[RFI] Ferrite Loops
Jim Brown
jim at audiosystemsgroup.com
Sat Aug 14 23:00:14 EDT 2004
On Sat, 14 Aug 2004 20:45:06 -0400, Tom Rauch wrote:
>If you have access to an antenna analyzer like the MFJ259B,
>you could measure them. Otherwise, you may just be wasting
>time.
It is also important that they be measured over the range of frequencies at
which you hope to use them. A wire running through a ferrite (or wrapped
around it) can be analyzed as if it were a series R and L (or XL), but for any
given ferrite material, and any given physical size, the values of R and L (or XL)
will vary with frequency. That's why you must measure over a range of
frequencies.
There's another important set of issues. In general, you are using a ferrite to
make a choke to eliminate RFI at low power levels, you want the ferrite choke to
look mostly resistive. That prevents the inductive reactance of the choke from
resonating with the capactive reactance of a short line and making the problem
worse .
But that may not work well if you are using the choke with a transmitter -- the
ferrite material may overheat, and ferrite materials don't react well to excessive
heat. So ferrites that are used with transmitters, at least at high power, should
look mostly reactive with relatively small resistive component to their
impedance.
The Fair-Rite website (they make most of the ferrite materials that are sold in
the ham marketplace) has their entire catalog online as a pdf. It is a wealth of
information, both data and tutorials, on ferrites. It is required reading for anyone
trying to understand how these very unique and useful components work.
The obvious question is, how do you measure them? Easy. Simply pass a
single piece of wire through the ferrite loop. Connect the two ends to the
terminals of the analyzer (MFJ or AEA), set the analyzer to the frequency, read
the result, and write it down (R and X). Do that for the range of frequencies you
care about -- I typically measure in steps of 1 MHz or so up to 10 MHz, then 2
MHz up to 20, then 4 or 5 MHz steps to 40 MHz.
Then I put the values in a spreadsheet and plot them vs frequency. I also let the
spreadsheet compute Z (the phasor sum of R and jX). As Tom has noted,
neither the MFJ analyzer nor the AEA analyzer knows what the sign of the
reactance is. You have to guess (and the AEA guesses for you, sometimes
correctly).
In general, the impedance of a wire passed through any piece of the same
ferrite material is proportional to the length of the ferrite and the square of the
number of turns (that is, the number of times the same current-carrying wire
goes through the ferrite).
One thing bothers me about the original question, which stated that he hoped to
reduce RFI/TVI caused by the transmitter. I'm not at all sure that a choke balun
(which is more or less that this would be) will do a lot about that (although it may
help to the extent that it reduces common mode shield current).
On the other hand, chokes on the wiring to the equipment receiving interference
may be more effective, and will probably be easier to implement. Multiple turns
around the ferrites sold by folks like Amidon can be quite effective in reducing
common mode current on the wiring of audio systems and on telephone lines.
Another important point. It is VERY important to use TWISTED PAIR cable for
loudspeaker wiring if you have an RFI problem. Twisted pair cable reduces RF
transferred from the cable to the equipment a LOT as compared to parallel wire
cable (zip cord). Yeah, I know the exotic/expensive speaker cables are zip cord,
but the people who market it only know how to SELL it, they don't know anything
about how it works.
Hope this helps.
Jim Brown K9YC
More information about the RFI
mailing list