Gene,

I think you know all this but I wonder if you implemented it.....

I wonder if the issue is not with grounding.   If a signal moves
relative to ground maybe it is the ground that moves?   The only way
to avoid a moving ground is tomato absolutely certain that it is
impossible for current to flow through a ground line.  The ground
moves because volts is current times reactance (Ohms law) Yes
"reactance" not resistance because we are dealing with AC.

Every positive power supply line needs a dedicated return line and
keep them inside the same cable assembly or even twisted.   The fields
cancel that way,  no common ground return paths

The only way I know to prove there is not current in a ground line is
to remember Kirkoff:  Current flows in a loop.  Make the ground such
that there are no loops.   This is VERY hard to do an requires things
like lifting the connect to a shielded cable on one end and

I think the most serious problems are common ground current returns

My experience with this is mostly with professional level audio where
we try and keep noise at the -100 dB level.  It can be done even with
switching mode power supplies
Noise can also readout but never through a grounded metal box

A good thing to invest in is a lab grade bend power supply.  Even the
cheap $100 units are decent.  When you are building your prototype
systems use that known-good power supply.  I think you said you were
using a re-cycled computer power supply.  These are the worst things
on Earth.    I use one to power my Lithium battery charger

If the grounding is done well and you've fixed radiated EMI by maybe
using all balanced power cables and metal boxes then LC filters really
work as designed.  A large iron core choke will seriously attenuate
noise


All that said, most of my motion control work is robots and I use
battery power.  The battery powers a set of switching mode DC/DC
converters and these power the electonics these switch at the MHz
level and are easy to filter  the other trick is not using one central
power supply but placing smaller DC/DC units near their loads

On Sat, Dec 31, 2016 at 8:54 AM, Gene Heskett <ghesk...@shentel.net> wrote:
> Greetings all;
>
> While I like the SPI interface as its faster that even a parport, I am
> finding it quite susceptible to mistakes cause by emi from a switching
> psu's typical of stepper motor psu's. These things it seems can radiate
> back out the line input connections, while switching at 17 kilohertz, is
> making the whole system bounce around with high frequency switching
> noises of 5 or more volts with rise times in the sub 5ns range, 2 feet
> of ground braid away from the common bolt, totally tearing up the data
> integrity of the spi bus.
>
> With the z motor psu unplugged, the system runs fine although the noise
> from the 5 v 4a computer and interface supply is still in the 1.5 volt
> range, measured from the common ground bolt to a grounded pin on the
> 7i90's center i/o connector. A filter of sorts, with too long (about a
> foot) interconnection leads from that 5v supply to the filter reduces
> that noise to about 400 millivolts, so its definitely helpfull. So I
> have bought 2 of corcoms 20 amp rated brick wall's, should be here the
> end of next week. $24/copy.
>
> So my question is:
>
> If I mount one adjacent to the psu's for the motors, and feed both psu's
> from that filter with lead wires in the 2 or 3 inch range, is this going
> to cause the crap from one supply to be rejected into the other supply,
> possibly damaging it?
>
> Both supplies carry universal, up to 250 volt input ratings. In fact I'd
> feed them 254 from the wall plug, except the inputs are labeled L & N,
> not L & L. Being a std residential hookup here, my 254 is centertapped,
> balanced to ground. 127 vac to dirt ground from either leg.
>
> The 2nd, I will strip room for it on the inside of the lid, so that the
> little supply also has very short leads to the filter.
>
> I quit adding a third one for the 2nd supply because I an running out of
> mounting space for it in this box. So I almost have to make one filter
> do for both supplies.
>
> 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



-- 

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

Reply via email to