Keyboard Cables? / external disk RF signal

[was-John McMillan]

In article <461 at limbic.UUCP> gil at limbic.UUCP (Gil Kloepfer Jr.) writes:
>In article <116 at gnosys.UUCP> gst at gnosys.UUCP (Gary S. Trujillo) writes:
>...
>>He says that once you start running external cables around that whole situation
>>can change very drastically.  Have you given any thought to that problem?  Have
>>you measured RF radiation, even by means of noting the effect of the external
>>cable on nearby radio and television receivers?  Same question to jbm.
>
>Ugh...  I've already read John Milton's reply, and my position on it stands
>equally with his.  In the hard disk upgrade instructions I'm distributing,
>I mention briefly the problems with RF interference.  In that brief section,
>I discuss how BIG of a problem it is.
>
>I disagree with John on one point:  I don't think shielding the cables will
>solve the problem completely, or even sufficiently.
...
>  What further baffles me is
>that there are not constantly signals going to/from the hard disk.  Why
>is there constant interference on a TV?  I suspect that the cables are
>leaking other-than-disk signals outside the shielded cabinet. (see below)

Wrapping them in duct tape?  O brother!  What is the attenuation of duct
tape and how was it grounded?  I'm sure EE's will respond to the issues
here, so I'll skip on.

RF isn't the source of all evils.
In one case:
	Some years ago, and at a different company, an adjacent lab
	was troubled by mysterious crashes.  ...Which stopped when
	they electrically isolated their equipment with line filters.
	A scope on the line showed they were pushing enough hi-frequency
	signal down the wire to be causing some sort of interference
	between machines.  (This is what I was told, and accept.)

In another case:
	Two years ago, a friend bought a maximally-cheap clone --
	of Korean manufacture, if I recall -- from an odd little
	shop near his home in Maryland.  He took the first unit
	back because it repeatedly died.  The store, admirably,
	replaced the unit.  Again, it frequently locked up: this
	time the store threw in an upgrade to a faster model --
	for free!
	
	And it still belly'd up.  I was visiting, and -- back at his
	friendly store -- convinced him to buy -- and Tom AIN'T
	easy to get in a BUYing mode -- one of those CRT-platforms
	that contains switched outlets and isolates each component
	on a spike-protected, noise-isolated circuit. (~$40?)  Of
	course,  at the same time the store also replaced the guts
	of the machine (what does their back-room look like with all
	these swapped parts?).  Since then, the clone's been untroubled.

				-  -  -  -

Whew -- didn't anticipate the length of the above! Apologies.

So: neither underestimate the effects of RF propagation from unshielded
cables, nor depend on the INTERNAL isolation of your equipment -- in
either direction!  At home, use a spike-arrester with built-in noise
filters to minimize line-carried interference.  The TV interference
on your neighbor's set may be RF, but on YOURS it may be house wiring.

jc mcmillan	-- att!mtunb!jcm	-- speaking for himself only