On Friday 13 July 2018 01:48:57 Erik Christiansen wrote:
> On 12.07.18 15:33, John Dammeyer wrote:
> > So what happens when the equipment with the 24V supply is 30m long
> > in multiple steel frames? There would be a bonding wire from frame
> > to frame since you wouldn't want to bond one end to one AC ground
> > outlet and the other end to a different AC ground outlet.
>
> Rather than chain a whole lot of modules on one long ground wire,
> where each can interfere with the others by imposing HF noise on the
> common ground impedance, thereby coupling it into the others, I'd wire
> them individually back to the power supply. ("Star Earth", as Gene so
> rightly points out.)
>
> If there is internal DC connection to the equipment case, I'd provide
> isolated mounts to defeat the earth loops which would most likely
> otherwise cause problems.
As I describe in a clarifying post. And I'd delete the "most likely
otherwise" above. This problem in my sheldon lathe build caused me to
ditch the switching supplies for the motors in favor of some toroid
transformers, bridge rectifiers and large filter caps, simply because I
could much more easily ground the negative rails to that common point
bolt w/o cause ground loops. The switchers were injecting noise with a
bandwidth exceeding my gigahertz sampling scope, at voltages (30 volts
peak) sufficient to blow an unprotected 7i90HD instantly.
I changed the supplies to nice quiet discrete analog, and added the
7i42TA's, giving me a much more easily wired interface, reducing that
noise source to the drivers current regulation switching and under 200
millivolts at the 7i42TA inputs. And modified the config to work around
the blown gates in the last of 3 7i90's I had bought trying to make it
work. A year later its still running well on a half blown 7i90HD.
As I am a Certified Electronics Technician, I should have done the analog
supplies in the first place, but I was in a hurry to make it move.
Presumably I could have glued the switchers onto a couple strips of
double sided pcb material, also glued to the mounting subplate in the
motor driver box, but I like solid mounts. And that reminded me of the
stupidity of short cuts. Not to mention that the pcb material used in
that fashion would have represented a good sized capacitor coupling the
switching noise they created into the chassis, not at all a good idea
once the ramifications of that capacitance were considered.
I did hook up a breadboard I could drive with a function generator, with
a common ground bolt hanging in the breeze. It worked to drive a 425oz 8
wire motor to just over 3000 revs, but noise from the bolt to the case
of the switcher sitting on wood with its ground terminal connected to
that common bolt, was over 100 volts p-p on the scope, with rise and
fall times equal to the scopes bandwidth. Theres 3 of those supplies in
the G0704 power box, and one AC powered driver for the z motor and my
portable phone is useless within 12 feet of it. They are that damned
noisy. It rings raggedly, and if I want to talk on it, I have to be 12
to 20 feet away. The sheldon's build does not bother the phone. I have
another of those phones out in the shop building, but there the
switchers are bolted down in closed metal boxes, so the phone works
fine.
> If there's RF EMI into the equipment from a hostile environment, then
> I'd be tempted to connect the metallic case to its internal
> equipment's earth via a good RF bypass capacitor, to put the
> circuitry's Faraday cage at earth potential.
That capacitor is what the pcb material would have been had I used it for
an isolation mount. If I were to do it again, I'd put the switchers in
their own box, on isolated mounts, and make an expanded metal faraday
cage around them so they could bounce w/o radiating that noise into the
rest of the circuitry. Hind sight, always 20-05.
> ...
>
> > What about if you have a vehicle instead. Might have equipment
> > mounted on frames that need to be bonded together. If they run an
> > independent battery pack and/or genset then the DC ground doesn't
> > need to touch the frame. But what if the vehicle 12V battery which
> > does have negative connected to the frame also provides some sort of
> > vehicle connection. Say a radio that has a modem that connects to a
> > PC.
>
> One connection, as at the battery would be fine. It's earth loops
> which best radiate EMI, proportional to the area of the loop antenna.
>
> > Logic would dictate you want the DC ground of everything connected
> > to the frame with some bond wires even if just for lightning
> > protection. But now you run risk of ground loops on the 12V circuits
> > interfering with the system battery pack.
>
> I'd put gaseous arrestors on external lines as primary protection,
> followed by MOVs or transorbs further inboard, with some impedance
> between, to further clamp the surge not entirely swallowed by the
> arrestors. But I'd try to read up on that bit first. Grounded
> shielding on external wiring has to help too.
>
> 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
--
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 <http://geneslinuxbox.net:6309/gene>
------------------------------------------------------------------------------
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
