Ferrite on Reducing RFI

By W4AN

(archived by KA9FOX)

I recently posted a question to CQ-CONTEST regarding RFI I was experiencing in the shack generated by a delta loop I use on 80 meters. The RF would light up a few of the boxes in the shack, mostly the TopTen Device boxes. Here are a few of the answers I received...

From George, W2VJN:

The split cores to order from Amidon are:
2x-43-251 for 1/4 inch cables and
2x-43-141 for 1/2 inch cables.

You can also use non split cores as they are cheaper:
FB-43-5621 for 1/4 inch cables and
FB-43-1020 for 1/2 inch cables.

There are also toroidal cores made of 43 material, preferred for HF, which can
be used for winding several turns of control cables on. The inserted impedance
goes up by the square of the number of turns, so you can see the advantage.

BTW, Amadon is a distributor. The cores are manufactured by Fair-Rite Products Corp and are available through industrial suppliers. They also supply kits of various types of their EMI suppression products.

From Pete, N4ZR:

For what it's worth, here's what Amidon's literature says about core material -

43 - "Widely used for medium frequency medium frequency inductors and wideband transformers up to 50 MHz. Very good frequency attenuation from 30 MHz to 400 MHz. toroids and ferrite beads."

73 - "Primarily a ferrite bead material. Very good attenuation properties from 0.5 MHz through 50 MHz. Available in ferrite bead form only."

And under Ferrite cores for RFI suppression ... "The 43 material is a good all-around material for most RFI problems. However, the lower frequencies from .5 to 10 MHz can best be served with the "J" or 75 material. The 77 material can provide excellent attenuation of RFI caused by amateur radio frequencies from 2 to 30 MHz and the 43 material is best for everything above 30 MHz. However, it is still very effective across the entire amateur band but not quite as good as the 77 material. The 73 material is specifically a small ferrite bead material having a permeability of 2500 and can provide RF attenuation very similar to the 77 ferrite core material."

I just built 6 bead baluns using the 73 beads with .2 inch ID - at $4.50 a dozen or 15 cents apiece in 300 quantity the price was sure right. Amidon also sells split beads in the 43 material with .250 and .500 ID that show decent impedance at 10 MHz though they clearly peak higher.

Finally, there's a chart in their literature that shows the impedance for a reference bead. The peak for 75 is about 8 MHz, for 73 material about 25 MHz and for 43 material around 200 MHz. More important for an 80-meter application, the 4 MHz values are: #75 - 27 ohms; #73 17 ohms; and #43 just under 10 ohms.

And of course, just to confuse things, not all of the relevant pieces are available in all materials (at least from Amidon). The .2 id beads (FB-X-2401, where "X" is the material number) are only available in 43 and 73, while the larger and more expensive ones are only available in 43, and so are the various split cores.

From John, WA2WYR:

I just went through this investigation, as I had the same problems you are having. I was getting excessive RFI on 160, 80 and somewhat on 40 meters. I run 1500 watts output here.

Here goes:

The common split ferrite (snap-on type) chokes sold at flea markets etc. are NO GOOD for lower freq. HF attenuation. If you look at the impedance curve of freq. vis impedance, you will find that these number 43 type materials are good for VHF attenuation and offer very little attenuation at the lower HF frequencies.

Therefore now knowing the problem, I went to the catalogs and selected a ferrite material that had a high impedance for the 160 and 80 meter bands. The result was number 75 and number 77 material from Amidon Inc. I bought several sizes of these ferrite cores from Amidon. The only problem is that "SPLIT" type ferrite cores in a snap on arrangement don't seem to be available from any manufacturer. I assume this is due to the fact that the impedance properties would most likely be severely degraded by splitting this material.

Part numbers are:

small ferrite bead, number 75 material - FFB-75B-101 $4.50 PK/12
larger core unit, number 77 material - FB-77-1024 $2.00 EA
very large core, number 77 material - FT-240-77 (great for cables) $9.00
EA

Results:

I choked all my computer I/O cables with these units and I no longer have RFI
problems with my computer.

Address of Amidon Inc.
P.O. Box 25867
Santa Ana, CA 92799

Telephone: (714) 850-4660

From Pete, KS4XG:

I'd go with type 75 material or maybe type 77. But the Amidon folks or others on the reflector can point you to the very best choice for suppressing RFI at 80m frequencies. Get a copy of their data sheets with your order and it will be helpful for next time. I found 1/2 inch toroids made of type 75 material to be very powerful for RF getting into my phone wires from 80m up. You'll need a larger size of course to get many turns with house wiring and rotor cables.

If you've got a decent ground I wonder if bypass caps might help also? If your cables pass through a bulkhead/ground window it should be easy to experiment with this. In combination with the ferrites, bypass caps will effectively create low pass filters. You'd want to use decent quality caps with very low self-impedance like polypropylene. By cascading sections like this, you can make as steep a filter as you like:

---Picture that looked good in email but didn't translate on the web!---

The Amidon charts can show you how much inductance you'll get and then you can select caps of the right value to cut off well below 3.5 MHz using filter design charts in the ARRL handbook or elsewhere. But whether the caps to ground would have any effect depends on the mode of the RF (i.e. whether it has a ground return path). If its common mode, the toroids by themselves are the way to go I think.
One word of advice, though. I've found that the hardest part of fighting RFI is identifying the real paths of the RF and whether they're common or differential mode. If you have done a good experiment to confirm which wires are the path, fine. But otherwise confirming the paths will save frustration later. When all the dust has settled and you've positively, absolutely cured your problem with lots of on-air testing, please publish the solution as well as the "well, that didn't work" failures so we can learn from your experience.

From Bill, K1JKS:

The Rat Shack snap ferrites are very pricey, but they really work on killing RFI in telephone answering machines! one or two on the power supply lead (AC line or P/S brick lead) has solved the universal RFI problem into telephones here.

You can use bypass caps at the rotor and control cables. While you're at it, install MOVs at the same time in case of a near miss by lightning. .01 discs should clean up any RF in the wrong places. Dont forget to check
ur power supply for RFI! put a scope on the P/S output while transmitting on 80M see what u get for AC (RF)

You can get catalogs from Amidon, Palomar and others. Plus there's
always a lot of ferrite at flea markets.