Am 18.09.2012 23:21, schrieb [email protected]:

First results are looking fine though. For later regulat transmissions i shall use the program, unless it is possible to generate a 12450 Hz TX tone. For the RX, it is not possible to set the audio BFO tone to 12500 Hz :-( Don't know why it seems to be such a problem to add that possibility. SpecLab and VAC has to do the rest of the work. OK, maybe i would need SpecLab anyway, to realise the 2.5 kHz SSB filter...

Sri, do not know why you need 12kHz??

Like many other stations (e.g. TF3HZ, G4WGT and YV7MAE) i do not use a HF TRX at 136 kHz "dial" to receive on LF. We are using a wideband DC receiver with different LO frequencies. For example, here it is 125 kHz, for YV7MAE it is 130 kHz. We also are not limited to a SSB bandwidth but can cover several kHz of LF spectrum. So we can not simply set to 136 kHz and all is fine. A tone generated at 137.5 kHz appears at 12500 Hz in my soundcard input. For TF3HZ it is 7500 Hz and so on.
So what we have to do is to substract a specific frequency in Spectrum Lab and also pass this through a 2.5 kHz wide filter. I have to substract 11 kHz, then the tone comes out at 1500 Hz. The sound output of SpecLab must then be fed to the sound input of the WSPR or Opera program by using a VAC (virtual audio cable software)....
Obviously it works because i got several decodes of PA0A now.

On my side it is the same for transmitting. I do neither use a HF TRX + transverter nor a DDS VFO to generate my TX signal but i'm generating the transmit frequency in my PC, either by SpecLab or now by using Markus' tool which is using a tool by DL4YHF of course :-) Now i am transmitting on 137.45 kHz. This is done by generating a 12450 Hz tone in the soundcard. The tone is fed to my IQ transmit converter (https://dl.dropbox.com/u/19882028/LF/schematics/TX%20IQ%20converter.png) where 125 kHz are "added". So then the signal appears at 137.45 kHz and is fed to the PA.

The 125 kHz signal for the RX and TX are generated by downdividing a 8 MHz xtal signal which is part of a PLL that is locked to a 10 kHz GPS reference signal. This is why i can generate DFCW-180 without any noticable drift. Also the soundcard xtal drift of all PCs is compensated by
this 10 kHz signal.

Now, if i would like to transmit WSPR by using the WSPR program, the program can't do that, i.e. can't generate a frequency on 12450 Hz... I would have to generate a tone on 1450 Hz and feed this to SpecLab using VAC and then add 11 kHz and feed this to the IQ converter. I would work but quite complex :-)
Anyway the way to generate CW, QRSS, DFCW, slow HELL or what ever in SpecLab is the perfect way to go for me because i can use all the other important functions in SpecLab, e.g. generating a quasi-GPS locked tone on a desired frequency, using all the periodic and scheduled actions and so on...

As long as some reports are coming in, the time delay cannot be to high. However i think the displayed SNR suffers from this delay.

Dont't think so: WSPR compares power during times of no signal with those with signal.

OK, thanks.

73, Stefan