I have a very simple new Rheem oil furnace and A/C system with a PSC Fan.
The thermostat is my existing Nest Heating and cooling one. It's common
terminal is used, so connections are 24VAC, Common, Fan, Heat, A/C.
My symptom is that the fan and furnace will initiate normal heat timing
cycle (burner and fan) as if commanded by the thermostat when I run high
power for about 4 seconds or more on the HF bands. CW characters do not do
it. The antennas are coax fed and well matched.
My HVAC installer/Service guy and I believe we have tracked the problem
down to the thermostat wire acting as an antenna and routing enough RF into
the fan/controller board to react as if the thermostat called for heat.
This belief is supported by observing the fan and heat indicator lights on
the board flickering and then turning full on (during the cycle) as well as
observing RF at the input to the board with an oscilloscope. CW characters
cause flickering but do not trigger a heat cycle. A voltmeter check
indicated no 24VAC on either the fan or command input terminals at the
board, ruling out the thermostat sending false commands in reaction to RF
at its end
Bypassing the fan command input terminal with 0.01uF to Common at the board
suppressed fan (no lights). It took 2uF to similarly kill the heat.
>From the above we concluded that RF on the thermostat wires was causing the
thermostat interfaces on the board to react as if demand commands had been
sent.
My HVAC guy was not comfortable with using 2uF as a long term solution
because the thermostat would look into an uncharged capacitor each time its
electronic switch closure operated. I let it be his call since he will be
responsible for warranty service.
Thus, we decided that using intermediate relays for isolation on the
thermostat wires at the control board input would be a safe
solution. Before we do that though, I'd like to try some snap-on ferrites
cores at the control board end to see if they would present a high enough
impedance to reliably kill the errant RF. I might also try some on the
line to break up its RF length.
Can the group recommend a specific manufacturer and model of core(s) to
try? I'm guessing 31 mix since I'm only interested in 1.8-30 MHz.
TIA and 73,
Dan
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|