Ah, yes. That is a slightly different problem...
The IF is determined by the VFO frequency +/- the BFO, not the nominal filter frequency. Note that you can get either out of the mixer and the function of the filter is to choose which one to pass. In the BITX40, that is 11.998.600 (roughly actual) - 4.998.600 = 7.000.000 (40m). This is filtered on 40m as LSB because of sideband inversion; the only trouble with it is that it will tune in reverse. (I am sure you have noticed this in the BITX40. It tunes "the right way" around 16 MHz with an analog VFO). In a real-world rig, one still has to adjust the BFO (or the IF offset) slightly to clarify things.
Note that by adding another VFO (which is a simple thing to do with a
DDS), one can also get other bands: 11.998.600 + 2.002.200 = 14.000.000
(20m). On 20m, it would be direct USB, and the BFO would have to be adjusted to reflect that.
Eliamady recognized this on his board by providing 3 caps with diode switching in his BFO. One can tune one to LSB, one to USB, and one to CW (or digital if the bandwidth of the filter were good enough). The VFO was off-board, and could be anything one designs for it. Note that one is tuning for an IF at the actual slope or position in the passband, nothing else. One tunes below the center frequency for USB and vice-versa for LSB for IF+VFO. For IF-VFO, inversion takes place.
But, you say, isn't the 11.998.600 frequency below 12 MHz? Yes, it is. But sideband inversion takes place because the IF is more than the VFO! If it were the other way around, the result would be USB, and one would tune the signal to another slope in the filter to pass it. Remember that the actual received or transmitted signal is determined by the mixer output; it is there that the selection is made. Both sidebands are actually available in the chosen IF --- the filter decides which one is passed through. If the received signal is LSB (say) then the USB is gibberish and nothing is lost in the choice. But it also means that we can choose USB if that is where the real signal is...or DSB for that matter.
All the 20m BITX rigs I have heard about use an IF+VFO design.
It is very difficult to make a wide-ranging analog VFO, even one 1 MHz wide. The only widely-used practical one that has been devised uses an IF around 9 MHz; 9+5 = 14 and 9-5 = 4, hence a 20m/80m pairing. And that one covers only a portion of 80m. One covering several MHz is going to be a nightmare design. But a DDS provides a relatively simple solution...provided one can get the BFO right.
john
AD5YE?
---In
BITX20@..., wrote :
Thanks John
I already have the 5351 running as a VFO and I may in the future also have the 5351 run the BFO.
In my sketch I assumed that the "BFO" value meant 12 MHZ, the IF. It should be the actual BFO frequency, in my case 11.9986 MHZ. This would explain why my readout was off slightly.
73 Ken
