Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Gene Heskett]

On Wednesday 09 March 2016 22:30:19 Chris Albertson wrote:

> I read the data sheet and found this quote:
>
> " If the direction of rotation is reversed, so that the gear rotates
> from the pin 4 side to the pin 1 side, then the output polarity
> inverts."
>
> Thanks for pointing out this device.  It is NOT a simple Hall effect
> sensor.  There are two hall effect sensors, the outputs of each are
> digitized and process with digital logic.   There is quite a lot going
> on inside the device to make it easy to use.  It self calibrates and
> works perfects down to zero RPM.   Looks like it would be very easy to
> install on any lather of mill that had a metal gear on it as you don't
> even have to place it accurately, it will self-adjust the gain based
> on the distance from the gear.
>
> So the answer is the observer behavior is because the device is
> designed to behave that way.

This sounds like an interesting device. Are you at liberty to give a 
range of pricing for one gear and sensor combo?  Or possibly just the 
sensor if the gear being ferrous is sufficient to make it work. I could 
possibly set the sensor on the quills driving gear down inside the head 
casting, which would free up the space now used by my optical encoder on 
this G0704, which in turn would allow it to be converted to a much 
quieter belt drive.

How would an index pulse be obtained?  Third sensor watching a drilled 
hole go by?

Thanks.
>
>
>
>
> On Wed, Mar 9, 2016 at 10:30 AM, Kirk Wallace
>
>  wrote:
> > On 03/09/2016 09:38 AM, andy pugh wrote:
> >> On 9 March 2016 at 16:54,   wrote:
> >>> The issue is that we drilled about a .130 hole for the index but
> >>> it acts odd.  If you rotate the spindle one direction it will pass
> >>> the index and stay high until it passes the hole then you get a
> >>> little index pulse low.  If you run it the other direction it goes
> >>> low after passing the index - and you get a positive index pulse
> >>> at every passing of the hole.
> >>
> >> This may be the internal logic at work. The sensors are rated to
> >> zero speed, and I think have an internal offsetting circuit.
> >> However I can't see how it can tell which way the gear is turning
> >> unless it has two sensors.
> >> Some automotive sensors do actively sense direction, and send a
> >> different pulse length in one direction compared to the other to
> >> encode the direction on a single wire.
> >
> > A very quick look at the datasheet seems to indicate that there are
> > two sensors about 2mm apart and that logic may use these to sense
> > whether the gear edge is metal to air or air to metal. I tend to
> > agree with Andy that the chip logic may be the problem. That and six
> > bits may get you a cuppa.
> >
> >
> > --
> > Kirk Wallace
> > http://www.wallacecompany.com/machine_shop/
> > http://www.wallacecompany.com/E45/
> >
> > 
> >-- Transform Data into Opportunity.
> > Accelerate data analysis in your applications with
> > Intel Data Analytics Acceleration Library.
> > Click to learn more.
> > http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> > ___
> > 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 

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Chris Albertson]

I read the data sheet and found this quote:

" If the direction of rotation is reversed, so that the gear rotates
from the pin 4 side to the pin 1 side, then the output polarity
inverts."

Thanks for pointing out this device.  It is NOT a simple Hall effect
sensor.  There are two hall effect sensors, the outputs of each are
digitized and process with digital logic.   There is quite a lot going
on inside the device to make it easy to use.  It self calibrates and
works perfects down to zero RPM.   Looks like it would be very easy to
install on any lather of mill that had a metal gear on it as you don't
even have to place it accurately, it will self-adjust the gain based
on the distance from the gear.

So the answer is the observer behavior is because the device is
designed to behave that way.







On Wed, Mar 9, 2016 at 10:30 AM, Kirk Wallace
 wrote:
> On 03/09/2016 09:38 AM, andy pugh wrote:
>> On 9 March 2016 at 16:54,   wrote:
>>> The issue is that we drilled about a .130 hole for the index but it acts 
>>> odd.  If you rotate the spindle one direction it will pass the index and 
>>> stay high until it passes the hole then you get a little index pulse low.  
>>> If you run it the other direction it goes low after passing the index - and 
>>> you get a positive index pulse at every passing of the hole.
>>
>> This may be the internal logic at work. The sensors are rated to zero
>> speed, and I think have an internal offsetting circuit. However I
>> can't see how it can tell which way the gear is turning unless it has
>> two sensors.
>> Some automotive sensors do actively sense direction, and send a
>> different pulse length in one direction compared to the other to
>> encode the direction on a single wire.
>>
>
> A very quick look at the datasheet seems to indicate that there are two
> sensors about 2mm apart and that logic may use these to sense whether
> the gear edge is metal to air or air to metal. I tend to agree with Andy
> that the chip logic may be the problem. That and six bits may get you a
> cuppa.
>
>
> --
> Kirk Wallace
> http://www.wallacecompany.com/machine_shop/
> http://www.wallacecompany.com/E45/
>
> --
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



-- 

Chris Albertson
Redondo Beach, California

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Jon Elson]

On 03/09/2016 11:38 AM, andy pugh wrote:
> On 9 March 2016 at 16:54,   wrote:
>> The issue is that we drilled about a .130 hole for the index but it acts 
>> odd.  If you rotate the spindle one direction it will pass the index and 
>> stay high until it passes the hole then you get a little index pulse low.  
>> If you run it the other direction it goes low after passing the index - and 
>> you get a positive index pulse at every passing of the hole.
> This may be the internal logic at work. The sensors are rated to zero
> speed, and I think have an internal offsetting circuit. However I
> can't see how it can tell which way the gear is turning unless it has
> two sensors.
> Some automotive sensors do actively sense direction, and send a
> different pulse length in one direction compared to the other to
> encode the direction on a single wire.
>
The sensors I used (Allegro ATS667LSGTN-T) have indeed, two 
Hall sensors separated by a mm or two, and a FF that flips 
one way when sensor 1 has more signal than sensor 2, and 
flips the other way when that condition is reversed.  (I 
think, maybe check the data sheet on your sensor to be sure 
how YOURS works.)  This reliably detects the passing of a 
gear tooth.  It does have the side effect of showing a 
reversed polarity when run the other direction.  My encoder 
counter treats the index signal as edge-triggered, so it 
won't have random alignment if the index is high most of the 
cycle.  Some other interface boards may not work like that.

It shouldn't matter, as the sensor is only used once at the 
beginning of the spindle-synched move.  if your encoder 
counter DOES act in a level-triggered manner, you would have 
to add a one-shot to detect the rising edge of the index pulse.

Jon

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[TJoseph Powderly]

a friend owns this company, i make nothing off its sale
but i think i should let people know about it as a resource

its designed for spindle speed control ( using little space )
and he has been successful selling it to oems in Taiwan

http://www.chiarey.com/products-en.html


  GUBOA IGS sensor  magnetic gear sensor

tomp


On 03/10/2016 01:26 AM, Chris Albertson wrote:
> Ideas:
>
> 1) How close to the motor is the sensor.  The motor might have a large
> external magnetic field.
>
> 2) I wonder if the shaft is magnetized while is spinning?  Is this reversed
> if the shaft spins the other way?
>
> 3)  Is the sensor powered from a reliable DC supply that is in no way
> connected or near the motor spindle power?
>
>
>
> On Wed, Mar 9, 2016 at 8:54 AM,  wrote:
>
>> We purchased some gear tooth sensors in hopes of using one of the gears in
>> the K&T for a spindle encoder.  Well - we ended up putting a shaft encoder
>> in the gearbox.
>>
>> http://www.digikey.com/product-search/en?keywords=620-1328-1-nd
>>
>> Well - move ahead a few years - We are putting a gear in the matsuura
>> spindle for an encoder.  the 2 sensors on the gear seem to work great for
>> quadrature signal.
>>
>> http://electronicsam.com/images/matsuura/sensors.jpg
>> http://electronicsam.com/images/matsuura/oscope.jpg
>>
>> (that is 6000rpm)
>>
>>
>> The issue is that we drilled about a .130 hole for the index but it acts
>> odd.  If you rotate the spindle one direction it will pass the index and
>> stay high until it passes the hole then you get a little index pulse low.
>> If you run it the other direction it goes low after passing the index - and
>> you get a positive index pulse at every passing of the hole.  I think it
>> should stay low (or high) no matter what direction the spindle is turning
>> (and get a pulse every pass of the index hole).
>>
>> clear as mud?
>> thanks
>> sam
>>
>>
>> --
>> Transform Data into Opportunity.
>> Accelerate data analysis in your applications with
>> Intel Data Analytics Acceleration Library.
>> Click to learn more.
>> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Gene Heskett]

On Wednesday 09 March 2016 11:54:06 sa...@empirescreen.com wrote:

> We purchased some gear tooth sensors in hopes of using one of the
> gears in the K&T for a spindle encoder.  Well - we ended up putting a
> shaft encoder in the gearbox.
>
> http://www.digikey.com/product-search/en?keywords=620-1328-1-nd
>
> Well - move ahead a few years - We are putting a gear in the matsuura
> spindle for an encoder.  the 2 sensors on the gear seem to work great
> for quadrature signal.
>
> http://electronicsam.com/images/matsuura/sensors.jpg
> http://electronicsam.com/images/matsuura/oscope.jpg
>
> (that is 6000rpm)
>
>
> The issue is that we drilled about a .130 hole for the index but it
> acts odd.  If you rotate the spindle one direction it will pass the
> index and stay high until it passes the hole then you get a little
> index pulse low.  If you run it the other direction it goes low after
> passing the index - and you get a positive index pulse at every
> passing of the hole.  I think it should stay low (or high) no matter
> what direction the spindle is turning (and get a pulse every pass of
> the index hole).
>
> clear as mud?

A bit dense & murky maybe, but since these are likely hall effect 
sensors, a good degaussing of the area might be in order. I'd almost bet 
the gear is magnetised.  Use an AC powered coil to make it buzz, and 
remove the coil very slowly to a 4 or 5 foot distance before turning it 
off.  And it would probably be a good idea to move the hall effect 
devices, and their excitation magnets, a similar distance in the other 
direction during the time that the coil is energized.  With cheaper 
magnets, the coil could be strong enough to wipe them clean too.  So get 
them out of the battlefield for the duration, then put them back & test.

> thanks
> sam
>
> --
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> ___
> 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 

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Old tek stuff

[Gene Heskett]

On Wednesday 09 March 2016 11:28:07 Chris Albertson wrote:

> On Wed, Mar 9, 2016 at 8:01 AM, Cecil Thomas  
wrote:
> > I can't think of a much better testimonial for a quality piece of
> > equipment.  During that time period they used nothing but the best
> > components and if no one made one good enough then they made it
> > themselves.
> >
> > By the way, it still works.
>
> I used one of those in  a lab at   UCLA  in the late 70's to fix some
> small computers.   Computer clocks were slow enough (1MHx or so)  in
> those days that an old scope would work.
>
> The down side of Tek building their old equipment so well was that
> only government labs and large companies could afford them

The upside of that was back in those days, the stuff they built had a 
lifetime warranty.  We had nearly a dozen of their broadcast waveform 
monitors, 529's IIRC, 2 at the transmitter, and in the middle my shift 
one morning one of them let all the smoke out while I was trying to baby 
a 4cx5000 whose screen was no longer in the shadow of the control grid.

I suppose a heat stress crack in the crt glass near the big rear socket, 
1/2" long I found when I pulled the old one out, so it let in the air 
rather slowly, but it eventually arced and crowbared the power 
transformer, killing it too. Both were tek made. I don't think the rep 
was happy, but in a week I had a fresh crt and a power tranny for it, no 
charge, and it was well past 5 yo.  There were a few other small parts 
cooked but we had those & I didn't fuss.

Today we're  lucky to get a 90 day warranty on a $500K transmitter.

I do not call that progress.

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 

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Learning LinuxCNC Youtube Channel

[Jim Craig]

I have added another video to the Learning LinuxCNC Youtube channel. 
Feel free to view it at your convenience. I should have another couple 
videos posted in the next week or two.

https://www.youtube.com/watch?v=A6y4DJasCJ4

Jim

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Old tek stuff

[Marcus Bowman]

Yes; it is only the ageing capacitors in some of the equipment from that era 
that lets the kit down.
I never used Tek equipment, but did use the early Data General SuperNova in the 
early 70s, then the DEC stuff, all of which was very well made. I smile when 
folk say "Why does such-and-such a computer manufacturer have such high 
prices?". Quality always costs. It usually pays too, although history tells us 
those companies don't always survive (with the exception of Apple, of course). 
The general public are price sensitive. It's pretty much the same story with 
quality tools.

Marcus

 
On 9 Mar 2016, at 16:01, Cecil Thomas wrote:

> 
> Mark,
> This is completely off topic but I thought you might be interested.
> I have a Tek 511-AD (Tek's first model) vintage somewhere around 1950 
> or 51 that I managed to rescue from an outside pile of electrical 
> scrap at McConnel AFB in Wichita Ks in 1971.
> It had been laying outside in the rain and snow for at least 3 months 
> if not longer.
> 
> I replaced the smashed CRT and made new shafts for a couple of 
> control pots that had been sheared off when it was tossed of the 
> truck and moved with the other trash in the bucket of a front end 
> loader.  I went to the Base Precision Measuring Equipment Lab and 
> asked if they still had the manual and they did.  It was they who 
> surplussed the scope.  They didn't need the manual any more so they 
> tossed it as well (to me).
> 
> I brought the caps up slow with a variac and it fired right up.
> 
> I can't think of a much better testimonial for a quality piece of 
> equipment.  During that time period they used nothing but the best 
> components and if no one made one good enough then they made it themselves.
> 
> By the way, it still works.
> 
> Cecil
> 
> 
> --
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] [OT] Group recommendations

[Nicklas Karlsson]

Look on debain, ubuntu, etc home page.

> Can anyone recommend a group that covers general Linux discussions. 
> 
> Distro specific is OK - Ubuntu, Debian, etc
> 
> TIA,
> 
> Richard

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] [OT] Group recommendations

[R.L. Wurdack]

Can anyone recommend a group that covers general Linux discussions. 

Distro specific is OK - Ubuntu, Debian, etc

TIA,

Richard


--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Kirk Wallace]

On 03/09/2016 09:38 AM, andy pugh wrote:
> On 9 March 2016 at 16:54,   wrote:
>> The issue is that we drilled about a .130 hole for the index but it acts 
>> odd.  If you rotate the spindle one direction it will pass the index and 
>> stay high until it passes the hole then you get a little index pulse low.  
>> If you run it the other direction it goes low after passing the index - and 
>> you get a positive index pulse at every passing of the hole.
>
> This may be the internal logic at work. The sensors are rated to zero
> speed, and I think have an internal offsetting circuit. However I
> can't see how it can tell which way the gear is turning unless it has
> two sensors.
> Some automotive sensors do actively sense direction, and send a
> different pulse length in one direction compared to the other to
> encode the direction on a single wire.
>

A very quick look at the datasheet seems to indicate that there are two 
sensors about 2mm apart and that logic may use these to sense whether 
the gear edge is metal to air or air to metal. I tend to agree with Andy 
that the chip logic may be the problem. That and six bits may get you a 
cuppa.


-- 
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[samco]

So - maybe it really doesn't matter..  It should be the same edge for 
threading/rigid tapping...

sam

On Wed, 9 Mar 2016 17:38:19 +
 andy pugh  wrote:
> On 9 March 2016 at 16:54,   wrote:
> > The issue is that we drilled about a .130 hole for the index but it acts 
> > odd.  If you rotate the spindle one direction it will pass the index and 
> > stay high until it passes the hole then you get a little index pulse low.  
> > If you run it the other direction it goes low after passing the index - and 
> > you get a positive index pulse at every passing of the hole.
> 
> This may be the internal logic at work. The sensors are rated to zero
> speed, and I think have an internal offsetting circuit. However I
> can't see how it can tell which way the gear is turning unless it has
> two sensors.
> Some automotive sensors do actively sense direction, and send a
> different pulse length in one direction compared to the other to
> encode the direction on a single wire.
> 
> -- 
> 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
> 
> --
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[andy pugh]

On 9 March 2016 at 16:54,   wrote:
> The issue is that we drilled about a .130 hole for the index but it acts odd. 
>  If you rotate the spindle one direction it will pass the index and stay high 
> until it passes the hole then you get a little index pulse low.  If you run 
> it the other direction it goes low after passing the index - and you get a 
> positive index pulse at every passing of the hole.

This may be the internal logic at work. The sensors are rated to zero
speed, and I think have an internal offsetting circuit. However I
can't see how it can tell which way the gear is turning unless it has
two sensors.
Some automotive sensors do actively sense direction, and send a
different pulse length in one direction compared to the other to
encode the direction on a single wire.

-- 
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

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[Chris Albertson]

Ideas:

1) How close to the motor is the sensor.  The motor might have a large
external magnetic field.

2) I wonder if the shaft is magnetized while is spinning?  Is this reversed
if the shaft spins the other way?

3)  Is the sensor powered from a reliable DC supply that is in no way
connected or near the motor spindle power?



On Wed, Mar 9, 2016 at 8:54 AM,  wrote:

> We purchased some gear tooth sensors in hopes of using one of the gears in
> the K&T for a spindle encoder.  Well - we ended up putting a shaft encoder
> in the gearbox.
>
> http://www.digikey.com/product-search/en?keywords=620-1328-1-nd
>
> Well - move ahead a few years - We are putting a gear in the matsuura
> spindle for an encoder.  the 2 sensors on the gear seem to work great for
> quadrature signal.
>
> http://electronicsam.com/images/matsuura/sensors.jpg
> http://electronicsam.com/images/matsuura/oscope.jpg
>
> (that is 6000rpm)
>
>
> The issue is that we drilled about a .130 hole for the index but it acts
> odd.  If you rotate the spindle one direction it will pass the index and
> stay high until it passes the hole then you get a little index pulse low.
> If you run it the other direction it goes low after passing the index - and
> you get a positive index pulse at every passing of the hole.  I think it
> should stay low (or high) no matter what direction the spindle is turning
> (and get a pulse every pass of the index hole).
>
> clear as mud?
> thanks
> sam
>
>
> --
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

Chris Albertson
Redondo Beach, California
--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] joint 2 exceeded errors

[Jon Elson]

> On Wednesday 09 March 2016 05:03:59 andy pugh wrote:
>
>>
>> If you had jogged a reference tool down to the PCB then touched-off
>> the G54 to that  position before touching-off the rest of the tools
>> you would have ended up with a tool table of offsets relative to the
>> reference tool, rather than referenced to the Z-axis home position.
>
This is how I learned to do it on the Allen-Bradley 7320.  
It worked on that, and works on LinuxCNC pretty much the 
same way.  (Some VERY old EMC(1) docs were actually an 
Allen-Bradley 8300 manual with hand scribbling.)

Jon

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Random quesion in regards to a gear sensor (maybe JonE)

[samco]

We purchased some gear tooth sensors in hopes of using one of the gears in the 
K&T for a spindle encoder.  Well - we ended up putting a shaft encoder in the 
gearbox.

http://www.digikey.com/product-search/en?keywords=620-1328-1-nd

Well - move ahead a few years - We are putting a gear in the matsuura spindle 
for an encoder.  the 2 sensors on the gear seem to work great for quadrature 
signal.  

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

(that is 6000rpm)


The issue is that we drilled about a .130 hole for the index but it acts odd.  
If you rotate the spindle one direction it will pass the index and stay high 
until it passes the hole then you get a little index pulse low.  If you run it 
the other direction it goes low after passing the index - and you get a 
positive index pulse at every passing of the hole.  I think it should stay low 
(or high) no matter what direction the spindle is turning (and get a pulse 
every pass of the index hole).

clear as mud?
thanks
sam

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Old tek stuff

[Chris Albertson]

On Wed, Mar 9, 2016 at 8:01 AM, Cecil Thomas  wrote:

>
>
>
> I can't think of a much better testimonial for a quality piece of
> equipment.  During that time period they used nothing but the best
> components and if no one made one good enough then they made it themselves.
>
> By the way, it still works.
>
>
>

I used one of those in  a lab at   UCLA  in the late 70's to fix some small
computers.   Computer clocks were slow enough (1MHx or so)  in those days
that an old scope would work.

The down side of Tek building their old equipment so well was that only
government labs and large companies could afford them


-- 

Chris Albertson
Redondo Beach, California
--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Old tek stuff

[Cecil Thomas]

Mark,
This is completely off topic but I thought you might be interested.
I have a Tek 511-AD (Tek's first model) vintage somewhere around 1950 
or 51 that I managed to rescue from an outside pile of electrical 
scrap at McConnel AFB in Wichita Ks in 1971.
It had been laying outside in the rain and snow for at least 3 months 
if not longer.

I replaced the smashed CRT and made new shafts for a couple of 
control pots that had been sheared off when it was tossed of the 
truck and moved with the other trash in the bucket of a front end 
loader.  I went to the Base Precision Measuring Equipment Lab and 
asked if they still had the manual and they did.  It was they who 
surplussed the scope.  They didn't need the manual any more so they 
tossed it as well (to me).

I brought the caps up slow with a variac and it fired right up.

I can't think of a much better testimonial for a quality piece of 
equipment.  During that time period they used nothing but the best 
components and if no one made one good enough then they made it themselves.

By the way, it still works.

Cecil


--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Need help with X200 VFD modbus

[Mark]

On 03/08/2016 06:33 PM, Gene Heskett wrote:
> Never had any issues with it turning on unwanted components.  Works
> fine in circuit too.  A lot of folks, on both the Tek and HP/Agilent
> groups are using the instrument, and I kinda went with the flow and
> bought a kit.  An hour or two in the afternoon, and it's all assembled
> up and ready to use.  Very handy little instruments, these hand held
> ESR meters.
>
> Mark
> Yes, they do an excellent job of measuring the most common electrolytic
> capacitor failure, detecting problems 2 years before a loss of
> capacitance will be measured by the common DVM in capacitance mode. They
> don't have a range to check the psu stuff, so generally speaking, that
> mode is similar in usefullness to those appendages on the belly of a
> boar hog.
>
> That $179 Wizard probably saved the station 500k USD in keeping about 15
> DVC-PRO broadcast VCR's going in years of their heyday. Capacitors and
> headwheels still cost us at least $100k though. Pinch rollers and other
> rubber parts were peanuts in comparison. Panasonic usually asked about
> $1500-$2500 a board for a new one, with 11 to 14 boards in a machine,
> and a minimum of a thou & parts at MSRP to repair one sent in.  That
> sort of outgo resulted in permission to buy the wizard in about 30
> seconds. Wasn't hard at all. :)  We still have, kept them like trophy
> deer horns, 3 ea, 2 lb coffee cans 2/3rds full of those capacitors that
> failed.  Thats a boatload because most were smaller than the eraser on a
> #2 lead pencil, so there are thousands in those coffee cans.  All
> surface mounted too.  And yours truly changed about 75% of them, with my
> assistant doing the last 25% after I gave him my office keys & a badly
> worn red office chair at a retirement party, on June 30th 2002.  But by
> 2004 much of that video gear had been replaced with servers with banks
> of hard drives.  Even the news cameras were insertable HD storage by
> then, still are today but replaced with full hidef stuff now.
>
> Cheers, Gene Heskett

As you know, I like to futz around with older Tek equipment.  I usually 
snatch up the "for parts or not working" units on the Bay of E.  Usually 
the problem(s) is centered in the power supply.  First thing I look at 
after dragging the schematics out are the caps. Visual, then an ESR 
search.  A lot of the units that were made in the era I like to fix have 
bad electrolytic caps, and a good many have those bad caps that were 
installed in a lot of other hardware like computers and such.  I forget 
if it was the Chinese or Koreans that tried to make caps based on stolen 
tech, but a lot of those caps found their way into our electronics.  Tek 
was no different. At any rate, the ESR meter makes troubleshooting power 
supplies bunches easier.

Mark

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] joint 2 exceeded errors

[Gene Heskett]

On Wednesday 09 March 2016 05:03:59 andy pugh wrote:

> On 9 March 2016 at 00:24, Gene Heskett  wrote:
> > Just one problem, the limit box in the backplot says the tool tip is
> > only about 150 thou below the z axis upper limit!, and my code
> > refuses to run because it runs up 4" to a tool change height, which
> > there are 3 of in the code..
> >
> > This does not make sense, the machine, when at the material top
> > surface, is almost 5" below the upper limit/home switch.
>
> What I think has happened here is that you have filled the tool table
> with tool-lengths that are approximately the full travel of the
> machine (and negative)
> What numbers do you see in the tool table?
>
> That shouldn't be a problem, but you need to then set up the G54
> offsets too to pull everything down again by touching-off to the work
> with a tool in the spindle and with the tool offsets (G43) enabled so
> that the system knows to factor the tool-length into the calculations.
>
> Starting with a blank canvas, you have a situation where Z=0 is at the
> top of machine travel.
> You then inserted a tool and jogged down several inches, and told the
> system that Z was now at zero. The system still thinks that Z=0 is at
> the top of machine travel, so the tool offset end up being negative
> (imagine a virtual tool pointing back up the spindle to the top home
> switch location :-)
>
> As long as the tool table is consistent, this is probably OK. But I
> prefer to work a slightly different way.
>
> I have my T99 which is a touch-probe. I have decided that that tool is
> of zero length. Every other tool has a Z-offset above or below that
> nominal zero.
> (On the lathe the reference tool is the turn/face T1). I know that
> both these tools should have zero offset, and if I mess up and put an
> offset in them then I can clear it with G10.
>
> If you had jogged a reference tool down to the PCB then touched-off
> the G54 to that  position before touching-off the rest of the tools
> you would have ended up with a tool table of offsets relative to the
> reference tool, rather than referenced to the Z-axis home position.

And that I believe explains it, I put the measured offset into all 3 
tools in the tool table. I have also rewrote the homing to put z=0 at 
50% of the Z travel.  I think I am good now if I can get my code to 
behave itself.

thank you Andy.

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 

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] joint 2 exceeded errors

[andy pugh]

On 9 March 2016 at 05:29, Gene Heskett  wrote:
> One thing I learned, it seems a g43 TLO comp enable, is cumulative,

No, it isn't. But you can end up increasing the tool length offset by
a sequence of tool-touch-off and work-touch-off actions.

I think you need to sit back and clarify in your head the way that the
absolute axis positions, G54 offsets and tool-length offsets interact,
and how LinuxCNC uses those numbers in calculating values for the
various varieties of G10 command.

In the Axis GUI Tool-Touch-Off is equivalent to G10 L10
http://linuxcnc.org/docs/2.7/html/gcode/g-code.html#gcode:g10-l10

The "other" touch-off (I am still on 2.6 so only have one button
still) is G10 L20
http://linuxcnc.org/docs/2.7/html/gcode/g-code.html#gcode:g10-l20

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] joint 2 exceeded errors

[andy pugh]

On 9 March 2016 at 00:24, Gene Heskett  wrote:
> Just one problem, the limit box in the backplot says the tool tip is only
> about 150 thou below the z axis upper limit!, and my code refuses to run
> because it runs up 4" to a tool change height, which there are 3 of in
> the code..
>
> This does not make sense, the machine, when at the material top surface,
> is almost 5" below the upper limit/home switch.

What I think has happened here is that you have filled the tool table
with tool-lengths that are approximately the full travel of the
machine (and negative)
What numbers do you see in the tool table?

That shouldn't be a problem, but you need to then set up the G54
offsets too to pull everything down again by touching-off to the work
with a tool in the spindle and with the tool offsets (G43) enabled so
that the system knows to factor the tool-length into the calculations.

Starting with a blank canvas, you have a situation where Z=0 is at the
top of machine travel.
You then inserted a tool and jogged down several inches, and told the
system that Z was now at zero. The system still thinks that Z=0 is at
the top of machine travel, so the tool offset end up being negative
(imagine a virtual tool pointing back up the spindle to the top home
switch location :-)

As long as the tool table is consistent, this is probably OK. But I
prefer to work a slightly different way.

I have my T99 which is a touch-probe. I have decided that that tool is
of zero length. Every other tool has a Z-offset above or below that
nominal zero.
(On the lathe the reference tool is the turn/face T1). I know that
both these tools should have zero offset, and if I mess up and put an
offset in them then I can clear it with G10.

If you had jogged a reference tool down to the PCB then touched-off
the G54 to that  position before touching-off the rest of the tools
you would have ended up with a tool table of offsets relative to the
reference tool, rather than referenced to the Z-axis home position.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users