[RFI] AC Line filters- Direct-Drive Washers

Jim Brown jim at audiosystemsgroup.com
Fri Apr 4 17:43:43 EDT 2008

Dale, WA9ENA, has given you some great advice. Larry, W0QE, made 
an excellent observation about this in a conversation we had 
offline. If the interfering signal is 50dB over S9, you've got to 
do a MASSIVE job of noise suppression to get that down to the 
point where you can hear signals of normal strength on the ham 
bands. An S-unit is "officially" 6dB, so S3 is 36dB below S9. To 
get the noise down to S3, you're looking for 86dB of suppression! 

A REAL 50dB over S9 signal is huge. S-meters on ham rigs are 
notoriously poor for linearity, so the actual signal level may be 
a lot lower. But even if there's a 20 dB error between S3 and 50 
over, you're looking for more than 60 dB of suppression. Such an 
effort is not for the faint of heart. You'll need to start with at 
least that level of suppression on the power line. Once you've 
done that, you're VERY likely to hear trash radiated from wiring 
within the box.

These offending motors are variable speed drives, which are 
essentially driven by DC power supplies followed by circuitry that 
chop the waveform into square waves of varying width, and 
frequencies on the order of 10-20 kHz. It's the harmonics of those 
square waves that we hear. 
As it turns out, Larry has done a lot of looking at this. He 
observes that a typical machine is built to make it easy to work 
on, and the wiring that drives the motor is a big loop. Because 
that loop is large, current it carries generates a very strong 
magnetic field, which is picked up on our antennas. The field is 
proportional to the loop area. To reduce that field we must make 
that loop a LOT SMALLER! 

Bottom line -- you may do the world's most wonderful job of 
filtering the power line and still have a lot of trash radiated by 
that wiring inside the box. If that's what's happening, a possible 
cure is to go inside the box and re-wire it so that the current 
driving the motor is carried by a twisted pair, not a big loop. 
But that might not work if part of the current path is capacity 
coupling via the motor frame and the chassis. You could also 
reduce the strength of those harmonics that we hear by slowing 
down the rise time of the square waves.  

Like I said -- not for the faint of heart. :)


Jim Brown K9YC

More information about the RFI mailing list