[RFI] Filtered connectors for video cable

dgsvetan@rockwellcollins.com dgsvetan@rockwellcollins.com
Thu, 25 Oct 2001 10:33:30 -0500

Pete and All,

Spectrum Control, as well as AMP, Amphenol, and other makers of filtered
Sub-D connectors do have excellent web sites with good tech info on the
connectors.  I worked with products from several of these vendors for many
years, and can attest to the fact that yes, they have poor attenuation
below 30 MHz.  If you look at the values of shunt cap and series
inductance, you can see why.  (Bear in mind that radiated emissions
controls mandated by the FCC do not start until 30 MHz.)  Also, most of the
filters are absolutely guaranteed to render any high speed digital (or
video) signals that you attempt to pass thru them as unusable.  They work
best on power and analog control lines, as well as low speed data.  (Some
of the filtered connectors will allow 9600 baud RS-232 to operate properly,
for example.)   Before attempting to use any of these filtered connectors
on data or signal line applications, you must always be certain that the
line driver can handle the shunt capacitance of the filter.  I know some
folks who got unpleasant surprises because a line driver could not "take a

Back to the video-related noise.  It would help if you could do some
sniffing to determine the main contributor.  If possible, take a female
coax connector, like an SO-239, and solder about a 6 inch loop of insulated
hook-up wire between the center pin and the mounting flange.  (An alternate
approach is to solder about a 1.5 inch piece of straight wire to the center
pin of the connector and insulate the open end very well.  Because of
potential safety issues, I prefer the loop method.) Connect a length of
high quality coax between the "probe" and your receiver.  Sniff the video
card area, the cables between the card and monitor, and all around the
monitor itself.

You may recall that there have been several past postings from people
having RFI problems to or from their monitors.  I agree with KB6B about
your monitor being the more likely culprit.  If the above sniffing proves
him right,  find another monitor (or borrow one) to see if the problem goes
away, reduces, or significantly changes characteristics (such as amount of
specturm occupied).  If you are unable to sniff or there are NO changes
when substituting another monitor, then maybe you should try a different
video card (borrow one from a real good friend?).  Either way, substitution
(in lieu of the sniffer set-up) is probably the quickest way to find the
guilty party and get it fixed or replaced.

73, Dale

Pete Smith <n4zr@contesting.com>@contesting.com on 10/25/2001 07:08:31 AM

Sent by:  rfi-admin@contesting.com

To:   rfi@contesting.com

Subject:  [RFI] Filtered connectors for video cable

In a message yesterday I mentioned that the ARRL RFI book referred to these
devices as a possible solution for RFI caused by the peripheral connections
of a computer.

After reading that, and finding a reference to Spectrum Control, Inc.
filtered connectors in the Mouser catalogue, I went to SCI's excellent web
site and researched the characteristics of these devices.  So far as I can
determine, none of the units they make have very much attenuation at low to
medium HF frequencies.  Their most aggressive filter has only 10 dB
attenuation at 5 MHz, rising to 30 dB at 50 MHz.  Their target seems to be
RFI at television and V/UHF frequencies, rather than HF.

I also received a private e-mail from a subscriber explaining that video
signals will be adversely affected by any direct filtration that would have
any useful effect at low HF.  I'm bccing him on this message, and perhaps
he'll repost that explanation for the general audience, because I think it
would be of wider interest.

73, Pete N4ZR

RFI mailing list