[Elecraft] Suggestions for 10 MHz Signal Source for the K3?

[Paul Christensen]

>  What are the stability and accuracy of these devices in parts
>  per billion (10^12)?  For 80 bucks you're not going to get a
>  rubidium or cesium standard

The $80 Trimble Thuderbolts as well as most GPS-DO units will do 10^-12.  My 
long-term 24 hour average is 10^-13 as reported by KE5FX software.  So, as 
long as the unit is locked to the GPS satellites, you can achieve long-term 
Rubidium accuracy and about one decimal point away from cesium.  Short-term 
accuracy is around 10 ^-11 although one of my Thunderbolts shows 10^-12 at 
start up and stays there.  Locking the GPS-DO is as easy as throwing a tiny 
GPS antenna out the window.  Mine is resting on decorative brick wall and 
fed with about 15 feet of RG-174.  These antennas are plentiful on the 
Internet for about USD $15 each.

True, the GPS-DO together with the K3 will not provide anywhere near the 
accuracy stated above since the K3 will be frequency locked to the external 
oscillator, and not phase locked.  But, the output of the GPS-DO can be 
split and fed to other transceivers (e.g., Icom 77700/7800, ADAT, etc), or 
used as a test equipment reference.  The Trimble Thunderbolt has 
extraordinary phase noise performance as documented by KE5FX, typically 
about 10 dB better than other GPS-DO units in the area of 10Hz to 1 kHz from 
Fc.  So, it may be overkill for the K3, but the device may come in handy at 
a future date where its full potential can be exploited.

Paul, W9AC