Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

[...]
> So now I'm in the genesis stage of adding a couple timedelay's to
> TLM's hal file to detect a stopped spindle 100 milliseconds after the
> motor has been enabled.  The idea is to issue an estop, and hold it
> long enough (say 5 seconds?) the motor will have coasted to a stop
> before its released and I can restart from the F1 key.

I "think" I have the .hal logic working, but feeding the detected kill 
signal into estop.in does not do anything. So where should I feed this 
estop signal because the spindle isn't turning when its been commanded 
to turn?

Once thats been corrected, can I do away with an and2 by feeding 
motion.spindle-on into an edge input as a parameter, effectively 
disabling it when the code stops the spindle?

Thanks everybody.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Tuesday 13 September 2016 07:57:56 andy pugh wrote:

> On 13 September 2016 at 11:51, Gene Heskett  
wrote:
> > > ethernet-blue-data-cable-4x2x0-12mm-od-6-3mm-49-1790
> >
> > 5 bucks a meter? A week to ship ARO?
>
> Well, clearly, you would look for a local vendor. I am just pointing
> out that the stuff exists.

I will check at Tolley Electric in Clarksburg the next time I go there, 
which ATM looks like Friday as I've a medical appointment at the new 
hospital just north of there.
>
> Should I stop posting illustrative links? I can't be bothered to try
> to find US-based alternative sources. Especially if you can't.

I hope not, as I do find them very educational.

I did a test fit of the SS pipe but it can't clear the crossfeeds gib 
adjusting screws. I was hoping I could put it there so I wouldn't smash 
it up running into the tailstock. The alternative is to put the cable 
chain on the front and that won't float at all. The dog screws might 
clear if I just glued it, but that I've not checked, been busy with TLM 
since about 5ish this morning. I found that I did not have a common 
connection from the - end of the spindle supply to the rest of the 
system grounding.

So now I'm in the genesis stage of adding a couple timedelay's to TLM's 
hal file to detect a stopped spindle 100 milliseconds after the motor 
has been enabled.  The idea is to issue an estop, and hold it long 
enough (say 5 seconds?) the motor will have coasted to a stop before its 
released and I can restart from the F1 key.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Jon Elson]

On 09/13/2016 12:07 PM, Jon Elson wrote:
> On 09/12/2016 10:25 PM, Gene Heskett wrote:
>> On Monday 12 September 2016 23:08:53 Gregg Eshelman wrote:
>>
>>> Why not use twisted pair ethernet cable?
>>>
>> Because its single strand solid and supposedly fragile?  The ribbon is 7
>> strand & able to stand more flexing.
>>
>>
> There is "jumper" cable that is stranded, although the

I meant "patch" above.
> strands are not really fine.  I got a 250' box of Belkin
> brand cable at MicroCenter to wire up my phone system.
>
>
Even our local Ace hardware store has patch cables, now.  
Digi-Key, Newark, Allied, Mouser, etc. all stock this 
networking stuff, too, and the prices are quite cheap.

Jon

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

Re: [Emc-users] output sink of the 667 hall effect?

[Chris Albertson]

On Mon, Sep 12, 2016 at 8:25 PM, Gene Heskett  wrote:

> On Monday 12 September 2016 23:08:53 Gregg Eshelman wrote:
>
> > Why not use twisted pair ethernet cable?
> >
> Because its single strand solid and supposedly fragile?  The ribbon is 7
> strand & able to stand more flexing.
>

There are two kinds of Ethernet cable.  (1) patch cords, there are flexible
with stranded wires inside and molded over RJ45 connectors on each end. and
(2) the kind that comes in 500 coils in a box and goes inside walls.  The
stuff is sold core.

For data cables you would only use the flexible patch cords

That said for short run of a single ended signal they work fine.  I would
go with the twisted pairs as are found in Ethernet cable for differential
signals.   After all Ethernet uses a fast baseband differential signal

Gear teeth sensing does not require a high performance solution the pulse
rate is slow, a couple KHz at most and the signal is very strong compared
to any noise that might be present.



> And I found, at Wallies of all places tonight, some glue pens that
> dispense uv setting resin with an included uv light to cure the stuff.
>
> I intend to make some use of that to prevent wandering wires from
> wandering while they are inside the headstock.
> >
> >
> > From: Gene Heskett 
> >  To: emc-users@lists.sourceforge.net
> >  Sent: Monday, September 12, 2016 11:43 AM
> >  Subject: [Emc-users] output sink of the 667 hall effect?
> >
> > Greetings all;
> >
> > I see by the spec sheet that the allegro 667 gear sensor can sink up
> > to 25 mills.
> >
> > But with 3 such sensors on the end of a long cable back to the
> > breakout board, probably 7 foot by the time I get it routed, I am
> > concerned with the potential for crosstalk noise in spite of putting a
> > bypass capacitor on the encoder itself.
> >
> > This noise would be reduced by a lesser amount of current. So, for
> > those of you using this device, what value is the resistor are you
> > using for a pullup at the breakout board?
> >
> > And if using a piece of idc ribbon cable, it seems to make sense to
> > put a grounded wire between each of the A|Z|B conductors, making that
> > a 7 conductor cable rather than a 5. I have around 90 feet of 26 wire
> > ribbon I can split off what I need from.
> >
> > Good idea? Or off the wall & probably will be plagued by noise if I
> > shoot for a 10 milliamp pullup with a 470 ohm pullup, or even 5 mills
> > with a 1k pullup.
> >
> > Time to assemble that motor driver box I believe.
> >
> >
> > --
> > ___
> > 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 
>
> 
> --
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>



-- 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Jon Elson]

On 09/12/2016 10:25 PM, Gene Heskett wrote:
> On Monday 12 September 2016 23:08:53 Gregg Eshelman wrote:
>
>> Why not use twisted pair ethernet cable?
>>
> Because its single strand solid and supposedly fragile?  The ribbon is 7
> strand & able to stand more flexing.
>
>
There is "jumper" cable that is stranded, although the 
strands are not really fine.  I got a 250' box of Belkin 
brand cable at MicroCenter to wire up my phone system.

Jon

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

Re: [Emc-users] output sink of the 667 hall effect?

[John Kasunich]

On Tue, Sep 13, 2016, at 06:51 AM, Gene Heskett wrote:
> On Tuesday 13 September 2016 04:13:15 andy pugh wrote:
> 
> > ethernet-blue-data-cable-4x2x0-12mm-od-6-3mm-49-1790
> 
> 5 bucks a meter? A week to ship ARO? They obviously aren't "hungry".  
> Based on buying some cutting tools from a limey vendor a year ago, I'll 
> be 6 weeks getting it into my hands. Nyet.

Andy posts UK based suppliers because he is in the UK.  The point is
that the stuff is out there, and there is probably a US seller if you
use the right search terms.

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

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

Re: [Emc-users] output sink of the 667 hall effect?

[andy pugh]

On 13 September 2016 at 11:51, Gene Heskett  wrote:

> > ethernet-blue-data-cable-4x2x0-12mm-od-6-3mm-49-1790
>
> 5 bucks a meter? A week to ship ARO?
>

Well, clearly, you would look for a local vendor. I am just pointing out
that the stuff exists.

Should I stop posting illustrative links? I can't be bothered to try to
find US-based alternative sources. Especially if you can't.

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Tuesday 13 September 2016 07:19:23 Erik Christiansen wrote:

> On 13.09.16 06:51, Gene Heskett wrote:
> > The ideal cable would be a star-quad that Clark Wire in windy
> > carried 20 years ago, it was 18 gauge in a mylar film covered in 68%
> > braided shield, very limp stuff. But Susan is not importing the
> > heavier gauges anymore, and its a piece of her 24 gauge thats
> > heating while driving the x motor on the little monster.  So as I
> > revive the little monster, some of that grey stuff will be used to
> > replace it.  But that drag chain has a much smaller turn around,
> > abusing that cable, so that will fail faster.  How fast remains TBD.
>
> Will TLM remain in use, or be decommissioned once the Sheldon flies?
> If the former, perhaps parallel two runs of the flexible stuff¹; if
> the latter, then either way works?
>
> Erik

I intend to keep it running once revived. The tapered gibs have restored 
the accuracy to about a thou tolerance, at least a much as the worn z 
ball screw can muster. Its starting to go because it has no swarf 
shielding. If I replace that screw again, now I can find some rubber 
bellows type boots to cover it with. I couldn't find anything but 
satellite jack covers when I converted it, which are much too heavy and 
couldn't be crushed enough to allow normal motions so they never got 
installed. Now some of that stuff is showing up on fleabay. I am hoping 
that the ones I have on a boat someplace, have a vent like the jack 
covers have, else trapped air might be a problem. TBD.

> ¹ Two runs might lead to 1/9 of the heating, through tripling of the
>   power conductors, since there's no call to parallel the signal
>   conductors.
>
> --
> ___
> 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 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Erik Christiansen]

On 13.09.16 06:51, Gene Heskett wrote:
> The ideal cable would be a star-quad that Clark Wire in windy carried 20 
> years ago, it was 18 gauge in a mylar film covered in 68% braided 
> shield, very limp stuff. But Susan is not importing the heavier gauges 
> anymore, and its a piece of her 24 gauge thats heating while driving the 
> x motor on the little monster.  So as I revive the little monster, some 
> of that grey stuff will be used to replace it.  But that drag chain has 
> a much smaller turn around, abusing that cable, so that will fail 
> faster.  How fast remains TBD.

Will TLM remain in use, or be decommissioned once the Sheldon flies?
If the former, perhaps parallel two runs of the flexible stuff¹; if the
latter, then either way works?

Erik

¹ Two runs might lead to 1/9 of the heating, through tripling of the
  power conductors, since there's no call to parallel the signal
  conductors.

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Tuesday 13 September 2016 04:13:15 andy pugh wrote:

> ethernet-blue-data-cable-4x2x0-12mm-od-6-3mm-49-1790

5 bucks a meter? A week to ship ARO? They obviously aren't "hungry".  
Based on buying some cutting tools from a limey vendor a year ago, I'll 
be 6 weeks getting it into my hands. Nyet.

I have some over-large drag chain, and I can get a stiffer, grey poly 
jacketed, 4 conductors with shield thats 7 mm's in diameter at half that 
locally. Not drag chain rated, but it will probably outlast me as this 
chain turns around in about a 5 inch circle.  This chain will carry at 
least 4 of those but I only need two to the saddle, one for X motor & 
something smaller, mini zip cord works on the toy lathe for a home 
switch. I bought 100 feet of the 4+shield to wire the G0704 with & may 
have enough left to do 2 steppers and the encoder on this lathe.

I have some SS pipe, fuel line for a race car, a neighbor gave me that I 
can pull x motor power across the saddle in a piece of, and can probably 
rig a home switch on the rear, but that belongs in the crossfeed backed 
out position so that a home can be done with a workpiece mounted.  I 
don't have the saddle and crossfeed assembled well enough to survey that 
possibility yet. With the taper extension of the crossfeed still in 
place, its long enough I should be able to put a home switch back there 
and out of the way and still have it trip half a millimeter from the 
backout mechanical stop.  I see that in my mind as being on the same 
saddle extension I'll have to make to carry the drag chain.  That way, 
only the x motor power has to come across the saddle. If I can bend it, 
I might drill a hole thru the top of the saddle to get it into the apron 
and the motor. Its something I'll check out, but I suspect I'll have to 
make a mini-half shell elbow cover for.  Bridges not yet crossed, but it 
will need to be solved in due time.

The ideal cable would be a star-quad that Clark Wire in windy carried 20 
years ago, it was 18 gauge in a mylar film covered in 68% braided 
shield, very limp stuff. But Susan is not importing the heavier gauges 
anymore, and its a piece of her 24 gauge thats heating while driving the 
x motor on the little monster.  So as I revive the little monster, some 
of that grey stuff will be used to replace it.  But that drag chain has 
a much smaller turn around, abusing that cable, so that will fail 
faster.  How fast remains TBD.

So what I can get, or have already, will have to do.  Its not always 
ideal but does make the connections needed.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

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

Re: [Emc-users] output sink of the 667 hall effect?

[andy pugh]

On 13 September 2016 at 04:25, Gene Heskett  wrote:

> > Why not use twisted pair ethernet cable?
> >
> Because its single strand solid and supposedly fragile?


Fixed-installation ethernet cable is solid core. Patch cables are stranded.

If you want to get carried away you can get drag-chain rated,
oil-resistant, shielded Cat5. I did and did on my Holbrook build.
https://www.rapidonline.com/lappkabel-2170489-unitronic-
ethernet-blue-data-cable-4x2x0-12mm-od-6-3mm-49-1790

Though for the resolvers I used dedicated encoder + power cable, because I
found some cheap on eBay.

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gregg Eshelman]

There's stranded Ethernet cable.
For that UV curable stuff, get a 9 LED UV flashlight, cheap on eBay. Works much 
better than the twee little single LED that comes with the stuff. Reminds me I 
need to goop up the USB cord for my phone.



 
  From: Gene Heskett <ghesk...@shentel.net>
 To: emc-users@lists.sourceforge.net 
 Sent: Monday, September 12, 2016 9:25 PM
 Subject: Re: [Emc-users] output sink of the 667 hall effect?
   
On Monday 12 September 2016 23:08:53 Gregg Eshelman wrote:

> Why not use twisted pair ethernet cable?
>
Because its single strand solid and supposedly fragile?  The ribbon is 7 
strand & able to stand more flexing.

And I found, at Wallies of all places tonight, some glue pens that 
dispense uv setting resin with an included uv light to cure the stuff.

I intend to make some use of that to prevent wandering wires from 
wandering while they are inside the headstock.
   
 
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Monday 12 September 2016 23:08:53 Gregg Eshelman wrote:

> Why not use twisted pair ethernet cable?
>
Because its single strand solid and supposedly fragile?  The ribbon is 7 
strand & able to stand more flexing.

And I found, at Wallies of all places tonight, some glue pens that 
dispense uv setting resin with an included uv light to cure the stuff.

I intend to make some use of that to prevent wandering wires from 
wandering while they are inside the headstock.
>
>
> From: Gene Heskett 
>  To: emc-users@lists.sourceforge.net
>  Sent: Monday, September 12, 2016 11:43 AM
>  Subject: [Emc-users] output sink of the 667 hall effect?
>
> Greetings all;
>
> I see by the spec sheet that the allegro 667 gear sensor can sink up
> to 25 mills.
>
> But with 3 such sensors on the end of a long cable back to the
> breakout board, probably 7 foot by the time I get it routed, I am
> concerned with the potential for crosstalk noise in spite of putting a
> bypass capacitor on the encoder itself.
>
> This noise would be reduced by a lesser amount of current. So, for
> those of you using this device, what value is the resistor are you
> using for a pullup at the breakout board?
>
> And if using a piece of idc ribbon cable, it seems to make sense to
> put a grounded wire between each of the A|Z|B conductors, making that
> a 7 conductor cable rather than a 5. I have around 90 feet of 26 wire
> ribbon I can split off what I need from.
>
> Good idea? Or off the wall & probably will be plagued by noise if I
> shoot for a 10 milliamp pullup with a 470 ohm pullup, or even 5 mills
> with a 1k pullup.
>
> Time to assemble that motor driver box I believe.
>
>
> --
> ___
> 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 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Monday 12 September 2016 22:14:38 Jon Elson wrote:

> On 09/12/2016 12:43 PM, Gene Heskett wrote:
> > Greetings all;
> >
> > I see by the spec sheet that the allegro 667 gear sensor can sink up
> > to 25 mills.
> >
> > But with 3 such sensors on the end of a long cable back to the
> > breakout board, probably 7 foot by the time I get it routed, I am
> > concerned with the potential for crosstalk noise in spite of putting
> > a bypass capacitor on the encoder itself.
>
> Just don't source any more current than you actually need.
> I have 3 of these on my spindle encoder, the cable is
> certainly more than 7 feet, and right next to a VFD-powered
> motor.  No problems so far.
> I have the pull-ups at the sensor end, and they are 1 K
> Ohm.  I do have a .1 uF (or possibly 1 uF) cap across the
> power at the sensor end junction block, that is maybe 4"
> from the actual sensors.  I used 4-conductor shielded cable,
> with the shield as the ground return (I know, not best
> practice, but I didn't have 5-conductor shielded cable).
>
> Jon
>
I've had a 6 foot piece of star-quad cable doing that on the little 
monster lathe for several years.  It just works.  And its cooked 
spaghetti pliable to boot. But I just found an identical 3 foot hunk of 
it, running an 8 wire 425oz motor wired parallel, is heating noticeably. 
Unfortunately, the next heavier gauge of that cable has been "out of 
stock" for 4 or 5 years.  Its just enough bigger that its quite hard to 
put a neutrik connector on it.  So I'm stuck using a much stiffer cable 
I can get at the electrical wholesaler in Clarksburg, and bigger 
cable-chain too.

> --
> ___
> 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 

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gregg Eshelman]

Why not use twisted pair ethernet cable?




From: Gene Heskett 
 To: emc-users@lists.sourceforge.net 
 Sent: Monday, September 12, 2016 11:43 AM
 Subject: [Emc-users] output sink of the 667 hall effect?
   
Greetings all;

I see by the spec sheet that the allegro 667 gear sensor can sink up to 
25 mills.

But with 3 such sensors on the end of a long cable back to the breakout 
board, probably 7 foot by the time I get it routed, I am concerned with 
the potential for crosstalk noise in spite of putting a bypass capacitor 
on the encoder itself.

This noise would be reduced by a lesser amount of current. So, for those 
of you using this device, what value is the resistor are you using for a 
pullup at the breakout board? 

And if using a piece of idc ribbon cable, it seems to make sense to put a 
grounded wire between each of the A|Z|B conductors, making that a 7 
conductor cable rather than a 5. I have around 90 feet of 26 wire ribbon 
I can split off what I need from.

Good idea? Or off the wall & probably will be plagued by noise if I shoot 
for a 10 milliamp pullup with a 470 ohm pullup, or even 5 mills with a 
1k pullup.

Time to assemble that motor driver box I believe.
   
 
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] output sink of the 667 hall effect?

[Jon Elson]

On 09/12/2016 12:43 PM, Gene Heskett wrote:
> Greetings all;
>
> I see by the spec sheet that the allegro 667 gear sensor can sink up to
> 25 mills.
>
> But with 3 such sensors on the end of a long cable back to the breakout
> board, probably 7 foot by the time I get it routed, I am concerned with
> the potential for crosstalk noise in spite of putting a bypass capacitor
> on the encoder itself.
Just don't source any more current than you actually need.  
I have 3 of these on my spindle encoder, the cable is 
certainly more than 7 feet, and right next to a VFD-powered 
motor.  No problems so far.
I have the pull-ups at the sensor end, and they are 1 K 
Ohm.  I do have a .1 uF (or possibly 1 uF) cap across the 
power at the sensor end junction block, that is maybe 4" 
from the actual sensors.  I used 4-conductor shielded cable, 
with the shield as the ground return (I know, not best 
practice, but I didn't have 5-conductor shielded cable).

Jon

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

Re: [Emc-users] output sink of the 667 hall effect?

[Chris Albertson]

On Mon, Sep 12, 2016 at 3:32 PM, Gene Heskett  wrote:
> On Monday 12 September 2016 14:44:33 Chris Albertson wrote:
>
>> I think the RC filter idea would work OK.  Gear teach are after all a
>> low frequency signal, so a filter made with a cap size for a corner
>> freq of (say) 100KHz will keep RF out of the line.
>>
> I was thinking more in terms of a rail to rail bypass.

If the current goes through a series resistor and there is a capacitor
to ground it is a filter.   You can say "bypass" and many people do
and everyone will know what you mean but the voltage across the cap
will have frequency content defined by the RC constant of the filter.
The product of R and C is what matters.You design this likely by
selecting R based on desired DC characteristics then select C based on
desired frequency cut off.  Just like designing a filter so I call
them filters

In larger factory setting people use driver/receiver chips




-- 

Chris Albertson
Redondo Beach, California

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

Re: [Emc-users] output sink of the 667 hall effect?

[Gene Heskett]

On Monday 12 September 2016 14:44:33 Chris Albertson wrote:

> I think the RC filter idea would work OK.  Gear teach are after all a
> low frequency signal, so a filter made with a cap size for a corner
> freq of (say) 100KHz will keep RF out of the line.
>
I was thinking more in terms of a rail to rail bypass.

> That said the "correct" way to do this is with a pair of
> driver/receiver chips.   If you use RS422 (that is a differential
> signal) you can likely go over a 1/4 mile of trusted pair wire.
>
> As for wire why not use Cat 6 or better for a machine shop cat-6
> shielded.   It is terminated in RJ45 and it's dirt cheap.  And has
> enough pairs (4) for your use.
>
> Also I was reading LinuxCNC docs, it said a spindle encoder needs to
> be quadrature.  You need two sensors for that phase A and B plus a
> third for the index.Cat 6 has four pairs so you can use the 4th
> for power to the sensors.

Thats been the plan from the getgo, and its required for rigid tapping.
Only way to fly IMO.

> As I said, the simple RC filter like you suggested can work but the
> "classic" method for use in a shop filled with high power machines is
> to use a differential single as it is immune to common mode noise,
> then use twisted pair in a sealed that is grounded on ONE end only,

I can switch to shielded cable once I get outside the spindle housing, so 
if its needed, I will.

> On Mon, Sep 12, 2016 at 10:43 AM, Gene Heskett  
wrote:
> > Greetings all;
> >
> > I see by the spec sheet that the allegro 667 gear sensor can sink up
> > to 25 mills.
> >
> > But with 3 such sensors on the end of a long cable back to the
> > breakout board, probably 7 foot by the time I get it routed, I am
> > concerned with the potential for crosstalk noise in spite of putting
> > a bypass capacitor on the encoder itself.
> >
> > This noise would be reduced by a lesser amount of current. So, for
> > those of you using this device, what value is the resistor are you
> > using for a pullup at the breakout board?
> >
> > And if using a piece of idc ribbon cable, it seems to make sense to
> > put a grounded wire between each of the A|Z|B conductors, making
> > that a 7 conductor cable rather than a 5. I have around 90 feet of
> > 26 wire ribbon I can split off what I need from.
> >
> > Good idea? Or off the wall & probably will be plagued by noise if I
> > shoot for a 10 milliamp pullup with a 470 ohm pullup, or even 5
> > mills with a 1k pullup.
> >
> > Time to assemble that motor driver box I believe.

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 
> >
> > 
> >-- What NetFlow Analyzer can do for you? Monitors network
> > bandwidth and traffic patterns at an interface-level. Reveals which
> > users, apps, and protocols are consuming the most bandwidth.
> > Provides multi-vendor support for NetFlow, J-Flow, sFlow and other
> > flows. Make informed decisions using capacity planning reports.
> > http://sdm.link/zohodev2dev
> > ___
> > 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 

--
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity 
planning reports. http://sdm.link/zohodev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] output sink of the 667 hall effect?

[Chris Albertson]

I think the RC filter idea would work OK.  Gear teach are after all a
low frequency signal, so a filter made with a cap size for a corner
freq of (say) 100KHz will keep RF out of the line.

That said the "correct" way to do this is with a pair of
driver/receiver chips.   If you use RS422 (that is a differential
signal) you can likely go over a 1/4 mile of trusted pair wire.

As for wire why not use Cat 6 or better for a machine shop cat-6
shielded.   It is terminated in RJ45 and it's dirt cheap.  And has
enough pairs (4) for your use.

Also I was reading LinuxCNC docs, it said a spindle encoder needs to
be quadrature.  You need two sensors for that phase A and B plus a
third for the index.Cat 6 has four pairs so you can use the 4th
for power to the sensors.

As I said, the simple RC filter like you suggested can work but the
"classic" method for use in a shop filled with high power machines is
to use a differential single as it is immune to common mode noise,
then use twisted pair in a sealed that is grounded on ONE end only,





On Mon, Sep 12, 2016 at 10:43 AM, Gene Heskett  wrote:
> Greetings all;
>
> I see by the spec sheet that the allegro 667 gear sensor can sink up to
> 25 mills.
>
> But with 3 such sensors on the end of a long cable back to the breakout
> board, probably 7 foot by the time I get it routed, I am concerned with
> the potential for crosstalk noise in spite of putting a bypass capacitor
> on the encoder itself.
>
> This noise would be reduced by a lesser amount of current. So, for those
> of you using this device, what value is the resistor are you using for a
> pullup at the breakout board?
>
> And if using a piece of idc ribbon cable, it seems to make sense to put a
> grounded wire between each of the A|Z|B conductors, making that a 7
> conductor cable rather than a 5. I have around 90 feet of 26 wire ribbon
> I can split off what I need from.
>
> Good idea? Or off the wall & probably will be plagued by noise if I shoot
> for a 10 milliamp pullup with a 470 ohm pullup, or even 5 mills with a
> 1k pullup.
>
> Time to assemble that motor driver box I believe.
>
> 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 
>
> --
> What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
> patterns at an interface-level. Reveals which users, apps, and protocols are
> consuming the most bandwidth. Provides multi-vendor support for NetFlow,
> J-Flow, sFlow and other flows. Make informed decisions using capacity
> planning reports. http://sdm.link/zohodev2dev
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



-- 

Chris Albertson
Redondo Beach, California

--
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity 
planning reports. http://sdm.link/zohodev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] output sink of the 667 hall effect?

[sam sokolik]

I used a 1k ohm.  but..  We ran that right into a CUI line driver.. (I 
don't think you can get them anymore)  then into a mesa differential 
encoder counter...
http://www.digikey.com/product-search/en?keywords=102-1787-ND

here is our mounting.  You can see the index hole (the index sensor 
isn't mounted in this photo)

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

sam

On 9/12/2016 12:43 PM, Gene Heskett wrote:
> Greetings all;
>
> I see by the spec sheet that the allegro 667 gear sensor can sink up to
> 25 mills.
>
> But with 3 such sensors on the end of a long cable back to the breakout
> board, probably 7 foot by the time I get it routed, I am concerned with
> the potential for crosstalk noise in spite of putting a bypass capacitor
> on the encoder itself.
>
> This noise would be reduced by a lesser amount of current. So, for those
> of you using this device, what value is the resistor are you using for a
> pullup at the breakout board?
>
> And if using a piece of idc ribbon cable, it seems to make sense to put a
> grounded wire between each of the A|Z|B conductors, making that a 7
> conductor cable rather than a 5. I have around 90 feet of 26 wire ribbon
> I can split off what I need from.
>
> Good idea? Or off the wall & probably will be plagued by noise if I shoot
> for a 10 milliamp pullup with a 470 ohm pullup, or even 5 mills with a
> 1k pullup.
>
> Time to assemble that motor driver box I believe.
>
> Cheers, Gene Heskett


--
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are 
consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
J-Flow, sFlow and other flows. Make informed decisions using capacity 
planning reports. http://sdm.link/zohodev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users