Re: [Emc-users] advice on signal grounding between mesa 7i76e and dmm dyn4 servo drive

2017-03-26 Thread Chris Albertson 
Never ground a shield at both ends.  If you do current can flow in the
shield.

Also every conductor that does cary current should have a conductors that
caries the opposite current  very close and running in parallel, preferably
the two are twisted together.   So, the power supply's return is paired
with the positive conductor.   Same for power leads going to motors and
ideally even signal lines are all paired (use differential signaling.)
 Ideally, these would all be inside their own shields and al the shields
would be ground at the same end at one place.

Yes, it is hard to follow these rules 100%   Some time positive and
negative terminals are separated by to much distance.  But try to do as
much as you can.

One fact to keep in mind, A twisted pair, where conductors cary opposing
current, even if the current is large produce in effect no electric field 8
wires diameters away.   The field self-cancels that quickly.  Twisting is
very effective.




On Sun, Mar 26, 2017 at 12:01 AM, linden  wrote:

> Hello All
>
> Should I ground both ends of the shield for shielded pair? I am not sure
> what the best practice is see pdf attached for what I have now in black and
> the chunk in red I am not sure about.
>
> --

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] advice on signal grounding between mesa 7i76e and dmm dyn4 servo drive

2017-03-26 Thread Gene Heskett 
On Sunday 26 March 2017 03:01:17 linden wrote:

> Hello All
>
> Should I ground both ends of the shield for shielded pair?

No, only one end. Actually I think best practice is a single #10 bolt in 
the middle of it all, to which all grounds and all cable shields are 
connected, with the grounds going here and there both heavy and 
insulated so they won't setup a ground loop by touching something on the 
way to connect a ground on something else.

This is called a star ground.

Shielded cables such as those from a stepper driver to its motor, I cut 
into the plastic sheath and separate the drain wire, and solder a 2" 
piece of 1/4" flat braid, with an eyelet crimped on one end, the other 
end folded over the drain wire and soldered to it, using a 2 or 3% 
silver bearing solder. Good connection, and stronger than the eutectic 
alloys. The eyelet gets dropped over the bolt and the nut replaced and 
tightened.

> I am not 
> sure what the best practice is see pdf attached for what I have now in
> black and the chunk in red I am not sure about.

I would not depend on the connector shell, trimming the shielding short 
so it doesn't touch there, nor at the motor, but does have the short tab 
described above soldered to the drain wire as it goes past the bolt on 
its way out of the box, and connected to the common ground bolt. You 
should run a piece of the braid from this bolt to the ground symbol 
terminal on the power supplies, then feed the drivers power from the 
psu's via twisted pair, and it may help if a ferrite clamp-on is applied 
close to the drivers motor outputs, and another applied at the psu 
outputs, so the bead covers both wires in each case. This will block 
the "longitudinal" noise but has no effect on the psu's switching noise. 
In my case the vfd's motor wires are in the same 1" conduit as the power 
coming in is in, and I had to put a 3 phase corcom 20 amp "brick wall" 
in the circuit as close to the vfd's output terminals as was practical, 
getting rid of the vfd's noise it was injecting into the incoming power 
for all my electronics. This has not appeared to have affected the vfd.

One other thing, there is not any filtering or shielding in the 5v runs 
to the encoders I installed to replace the hand cranks removed from this 
Sheldon 11x36 lathe, its all in cat 5 cable, and the returns from the 
encoder were loaded with high voltage electrical spikes on the corners 
of what should have been the 4.9 volt or so square waves it produces. So 
those signal wires now go past a board with 6 each (I buy them in 100 
packs) 1n914 diodes to ground for each signal wire so as to clamp the 
positive peaks to about 4 volts, and while its working, I may add 
another diode faceing the other way to clamp any ground undershoots to 
about -0.55 volts to protect the 7i90's inputs.

This board in particular seems to be more easily damaged by the noise 
spikes and static electricity*, than a 5i25, which I also am familiar 
with, but the 7i90 gives you I/O to throw away, ( 72 minus the 
generators and encoders enabled ) whereas the input facilities of the 
5i25 are limited to the 5 the parport offers, per connector, which it 
has 2 of.
 
*static electricity, generated  by my butt rubbing on the garage door 
behind the lathe as its been covered by an additional layer of 2" blue 
styrofoam from Lowes, adding another R11 to most of the door making my 
otherwise well insulated garage very easy to heat or air condition. So I 
have to slip on one of those wrist grounding straps when I'm working on 
the electronics to keep my charges down else I can blow the 7i90 with my 
generated static. I've BTDT too. :(

> The only other connections I have are:
>   - 24 volt supply for the mesa board going into TB3 pin22 as
> +24vdc - the corresponding ground from the power supply going into TB3
> pin 23

This - supply wire to T3-23  s/b grounded as it goes by that bolt.

>   - a jumper going from IO ground pin 9 on JP4 of the servo drive
> to TB3 pin 24

As should this wire UNLESS its internally connected to the TB3-23, verify 
with an ohm-meter.

Proper grounding can be tricky because you don't always know whats in the 
box. So keep a meter handy.

> Thanks for any help and advice
> linden

Thats my $0.02, but adjust for inflation since 1934. :) I am also a 
C.E.T. 

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] advice on signal grounding between mesa 7i76e and dmm dyn4 servo drive

2017-03-26 Thread linden 
thanks Erik

I don't have any formal electronics back ground and have been 
struggling to get my head rapped around driving this servo with the mesa 
card. The documentation for both seems pretty good but it is all in a 
language I don't fully understand. It is very helpful having a second 
set of eyes going over my work and making sure things really work how i 
imagine they work in my head.

thanks again


linden


On 17-03-26 12:38 AM, Erik Christiansen wrote:
> On 26.03.17 00:01, linden wrote:
>> Hello All
>>
>> Should I ground both ends of the shield for shielded pair? I am not sure
>> what the best practice is see pdf attached for what I have now in black and
>> the chunk in red I am not sure about.
> Grounding the shield at both ends will create a ground loop, and
> therefore trouble. If the " JP4 Shell" in the drawing were only the
> connector shell, not connected to anything, then connecting to it would
> merely extend the screen right to the end. BUT you've added "Gnd" to the
> label, to indicate that it's grounded at that end, right? Draw a great
> looping line connecting the two "Gnd" labels. What do you have if the
> red connection is made?
>
> My preference is to ground screens at the driving end, and you've done
> that. No more is needed.
>
> Erik
>
> --
> 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] advice on signal grounding between mesa 7i76e and dmm dyn4 servo drive

2017-03-26 Thread Erik Christiansen 
On 26.03.17 00:01, linden wrote:
> Hello All
> 
> Should I ground both ends of the shield for shielded pair? I am not sure
> what the best practice is see pdf attached for what I have now in black and
> the chunk in red I am not sure about.

Grounding the shield at both ends will create a ground loop, and
therefore trouble. If the " JP4 Shell" in the drawing were only the
connector shell, not connected to anything, then connecting to it would
merely extend the screen right to the end. BUT you've added "Gnd" to the
label, to indicate that it's grounded at that end, right? Draw a great
looping line connecting the two "Gnd" labels. What do you have if the
red connection is made?

My preference is to ground screens at the driving end, and you've done
that. No more is needed.

Erik

--
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