VAX 6000 3-phase conversion (was: Re: Three Phase)

From: ajp166 <ajp166_at_bellatlantic.net>
Date: Sun Apr 22 14:45:07 2001

From: Clint Wolff (VAX collector) <vaxman_at_qwest.net>


>The cages I've seen are made from an expensive copper screen,
>like window screen, but they are for doing FCC interference
>measurements, not shielding. I can get 1/2 inch square fencing
>from the look home improvement center, but it is galvanized and
>quite a bit coarser...
>
>Comments?


First a faraday cage to run a vax is foolish. The FCC notice
refered to is basically a hint about whose reponseability it
is to fix RFI problems should they occur. Often it's not a problem
save for at point blank range.

RE: galvanized screening vs fine copper screening. If you cant do it
right dont bother. As the holes get larger and the resistivity increases
the attenuation decreases. Also as the holes get bigger the frequency
at which the screen becomes RF transparent also gets lower.

So to get an effective faraday screen fine copper screening with
well bonded (often soldered) joints and a door that has RF tight joints
are used. Also any wires that enter or leave have to be bypassed
to avoid it becomming a leakage path in either direction. I participated
in building one and it's a lot of work and after testing to find RF
leaks.

At the other extreme a cage of .500 galvanized will have effect but the
effect is far more limited. to prove it take a roll and make a mini cage
large anough to hold a battery AM/FM radio and see what I can hear
from inside it. Then take a tube and coat it with aluminum foil
and try it with the same radio. You will also find it very difficult to
solder or otherwise effectively bond the joints of a galvanized wire
cage.

Last item: The average system (even a PDP-8) will be much better
in the RFI derby if all the covers are on and the correct cables are
used. The later smaller one box systems are fairly decent if the
case is metal and all the screws/hardware are in place.

The killer items for RFI are SCSI cables, external disks that used
any unshielded cables (BC05s come to mind) or monitors that
used shielded cables but are incorrectly terminated. RS232 wires
that do not use an external shield can be a path to propagate noise
out of and otherwise tolerable box as well.

Worst offender awards:

    TRS80 with EI... ghastly both radiating and affected by nearby
    transmitters. (1w 144mhz ht at 4ft would crash the average TRS80).

   Some of the Apple][s

   Early S100s with non-stripline and unterminated busses (altair!!!!).

   SBCs, no case at all most often.

   Many of the early all plastic cased (no internal conductive spray
coat)
   with no internal shielding systems.

   Many (most) CRTs that are not metal boxed internally are noisy.


Allison
Received on Sun Apr 22 2001 - 14:45:07 BST