denbeste@cc5.bbn.com.BBN.COM (Steven Den Beste) (09/30/87)
[FCC, this line's for you!]
A few years ago I helped design a logic analyzer at Tektronix. In order to get
nice clean setup-and-hold specs at high speed, it was necessary to pump the
sampling clock up to the probe head, where a special purpose hybrid did signal
conditioning to send nice clean square waves down to the LA instead of ugly
dirty ramps and slopes.
We used shielded twisted-pair differential drive cables - which is as clean as
it is possible to get. Each strand of a twisted pair carries the same signal,
but in opposite directions, so they generate cancelling magnetic fields.
Any residual is mostly cancelled by the shield.
One day, after dealing closely with the FCC, our boss came into the area and
said (paraphrased - this was a long time ago, you understand):
"The only way we can make the FCC spec is by changing the universal electrical
constant. Anyone have any idea how we go about doing that?"
They ultimately proved to the FCC that it was physically impossible to meet the
spec, and got them to loosen it.
--
Steven C. Den Beste
Bolt Beranek & Newman, Cambridge MA
denbeste@bbn.com (ARPA or CSNET or UUCP)
harvard!bbn.com!denbeste (UUCP)
I don't think BBN cares what I think about this stuff.