To: mail11:;-at-cimcad.enet.dec-dot-com (-at-teslatech)
Subject: Radiated EMI
From: I am the NRA <pierson-at-cimcad.enet.dec-dot-com>
Date: Tue, 17 Jan 95 17:03:50 EST
>Received: from inet-gw-1.pa.dec-dot-com by csn-dot-net with SMTP id AA19754 (5.65c/IDA-1.4.4 for <tesla-at-grendel.objinc-dot-com>); Tue, 17 Jan 1995 15:12:37 -0700
>>After the base wire, the next sore spot is the spark gaps. The
>arc does produce some RFI. Fortunately the gaps are easily
>sheilded if RFI is a problem.
>I plan to build a rotary spark gap. It should be relatively easy to build a
>Faraday cage around it.
OK. This area (emi) is waht i get paid (sometimes) to do. Putting
a faraday cage (aka shield box) around something isd a good, neccesary,
first step. Sometimes it solves everything. Sometimes it does not.
Usually (always, in the case of a spark gap) wires have to enter and
leave the cage. ASSuming there is EMI INSIDE, it will conduct on
the wires, bypassing the cage, to some extent and radiating OUTSIDE.
The fix is easy (and i think Richard may have mentioned it, its worth
mentioning again, imo....). bypass caps, and maybe some rf chokes.
ASSuming the coil secondary is running at 100KHz or below, it is
possible to filter these leads heavily for frequencies above that. And
above that is where the neigbors will "hear" you. Small caps, rated
for the primary voltage, and above. Big caps can be a bad idea, they
will start bypassing the coil currents/resonant currents. ALSO, the
stray effects (stray resistance, stray, inductance of large caps are
large, and interfere with their proper functioning as bypasses.
Exact values are implementation dependent, on the order of 300 uuF.
(Richard, does that sound right) to start with. rough schematic
(I HATE ascii art... 8)>>)
Where L is an inductor (10-15 turns, air core, not much is needed)
C is the cap(s) tied toi the cage.
: is the cage wall.
( I only drew one half of the gap and a small segment of the wall.
Serious people run the L literally throgh the wall.) SHORT leads are
important. Caps too small are probably better than too big.
I would not START planning to do this, unless planning a large coil.
Even then, i might well add it only as neccesary...
Any books on EMI control.
Don White publishes a whole shelf full. Not that oen needs the whole