knob and some hal magic fixed that right up. That speed 
overpowered the opto's in the bob, so they've been excised. The encoder 
didn't say, but when it arrived it was differential outputs, so some 
dollar a copy rs485 gismo's were used as receivers.  Works fine to the 
spindles max of 1500 or 3000 revs which is a hair over 350 kilohertz at 
the 5i25 inputs.

I think they'll work fine for an index, but I'd have reservations about 
using them for the encoder A/B generators. As for the drive (I'm 
assuming a vfd) having a tach, its probably derived from the frequency 
being sent to the motor, meaning its going to be a few percent 
optimistic with a load on the motor. They always have a slip angle, its 
what makes the motors work, so a 4 pole motor at rated load, running on 
60 hz, will actually be turning between 1725 and 1750 as opposed to a 
fully sync motor which would be doing the full 1800 at light load. I'd 
much rather have proof of the spindles absolute position everytime an 
edge of the tone wheel goes by.

This is working so much better that the 67 slot opto disk in the g0704, 
that given enough time before I miss morning roll call, I will do the 
same for the teeny motor on my HF micromill. Except that's a good excuse 
to put the bigger 400 watt motor I took out of TLM when it grew a 1 
horse drive parts breaker.

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

http://ebay.com/itm/LJ12A3-4-Z-BX-Inductive-Proximity-Sensor
These were the winner in the opening comment. They have a five pack for
less than$20 which would leave 2 left over . Their mounting would work well
with a tone wheel on my spindle . My
Drive runs a tach and this would be a good solution for spindle index as I
have no great place for an encoder.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Gene Heskett]

On Sunday 11 February 2018 17:47:07 jeremy youngs wrote:

> How fast are these? Could they be utilized as a spindle encoder?

How fast are what? Need a link Jeremy...

-- 
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 

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[jeremy youngs]

How fast are these? Could they be utilized as a spindle encoder?
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[theman whosoldtheworld]

panasonic PM-25/45 ... for better result EZ series ...

PM series response time: Under light received condition: 20 μs or less
Under light interrupted condition: 80 μs or less
(Maximum response frequency: 3 kHz)(Note)
(Note) : The response frequency is the value when the disc, given in the
figure below, is rotated.
Sensing range 6 mm 0.236 in (fixed)
Minimum sensing object 0.8 × 1.2 mm 0.031 × 0.047 in opaque object
Hysteresis 0.05 mm 0.002 in or less
Repeatability 0.01 mm 0.0004 in or less
Supply voltage 5 to 24 V DC ±10 % Ripple P-P 10 % or less
Current consumption 15 mA or less
Output NPN open-collector transistor
・Maximum sink current: 50 mA
・Applied voltage: 30 V DC or less (between output and 0 V)
・Residual voltage: 2 V or less (at 50 mA sink current), 1 V or less (at 16
mA sink current)
Output:Output operation Incorporated with 2 outputs: Light-ON / Dark-ON
Output:Short-circuit protection Incorporated

omrom has the same.

There is version with shielded PUR cable, pvc cable and connector
automotive type (not reccomended, because the vibration produce some spike
into connector that broke the sensor in 1-2 year. Obviusly in me experience)



2018-01-26 1:33 GMT+01:00 John Dammeyer <jo...@autoartisans.com>:

> Part numbers?
>
> > -Original Message-
> > From: theman whosoldtheworld [mailto:bleachk...@gmail.com]
> > Sent: January-25-18 3:17 PM
> > To: Enhanced Machine Controller (EMC)
> > Subject: Re: [Emc-users] non-contact position sensors, surprisingly good
> > performance.
> >
> > normally i use omron or panasonic micro photocell ... 0.05mm precision of
> > positioning, hight performances of repetitions, hight vibration
> resistance,
> > low cos, and possibility of shielded chain type cable, or connectors
> > automotive 4 pin type. Not 2USD, but 19USD for 3meters cable type, or
> 9USD
> > for connectors type.
> >
> > bkt
> >
> > 2018-01-25 21:20 GMT+01:00 Ken Strauss <ken.stra...@gmail.com>:
> >
> > >
> > >
> > > > -Original Message-
> > > > From: Nicklas Karlsson [mailto:nicklas.karlsso...@gmail.com]
> > > > Sent: Thursday, January 25, 2018 2:55 PM
> > > > To: Enhanced Machine Controller (EMC)
> > > > Subject: Re: [Emc-users] non-contact position sensors, surprisingly
> good
> > > > performance.
> > > >
> > > >
> > > > Sounds promising, I do not want to compete with low salary.
> > > >
> > > Unfortunately, when the competition has technology, economies of scale
> > and
> > > low salaries there are few options other than lower salaries.
> > >
> > >
> > >
> > > 
> > > --
> > > 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
> > > Emc-users@lists.sourceforge.net
> > > https://lists.sourceforge.net/lists/listinfo/emc-users
> > >
> >
> 
> 
> --
> > 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
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[John Dammeyer]

Part numbers?

> -Original Message-
> From: theman whosoldtheworld [mailto:bleachk...@gmail.com]
> Sent: January-25-18 3:17 PM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] non-contact position sensors, surprisingly good
> performance.
> 
> normally i use omron or panasonic micro photocell ... 0.05mm precision of
> positioning, hight performances of repetitions, hight vibration
resistance,
> low cos, and possibility of shielded chain type cable, or connectors
> automotive 4 pin type. Not 2USD, but 19USD for 3meters cable type, or 9USD
> for connectors type.
> 
> bkt
> 
> 2018-01-25 21:20 GMT+01:00 Ken Strauss <ken.stra...@gmail.com>:
> 
> >
> >
> > > -Original Message-
> > > From: Nicklas Karlsson [mailto:nicklas.karlsso...@gmail.com]
> > > Sent: Thursday, January 25, 2018 2:55 PM
> > > To: Enhanced Machine Controller (EMC)
> > > Subject: Re: [Emc-users] non-contact position sensors, surprisingly
good
> > > performance.
> > >
> > >
> > > Sounds promising, I do not want to compete with low salary.
> > >
> > Unfortunately, when the competition has technology, economies of scale
> and
> > low salaries there are few options other than lower salaries.
> >
> >
> >
> > 
> > --
> > 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
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
>

--
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[theman whosoldtheworld]

normally i use omron or panasonic micro photocell ... 0.05mm precision of
positioning, hight performances of repetitions, hight vibration resistance,
low cos, and possibility of shielded chain type cable, or connectors
automotive 4 pin type. Not 2USD, but 19USD for 3meters cable type, or 9USD
for connectors type.

bkt

2018-01-25 21:20 GMT+01:00 Ken Strauss <ken.stra...@gmail.com>:

>
>
> > -Original Message-
> > From: Nicklas Karlsson [mailto:nicklas.karlsso...@gmail.com]
> > Sent: Thursday, January 25, 2018 2:55 PM
> > To: Enhanced Machine Controller (EMC)
> > Subject: Re: [Emc-users] non-contact position sensors, surprisingly good
> > performance.
> >
> >
> > Sounds promising, I do not want to compete with low salary.
> >
> Unfortunately, when the competition has technology, economies of scale and
> low salaries there are few options other than lower salaries.
>
>
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Ken Strauss]

> -Original Message-
> From: Nicklas Karlsson [mailto:nicklas.karlsso...@gmail.com]
> Sent: Thursday, January 25, 2018 2:55 PM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] non-contact position sensors, surprisingly good
> performance.
>
>
> Sounds promising, I do not want to compete with low salary.
>
Unfortunately, when the competition has technology, economies of scale and
low salaries there are few options other than lower salaries.



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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Nicklas Karlsson]

On Wed, 24 Jan 2018 23:18:05 -0500
Dave Cole  wrote:

> On 1/24/2018 7:38 PM, Chris Albertson wrote:
> > On Wed, Jan 24, 2018 at 4:06 PM, Dave Cole  wrote:
> >
> >> That is likely an unshielded prox switch.  That means that the sensing
> >> pattern is  a blob that sits right at the base of the plastic tip, not
> >> above it.
> >> That means that they can also sense to the side of that plastic tip as
> >> well.   You need to be careful that you don't have any metal next to the
> >> plastic tip.
> >>
> >> The prox switches which do not have a plasic tip, but where the plastic
> >> tip is within the sides of the metal sensor tube are considered shielded
> >> prox switches.   They only sense above the tip and not to the side.  Their
> >> sense pattern looks like a short flame that comes out of the tube.
> >>
> >> I tend not to use unshielded proxes as they can trigger sometimes when you
> >> don't expect it due to brackets nearby etc.
> >>
> >> Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!
> >>
> > When these show up I will test there pattern my intentionally misaligning
> > them.   But if used for a machine end-stop the target can be pretty well
> > controlled.
> >
> > That is $2 with FREE shipping from China.   It is pretty much the standard
> > price, not a special deal.
> >
> > Factory workers in China make about $3.50 per hour so $2 retail allows for
> > maybe 5 minutes of labor per unit.  Reasonable if the factory is automated.
> >
> > I was watching a video id Apple MacBook cases being milled from billet.
> > The Mac has just one structural part, the unibody case itself.  These are
> > made literally by a millions from one foot diameter aluminum "logs" that
> > are  the size of utilty poles.   VERY little human labor is required, the
> > "logs" are pressed into plates, cut and milled by a special purpose
> > machine.So even with labor at $3.50/hour they don't use much labor.
> >
> 
> A lot of people don't understand what is going on in China.  They have a 
> lot of automated production plants there and they are building more.   I 
> was there this last spring installing 1 of 3 robotic cells that were 
> being installed at about the same time. The plant is an automotive 
> supplier for a nearby car plant.   The plant has numerous robots.  Very 
> few people work at the plant, and it runs 24x7.    What more could you 
> want as a manufacturer;  cheap labor, a growing skilled labor force, 
> automation, and low electric rates.

Sounds promising, I do not want to compete with low salary.

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Nicklas Karlsson]

On Wed, 24 Jan 2018 19:06:26 -0500
Dave Cole  wrote:

> That is likely an unshielded prox switch.  That means that the sensing 
> pattern is  a blob that sits right at the base of the plastic tip, not 
> above it.
> That means that they can also sense to the side of that plastic tip as 
> well.   You need to be careful that you don't have any metal next to the 
> plastic tip.
> 
> The prox switches which do not have a plasic tip, but where the plastic 
> tip is within the sides of the metal sensor tube are considered shielded 
> prox switches.   They only sense above the tip and not to the side.  
> Their sense pattern looks like a short flame that comes out of the tube.
> 
> I tend not to use unshielded proxes as they can trigger sometimes when 
> you don't expect it due to brackets nearby etc.
> 
> Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!
> 
> Dave

As an electronic engineer it is really hard to find somewhere to earn money.

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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[andy pugh]

On 25 January 2018 at 02:06, Dave Cole  wrote:

>
> The prox switches which do not have a plasic tip, but where the plastic
> tip is within the sides of the metal sensor tube are considered shielded
> prox switches.


 I use these on all my machines. (typically combined with index homing for
the best possible repeatability).
They have the advantage that they can be buried inside the machine out of
harm's way. My target is normally a drilled hole in an non-bearing part of
the slide. Sometimes filled with a plastic plug.
I thought I had a better picture than this, but you can see the hole that
the X-axis one lives in to the left of the ball-nut block in this picture:
https://photos.app.goo.gl/zZSFMootmB2rN8GB2
And this picture shows the target hole on the actual lathe slide (for the Z
axis)
https://photos.app.goo.gl/Bbxfx8WoIZiIjc8l2
In the pic the plug is aluminium. That didn't work It is now Delrin.

Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!


They partly save the money on the cables. I have had many cable failures
and no sensor failures. I would suggest replacing the wires as close as
possible to the sensors with something like:

https://www.rapidonline.com/lappkabel-1027752-lflex-chain-808-cp-grey-drag-chain-cable-3-x-0-5mm-63-4350

Note that you are then spending much more on the wire than the sensor :-)
(There are cheaper 3 x 0.5mm cables in the same section of that web site)

-- 
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, 1916
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Dave Cole]

On 1/24/2018 7:38 PM, Chris Albertson wrote:

On Wed, Jan 24, 2018 at 4:06 PM, Dave Cole  wrote:


That is likely an unshielded prox switch.  That means that the sensing
pattern is  a blob that sits right at the base of the plastic tip, not
above it.
That means that they can also sense to the side of that plastic tip as
well.   You need to be careful that you don't have any metal next to the
plastic tip.

The prox switches which do not have a plasic tip, but where the plastic
tip is within the sides of the metal sensor tube are considered shielded
prox switches.   They only sense above the tip and not to the side.  Their
sense pattern looks like a short flame that comes out of the tube.

I tend not to use unshielded proxes as they can trigger sometimes when you
don't expect it due to brackets nearby etc.

Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!


When these show up I will test there pattern my intentionally misaligning
them.   But if used for a machine end-stop the target can be pretty well
controlled.

That is $2 with FREE shipping from China.   It is pretty much the standard
price, not a special deal.

Factory workers in China make about $3.50 per hour so $2 retail allows for
maybe 5 minutes of labor per unit.  Reasonable if the factory is automated.

I was watching a video id Apple MacBook cases being milled from billet.
The Mac has just one structural part, the unibody case itself.  These are
made literally by a millions from one foot diameter aluminum "logs" that
are  the size of utilty poles.   VERY little human labor is required, the
"logs" are pressed into plates, cut and milled by a special purpose
machine.So even with labor at $3.50/hour they don't use much labor.



A lot of people don't understand what is going on in China.  They have a 
lot of automated production plants there and they are building more.   I 
was there this last spring installing 1 of 3 robotic cells that were 
being installed at about the same time. The plant is an automotive 
supplier for a nearby car plant.   The plant has numerous robots.  Very 
few people work at the plant, and it runs 24x7.    What more could you 
want as a manufacturer;  cheap labor, a growing skilled labor force, 
automation, and low electric rates.



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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

On Wed, Jan 24, 2018 at 4:06 PM, Dave Cole  wrote:

> That is likely an unshielded prox switch.  That means that the sensing
> pattern is  a blob that sits right at the base of the plastic tip, not
> above it.
> That means that they can also sense to the side of that plastic tip as
> well.   You need to be careful that you don't have any metal next to the
> plastic tip.
>
> The prox switches which do not have a plasic tip, but where the plastic
> tip is within the sides of the metal sensor tube are considered shielded
> prox switches.   They only sense above the tip and not to the side.  Their
> sense pattern looks like a short flame that comes out of the tube.
>
> I tend not to use unshielded proxes as they can trigger sometimes when you
> don't expect it due to brackets nearby etc.
>
> Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!
>

When these show up I will test there pattern my intentionally misaligning
them.   But if used for a machine end-stop the target can be pretty well
controlled.

That is $2 with FREE shipping from China.   It is pretty much the standard
price, not a special deal.

Factory workers in China make about $3.50 per hour so $2 retail allows for
maybe 5 minutes of labor per unit.  Reasonable if the factory is automated.

I was watching a video id Apple MacBook cases being milled from billet.
The Mac has just one structural part, the unibody case itself.  These are
made literally by a millions from one foot diameter aluminum "logs" that
are  the size of utilty poles.   VERY little human labor is required, the
"logs" are pressed into plates, cut and milled by a special purpose
machine.So even with labor at $3.50/hour they don't use much labor.


I



>
> Dave
>
> On 1/24/2018 3:18 PM, Chris Albertson wrote:
>
>> An interesting you tube video by "Tom's 3D".  He talks about 3D printers
>> but much of the technology applies to machine tools.  After all they both
>> run G-code.
>>
>> In this video he measures the accuracy of end-stop or "home" sensors and
>> compare microswitches, with and without levers and also varies kinds of
>> non-contact sensors.He runs each kind into a stop at both high and low
>> speed and after 100 tries has data for standard deviations.
>>
>> Result is that removing the lever from the microswitch reduces standard
>> deviation by about 3X.   I assume because the per has about a 3X
>> mechanical
>> advantage.   But it's surprising is that a $2 inductive sensor beats even
>> the non-lever switch.
>>
>> It might be moot as all the measurements are in microns (micro meters)
>>
>> But on the other hand if these dirt-cheap inductive sensors are this good
>> and importantly they are water and oil tight with no moving parts we
>> should
>> be using them.
>>
>> Here is the winner at 0.27 microns standard deviation
>> ebay.com/itm/LJ12A3-4-Z-BX-Inductive-Proximity-Sensor
>> > Sensor-Switch-NPN-DC6-36V-Great/272907213327?hash=
>> item3f8a899a0f:g:6fwAAOSw~gRVyHZ->
>>
>> The blue end looks like a button switch but it isn't.  This is a
>> non-contact sensor.  It triggers when it senses a metal part at 4mm
>> distance (with a surprising 0.27 micron standard deviation)
>>
>> Here is the video, The application is different but look at his test
>> setup.  It pretty much mimics what you would do on a milling machine.
>> https://youtu.be/il9bNWn66BY
>>
>> How to use them:  There are three wires two are power, you apply between 6
>> and 36 volts, typically 12 volts.   The third wire is either normally 12V
>> or zero and then switches state when the sensor is tripped.   Of course 12
>> volts would fry your computer so people use a voltage divider of optical
>> isolator or even a switching diode.
>>
>> The typical use for these is industrial automation, food processing,
>> manufacturing or whatever.
>>
>> I have ordered a few and will experiment.
>>
>
> 
> --
> 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
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[Dave Cole]

That is likely an unshielded prox switch.  That means that the sensing 
pattern is  a blob that sits right at the base of the plastic tip, not 
above it.
That means that they can also sense to the side of that plastic tip as 
well.   You need to be careful that you don't have any metal next to the 
plastic tip.


The prox switches which do not have a plasic tip, but where the plastic 
tip is within the sides of the metal sensor tube are considered shielded 
prox switches.   They only sense above the tip and not to the side.  
Their sense pattern looks like a short flame that comes out of the tube.


I tend not to use unshielded proxes as they can trigger sometimes when 
you don't expect it due to brackets nearby etc.


Its amazing that they can sell those for just over $2 bucks.. Crazy cheap!

Dave

On 1/24/2018 3:18 PM, Chris Albertson wrote:

An interesting you tube video by "Tom's 3D".  He talks about 3D printers
but much of the technology applies to machine tools.  After all they both
run G-code.

In this video he measures the accuracy of end-stop or "home" sensors and
compare microswitches, with and without levers and also varies kinds of
non-contact sensors.He runs each kind into a stop at both high and low
speed and after 100 tries has data for standard deviations.

Result is that removing the lever from the microswitch reduces standard
deviation by about 3X.   I assume because the per has about a 3X mechanical
advantage.   But it's surprising is that a $2 inductive sensor beats even
the non-lever switch.

It might be moot as all the measurements are in microns (micro meters)

But on the other hand if these dirt-cheap inductive sensors are this good
and importantly they are water and oil tight with no moving parts we should
be using them.

Here is the winner at 0.27 microns standard deviation
ebay.com/itm/LJ12A3-4-Z-BX-Inductive-Proximity-Sensor


The blue end looks like a button switch but it isn't.  This is a
non-contact sensor.  It triggers when it senses a metal part at 4mm
distance (with a surprising 0.27 micron standard deviation)

Here is the video, The application is different but look at his test
setup.  It pretty much mimics what you would do on a milling machine.
https://youtu.be/il9bNWn66BY

How to use them:  There are three wires two are power, you apply between 6
and 36 volts, typically 12 volts.   The third wire is either normally 12V
or zero and then switches state when the sensor is tripped.   Of course 12
volts would fry your computer so people use a voltage divider of optical
isolator or even a switching diode.

The typical use for these is industrial automation, food processing,
manufacturing or whatever.

I have ordered a few and will experiment.


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

Re: [Emc-users] non-contact position sensors, surprisingly good performance.

[John Alexander Stewart]

Interesting - thanks, and tell us how you get on with them.

John.
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] non-contact position sensors, surprisingly good performance.

[Chris Albertson]

An interesting you tube video by "Tom's 3D".  He talks about 3D printers
but much of the technology applies to machine tools.  After all they both
run G-code.

In this video he measures the accuracy of end-stop or "home" sensors and
compare microswitches, with and without levers and also varies kinds of
non-contact sensors.He runs each kind into a stop at both high and low
speed and after 100 tries has data for standard deviations.

Result is that removing the lever from the microswitch reduces standard
deviation by about 3X.   I assume because the per has about a 3X mechanical
advantage.   But it's surprising is that a $2 inductive sensor beats even
the non-lever switch.

It might be moot as all the measurements are in microns (micro meters)

But on the other hand if these dirt-cheap inductive sensors are this good
and importantly they are water and oil tight with no moving parts we should
be using them.

Here is the winner at 0.27 microns standard deviation
ebay.com/itm/LJ12A3-4-Z-BX-Inductive-Proximity-Sensor


The blue end looks like a button switch but it isn't.  This is a
non-contact sensor.  It triggers when it senses a metal part at 4mm
distance (with a surprising 0.27 micron standard deviation)

Here is the video, The application is different but look at his test
setup.  It pretty much mimics what you would do on a milling machine.
https://youtu.be/il9bNWn66BY

How to use them:  There are three wires two are power, you apply between 6
and 36 volts, typically 12 volts.   The third wire is either normally 12V
or zero and then switches state when the sensor is tripped.   Of course 12
volts would fry your computer so people use a voltage divider of optical
isolator or even a switching diode.

The typical use for these is industrial automation, food processing,
manufacturing or whatever.

I have ordered a few and will experiment.
-- 

Chris Albertson
Redondo Beach, California
--
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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users