Shielding and EMI revisited ( was number of conductors ( was

Hi Jan, all,

got all the switches in place and the solenoids and such and am now
trying to get a section of cable track.

I found MECI has 16g, 12 conductor shielded wire. this would be fine
for my project.

1 section for the 3 steppers
1 section for the solenoids and switches

then run the router speed conttrols in home depot unshielded wire.

regarding EMI, I was thinking of putting on a battery back-up for the
PC as I have one laying around.

that would not protect the Gecko's or breakout board from stray
signals. I tried to use the power supply on there once and it did
not like the application.

I do have some EMI filters from some old stuff so I could put that on
the router line. it exceeds the router current, but not the Gecko's.

That leaves me with the router line noise contained. (hopefully)
the PC protected, and the Gecko's getting what's left. since I have
been running this system without any protection on a standard 110 AC
motor, I assume the arrangement would be fine.

any comments anywone ?

Dave

--- In CAD_CAM_EDM_DRO@yahoogroups.com, JanRwl@A... wrote:
> In a message dated 5/4/2003 9:36:25 AM Central Standard Time,
> davemucha@j... writes:
>
> > Now it sounds good, but I'm not sure if mixing all those in the
same bundle
> > is too bright. The router speed contorls are the cheap import
style that I
> > think are just a voltage change and run the motors on a variable
DC
> > (rectified only, no caps). to it might be running the motors on
DC ? which
> > might or might not make a difference on the bundling.<<
>
> Dave: To control the voltage/current of DC to a router, even a
fairly small
> "Trimmer router", you would need a large and QUITE expensive
variable
> resistor. The "cheap" style of "dimmer controls" use SCR-type
semiconductors
> to "chop" some of the sine-waves off as AC passes into them, and
these will
> NOT work on DC! AND, this type of SCR- or Triac "dimmer" makes
electrical
> hash (aka EMI noise) to beat the band, especially "cheap" ones!
The
> trigger-controls in a VSR drill is this same kind of thing. It
varies the
> "time ON" vs. the "cycle time" of the AC, not the "voltage" per se.
> >
> > Am I getting too far off base ? would I be better using seperate
> > bundles/cables ? should I seperate the 110 from all the other
signals ?<<
>
> Yes. Possibly. However, if you could SHIELD the signal-wires
separate of
> the AC power-lines, these could then be all in the same sheath, if
the shield
> were properly GROUNDED.
>
> For a large CNC device I once built, the ONE cable-carrier
(Gortite!)
> contained three-phase, 240 VAC, 90 VDC "chopped" for speed-control
of a
> quill-up/down motor; and limit-switch lines, all of which ONLY
controlled
> tiny DC relays which then controlled the "5 VDC logic signals".
There were
> NO "logic signal lines" ain that gob of conductors, and all worked
fine!
> Whew!
>
> Now one of my home-brew lathes (see "Home Brew..." in PHOTOS) was
constructed
> before I knew how to tell time ('84), and it worked OK in the
beginning.
> Then I had occasion to plug-in a VSR in the same AC circuit; the
lathe was
> turned ON, but was at rest. The instant I pushed the trigger on
the VSR
> drill, it began to "buzz" and the lead screws merrily went both
ways
> intermittantly with no apparent logic at all. This instantly
stopped when I
> let go of the drill-trigger, and resumed when I'd press, again!
So, I go
> back to my local surplus dealer and get a large-size (what, 5KVA?)
EMI
> line-filter---ya know, one of those things neatly packaged in a tin-
plated
> shell, and then the lid sweat-soldered all around. Put that on the
AC-mains
> feeding the lathe, and it behaved from that time, on!
>
> While I'm at it: Then I built the second home-brew lathe, using a
Superior
> Electric driver-box and bipolar steppers. That has worked
perfectly from
> day-one, and has NEVER "glitched"! But it has OPTO-ISOLATOR
inputs, and
> built-in EMI filtering on the AC-line coming in. For "shop
machinery" with
> AC motors, especially ones with "speed-controls" and/or "logic
lines", you
> just GOTTA do things "fussy" and THINK!!! Jan Rowland
>
>
>
>
>
> [Non-text portions of this message have been removed]