开云体育

Hi Philip

Out of interest, could one of these TX boards be used/modified
for use as a WSPR transmitter?

It could, but not easily!

Well, depends how you look at it. If you just look at the board as a plain 30m transmitter, and then decide to interface it to your PC, and have the PC generate the WSPR code and control the transmitter, then I think that's a relatively straightforward proposition.

On the other hand, perhaps the more interesting approach is to consider a WSPR transmitter that is standalone, not requiring a PC to control it. This is one of the things I'd like to look at in future when I have time. Changes that would need to be made, off the top of my head, would include:

1) Shift the frequency up 100Hz (just a matter of adjustment)

2) Arrange for the PA to be keyed, rather than just permanently on. This would need reprogramming of the microcontroller, such that one of the pins (probably one of the pins currently used for setting the QRSS speed) could control the PA. The PA should be switched on for the required 2 minute WSPR transmission frame then switched off for the rest of the 10 minute WSPR cycle.

3) Program the microcontroller to send WSPR. Seemingly the most likely way to accomplish this is to use the DOS program that's out there somewhere on the internet, which will take a set of parameters (i.e. your locator, transmitter power) and output a sequence of codes to control the transmitter frequency. There are four frequency steps in a WSPR message so it would like like 1, 0, 3, 2, 0, 1, 3, etc etc. (i.e. a sequency of 0's, 1's, 2's and 3's).

4) Modify the circuit slightly to allow the four frequencies rather than just 2-level FSK. This can be done using a PWM output from the micro and an integrator resistor/capacitor network between the microcontroller pin and the frequency tuning varactor diode (a.k.a. the 5mm red LED).

5) Seemingly the biggest challenge is the issue of timing. The keyer currently just runs on the microcontroller's internal RC oscillator, which would not be accurate enough for WSPR purposes. It would be necessary to use a crystal for the frequency control, at a minimum.

It would also be necessary to synchronise the circuit operation to the standard 10 minute WSPR cycle, the minute starts have to be synchronous to world time, to a fair accuracy. I think that a properly adjusted crystal oscillator would probably be accurate enough to allow 24 hours of WSPR operation, and perhaps add a button to allow synchronisation once every 24 hours with the actual time. I think the deviation would be within limits.

The alternative is much more complex, if you had a GPS receiver module for example, then that could be used to extract precise timing and synchronisation.

73 de Hans G0UPL

Hi Hans, Philip, fellow QRSS knights,

OE1FM has designed and built stand alone beacons that are running in Africa (V53ARC that I seldom receive).
Design is a bit complex but perhaps some idea could inspire a new work.



73,
Yan - XV4TUJ.
---
Yannick DEVOS - XV4TUJ
(Blog in french)
(web page in english)

--- In QRPLabs@..., "hanssummers2000" <hans.summers@> wrote:

Hi Philip

Out of interest, could one of these TX boards be used/modified
for use as a WSPR transmitter?

It could, but not easily!

Well, depends how you look at it. If you just look at the board as a plain 30m transmitter, and then decide to interface it to your PC, and have the PC generate the WSPR code and control the transmitter, then I think that's a relatively straightforward proposition.

On the other hand, perhaps the more interesting approach is to consider a WSPR transmitter that is standalone, not requiring a PC to control it. This is one of the things I'd like to look at in future when I have time. Changes that would need to be made, off the top of my head, would include:

1) Shift the frequency up 100Hz (just a matter of adjustment)

2) Arrange for the PA to be keyed, rather than just permanently on. This would need reprogramming of the microcontroller, such that one of the pins (probably one of the pins currently used for setting the QRSS speed) could control the PA. The PA should be switched on for the required 2 minute WSPR transmission frame then switched off for the rest of the 10 minute WSPR cycle.

3) Program the microcontroller to send WSPR. Seemingly the most likely way to accomplish this is to use the DOS program that's out there somewhere on the internet, which will take a set of parameters (i.e. your locator, transmitter power) and output a sequence of codes to control the transmitter frequency. There are four frequency steps in a WSPR message so it would like like 1, 0, 3, 2, 0, 1, 3, etc etc. (i.e. a sequency of 0's, 1's, 2's and 3's).

4) Modify the circuit slightly to allow the four frequencies rather than just 2-level FSK. This can be done using a PWM output from the micro and an integrator resistor/capacitor network between the microcontroller pin and the frequency tuning varactor diode (a.k.a. the 5mm red LED).

5) Seemingly the biggest challenge is the issue of timing. The keyer currently just runs on the microcontroller's internal RC oscillator, which would not be accurate enough for WSPR purposes. It would be necessary to use a crystal for the frequency control, at a minimum.

It would also be necessary to synchronise the circuit operation to the standard 10 minute WSPR cycle, the minute starts have to be synchronous to world time, to a fair accuracy. I think that a properly adjusted crystal oscillator would probably be accurate enough to allow 24 hours of WSPR operation, and perhaps add a button to allow synchronisation once every 24 hours with the actual time. I think the deviation would be within limits.

The alternative is much more complex, if you had a GPS receiver module for example, then that could be used to extract precise timing and synchronisation.

73 de Hans G0UPL