Hi Chris,

as WSPR is a constant envelope signal, overdriving the TX Audio would not produce intermodulation but only audio harmonics. With a fundamental near 1.5 kHz, the odd harmonics at 4.5 kHz etc. will be too far outside the passband of an SSB transmitter to cause noticable contamination.

However this may not be the whole truth: If the clipping already takes place in the digital domain (i.e. before the DAC and antialias filter), we’d still have the same harmonics but they could be aliased down with the samplerate. For example, a clipped 1.51 kHz signal contains components at 7f (10.57 kHz) and 9f (13.59) kHz. With 12 kHz samplerate, these could be folded back to 1.43 and 1.59 kHz, and thus create weak unwanted sidebands near the desired output frequency. BTW the very same effect can happen on the receive side when overdriving the ADC.

Best 73,

Markus

25 April 2016

Had a report that my LF WSPR2 signal was dirty with massive side bands

on Sunday. I had a look at my setup today found Windows had decided to

up the audio level from WSPR to maximum into the TS-590 and scoping

the output of the pre amp and frequency doubler before my G3YXM amp

showed it looking very ropey indeed.

I reset it and did a couple of normal WSPR2 transmissions interspersed

with using the WSPR TUNE function, whilst watching on the G3XXM and

G4WGT LF grabbers. I am no expert whatsoever, but it doesn't look bad

now to me. Is this an adequate test to make and to assume things are

now OK? Thanks.

Screenshots are at http://www.chriswilson.tv/signals/signals.html

--

Best regards,

Chris  2E0ILY                           mailto:[email protected]

My        part        time        LF        grabber       is       at

http://www.chriswilson.tv/grabber.html