Re: #wspr U3S for 2m WSPR: How to remove the frequency-swing at start of transmission #wspr


Hi all,

The Taitien TXETBLSANF-27.000000 worked much better than the Fox 924B I had been using for 2m WSPR TX and decodes.  I'm not completely sure that it's the TCXO that made the difference though, as the chunk of copper I glued to the Taitien ( 12.9 x 6.6 x 1.6 mm) is much bigger than that stuck to the Fox (3.4 x 6.9 x 1.6 mm).  Other than that, everything was the same. 

Before the changeover, I removed the foam I had installed to thermally isolate the TCXO from the rest of the circuit, and tried several TX; no decodes.  The signal trace jumped around and drifted too much, as it had before.  After changing to the Taitien, every TX has decoded, all but one with zero drift.  During the entire time, the house's furnace was slowly raising the room's temperature up from 5 C to about 20 C where it is now.

As noted in some other posts, I have the TCXO towering like a giraffe over the other components on the PCB, on 2-cm scraps of very thin wire.  Since the new TCXO doesn't jump around, even with no thermal isolation, the long legs might not be needed.

The Taitien TCXO seems to adjust its frequency much more gradually than the Fox does.  Since I don't see any jumps or drifts at all, even over several minutes of operation as it warms up, I think the difference is due to the TCXO's operational technique, not the larger piece of copper.

The Fox's center frequency is 12 Hz above 27 MHz.  The Taitien is 44 Hz below.  These are within the manufacturer's tolerance claims.

This TCXO's performance opens up the opportunity to run this unit at ~222.294 MHz!  By the time I procure a suitable receiver, maybe the community will have converged on a frequency to use on this band.


Halden NR7V

Join to automatically receive all group messages.