Here is the schematic for the little QRSS (visual) transmitter that is currently rockin' Europe's 30 meter band with an AWESOME 20 milliwatts of SLOOOW FSK. (As I type, it is 0415 UTC, 0615 local, and the first signs of my signal have just appeared on the ON5EX grabber up in Belgium.) The FSK modulation comes from Hans Summers' multivibrator circuit (see earlier posts).
It was a lot of fun to take this thing very quickly from LTSpice, to the workbench, then to the antenna, with Johan's grabber providing instant feedback. This started out as a one-stage Colpitts oscillator transmitter. But I needed more stability. Indeed, the separate oscillator with the source-follower buffer makes it much more stable. Before, any adjustment to the antenna tuner shifted the frequency. At one point I even suspected that wind blowing the antenna was shifting the frequency -- we are talking about a band that is 100 HERTZ wide, so even a few hz of instability is noticeable. But I find that crystal ovens and other extraordinary measures are not really necessary.
I had one unusual problem with this little rig: As I was doing my initial tests, I noticed that the output signal was sort of jumping up and down. The problem was in the PA. I isolated the problem to the base circuit. At first I thought that some small blob of solder was intermittently messing up the bias voltage (that's quite possible here in the N2CQR lab!). But no! It was that 4700 ohm resistor. It was bad, and kind of intermittently bad! I never had a resistor go south on me like this. It is an ordinary 1/4 resistor. It is not dissipating a lot of power.
I'll keep it around 10140010 today. Check it out on Johan's grabber:
http://www.on5ex.be/grabber/grabber.html
Look for a horizontal lines with little bumps (about 4-5 per minute). That's me.
An Unintended Post-Mortem
-
Many of us wind up in a similar situation, I'm sure: we impulse buy at
hamfests—great plans for some big old currently non-working piece of
gear—and then...
1 hour ago
Three transistors! That's quite a lot for 20mW, considering Roger G3XBM's XBM80-2 (which will work on 40 or 30m as well) manages a complete transceiver with just two.
ReplyDeleteI know, I know -- terribly extravagant. And there are two more transistors in the multivibrator! And don't forget the zener diode on the oscillator, and the LED. I hang my head in shame...
ReplyDeleteOn thing I have learned from a hf/microwave designer is avoid to use zener diodes into this kind of Receivers. We want to have the noise level as low as possible.
ReplyDeleteProblem of zener diodes that they are also excellent noise generators.
Beter is to use a small voltage regulator.