back to filter research

filter design and testing:


EMI interference can take the form of a general blanket of static, raising the "nose floor" of your receive,
but more often it results in a series of "birdies" appearing at evenly-spaced intervals on your bandscope.
These are caused by switched-mode power supplies, used by anything that is trying to efficiently convert one
voltage level to another. (up OR down) Most of this noise appears as "common-mode" interference, and can be
filtered out by a common-mode filter. More stages provide better filtering. The filter will need to be rated
to handle the amount of current you're going to be drawing.

I recently purchased some pre-made filters, filter boards, and filter components to experiment with. CUI filters
are available in a variety of current ratings, priced reasonably, on eBay and s. VFM-15C costs around $40
at Mouser, but I was able to to find them on eBay for $20. They work pretty well, and are single stage common
followed by plus-side differential (series) filtering.

I got better results from a generic made-in-china (YWBL-WH) brand bare filter board (no enclosure) that offers
two stage common mode filtering and is rated for 25 amps. You'll have to put it in an enclosure, but the extra
effort is worth it for the improved filtering.

I tested by powering my 7300 on a battery while receiving on the end-fed long-wire. Noise was provided by a stock
Edecoa sine wave inverter powering a 700 watt ceramic heater. I connected power-pole blocks on the radio and
the battery, so I could run two power connections to the radio. One was direct, the other loop went through a
filter. That allowed me to simply connect or disconnect the direct jumper to compare performance with and
without the filter, without powering the radio down.

Interference was present on all lower bands (tested 40, 80, and 160) and increased when the heater was turned
on, increasing load on the inverter. The CUI filter eliminated most of the noise, but a bit of a whine could
still be heard (though not seen on the bandscope) with no load on the inverter. Turning on the heater made the
EMI show up faintly on the bandscope. The test was repeated with the YWBL, which made the idle EMI vanish
completely, and only faintly be heard under load. YWBL was the clear winner, probably due to its dual stage
common-mode filtering. (it was in a plastic case, not a metal housing like the CUI)




last updated 08/19/2023 at 10:42:57