31 Jul 2012

Sub-harmonic mixer based 10m WSPR transceiver


My favourite HF band is 28MHz and I often put my FT817 on and run WSPR in the background whilst getting on with other things in the shack. You may be surprised how often good DX reports are received. This ties up the FT817 so I have been rethinking how to make a very simple WSPR transceiver for 28.1246MHz.

One neat idea is to use a 14.060MHz crystal (cheap from GQRP) pulled up to 14.0623MHz (it is possible) and use this to drive a Polyakov sub-harmonic mixer, which needs half frequency drive, in a simple direct conversion receiver feeding a PC sound card via an isolating transformer. The same oscillator, doubled would be fed into a single balanced DSB modulator and a small QRP PA stage. The resulting WSPR signal would appear as a DSB signal , not SSB, so half the power would be wasted but who cares with milliwatts?

The resulting full WSPR transceiver would be extremely small and simple, need no VFO or VXO and could be left running almost continuously when the PC is switched on. This would be ideal with a small netbook PC.

The idea is not new: see for example http://www.qsl.net/d/dl1gsj/html/qrssrx30.html .

Sounds like this is my next project......

10 comments:

PE4BAS, Bas said...

Sounds interesting! 73, Bas

YO5PBG - Adrian said...

Hello,
I don't understand exactly why are you using 2 mixers, not one for both RX and TX? Is it Plyakov Mixer better on RX side? IS it because of havingbetter RX - TX separation at the antenna conecion?
73! Adi

Roger G3XBM said...

Adi, The RX Polykov mixer needs half frequency injection whereas the TX mixer needs final frequency injection. Both mixers will only need a pair of 1N4148 diodes and a transformer, so inexpensive to implement.

YO5PBG - Adrian said...

Roger Roger :-) but i saw already a multiplier in your schematic for TX chain and I was thinking you use a balanced mixer for TX. it is also simple to use it for RX the same balanced mixer with only 2 diodes and a tri-wound transformer...
But i really wait to see how Polyakov mixer works in your test RX...
Adi

Roger G3XBM said...

Adi, you are right that just one mixer could be used, but I suspect it is as easy to make a second mixer (with less LO leakage out of the antenna for the RX side) as to diode switch between RX and TX around a single mixer.

YO5PBG - Adrian said...

Yes Roger, that is one of the questions in my first message, if it is about the LO leakege you use two mixers. Now it is very clear. About Polyakov mixer i still wait your results, feeling about using that etc.. I wanted to use this mixer in a VHF transverter so i will look for details.. :-)
Good luck with your project!
Adi

Roger G3XBM said...

With the RX injection at half frequency, even with simple RF front end selectivity there should be very little LO leakage with the Polyakov mixer.

Project is 2nd on the list as I've been asked to submit my 472/500kHz transverter design as a magazine article and it needs a bit more work.

Anonymous said...

I used a ne612 - just one, bidirectional with just a single transistor PA for a simple wspr trx a couple of years back. it worked ok. only problem was the usual hash coming from the pc.

Tony

Bob G3WKW said...

Why use a high powered netbook when you could run it on a Raspberry pi using 3.5 watts. I have a web interface running on it now so it runs stand alone and I can still watch the waterfall when I want to. Mind you it did crash unexpectedly after 19 hours continuous WSPR on 40m so I might have some more work to do yet!

Roger G3XBM said...

Excellent Bob. Don't you still need a display though?