[Moon] ts-2000x fan chirp

[richard allen]

When testing my new ts-2000 on some 437 MHz cw satellite signals I 
noticed some frequency variations that were abnormal.  

I was tracking several satellites and correcting for Doppler shift.  
When all is working correctly and the frequencies and Keplerian 
elements are correct this will produce a fairly straight line down my 
waterfall display.  When I saw these abnormalities, I fired up my 
z3801 oscillator and fed it into the ts-2000 first listening to the 
128th harmonic of the 10 MHz output.  The radio was tuned in upper 
sideband to 1279.999000 MHz which would yield a beat note of 1000 Hz. 
 I was glad to see that the radio was only about 150 Hz off but over 
time I noticed a cyclical frequency variation with about a 20 minute 
period.

The first thought was that the z3801 had not yet stablised but after 
doing a new survey and waiting for all to settle down, the signal was 
still varying.  After monitoring for several hours, I checked the 
local channel two  video carrier at 55.25000 MHz and could see a 
small variation on it also. As the channel two carrier shifted down 
by a few Hertz, I noticed that the fan in the ts-2000 had just turned 
on.  

Sure enough, the cyclical pattern matched the fan cycling on and off. 
 I posted an example waterfall on my website.  

You can view it at http://www.rcallen.com/ts2k-fan.gif ...

The plot starts at about 13:00 utc on April Fool's Day but, 
unfortunately, it is not a joke.

>From 13:00 thru about 13:33 shows several fan cycles.  The ts-2000 
was tuned to 1279.999000 and in upper sideband. As can be seen, the 
frequency changes down to 800 Hz when the fan comes on and then peaks 
back up to about 900 Hz at fan cutoff.  While this is well within the 
+-0.5 ppm specification of the radio, it may be a problem for 
operating.

After 13:41, I switched to tracking the Argentinian ham satellite 
LUSAT/LO-19.  It's cw beacon is nominally on 437.12545 MHz and with a 
M2 70 cm eggbeater, peaks about 20-30 dB out of the noise.  The 
apparent frequency smear is from the fact that I was only updating 
the radio's frequency every 10 Hz.  The Doppler shift makes the 
apparent frequency move more than the program can keep up with at 1 
Hz increments.

At marker 6, at about 13:47, the fan came on causing a drop in pitch. 
The satellite peaked in elevation at 13:49:49 at 25 degrees.  The fan 
then went off at about 13:53 shown by marker 7.  The satellite went 
on to set at 13:57.

At marker 8 I switched the radio to 439.999000 MHz USB, the 44th 
harmonic, and monitored another fan on/off sycle indicated 
approximately by markers 9 and 10.

Any suggestions as to how I can stabilse the radio would be 
appreciated.  I think I will first try to modify it so that the fan 
stays on continuosly.  The master oscillator is a TCXO at 15.6 MHz 
and it may be possible to replace it with an external oscillator 
slaved to my z3801.  Some of the shift might also be a voltage drop 
when the fan comes on.

Regards,

Richard W5SXD