[Elecraft] K2 BFO drift problem

[Phil Wheeler]

Good job of troubleshooting, Mikael. It's really 
good you found this rather obscure problem. Who 
would have guessed?

Perhaps there was a bad lot of rubber with 
contaminants that you particular bumper was made from.

Mainly you now have the problem identified and can 
correct it with a new part.

Phil w7ox

On 4/14/14, 11:15 PM, TheMG . wrote:
> Wow, you'll never believe what I just found out!
>
> There is absolutely nothing wrong at all with L33. It turns out it's
> the rubber bumper that is at fault!
>
> I found this out by pressing down on the stem of the bumper with the
> plastic tuning tool, instead of pressing on the inductor. Guess what?
> Frequency started to drift!!!
>
> On a hunch, I decided to get my multimeter and measure on the bottom
> of the rubber bumper. It's slightly conductive! Starts at about 3
> megohms and if I hold the multimeter probes in place the resistance
> slowly starts to drop down to 1 megohm and below. After a couple
> minutes holding the probes in place about 1/4 inch apart on the bottom
> of the bumper, the resistance got down to 500k!
>
> Since the solder pads for X3 BFO crystal are underneath the rubber
> bumper, it is effectively like putting a resistor in parallel with the
> crystal, which obviously has undesirable effects on the operation of
> the BFO oscillator.
>
> The slowly decreasing resistance as the meter probes are held in
> contact with the rubber is consistent with the fact that the BFO
> drifts over time.
>
> I suspected contaminants on the surface of the rubber, but after
> cleaning it with iso alcohol, there is no change. Something in the
> rubber is causing it to be conductive.
>
> I wonder if somewhere out there, there are hams with K2s experiencing
> BFO drift related to the same cause? If I got a conductive rubber
> bumper in my kit, it's highly likely someone else has also.
>
> Anyways, it's getting a little late right now. I'll communicate my
> findings to Elecraft tomorrow (in case they aren't already reading
> this).
>
> 73,
> Mikael
> VE8MT