Janne This is a real problem, for some (very high performance) SDRs it is a big problem.
It's mot just the second harmonic you have to worry about; if you have a very big signal on (say) 3.50MHz and another one on 3.51MHz then these will mix in the QSD to give a signal at 7.01MHz. The only solution (apart from a better QSD) is to use really good band-pass filters at the front end, at least that is if you want to build a really good receiver. This doesn't just apply to SDRs of course, it applies to all receivers. It is also one reason why an amateur-bands only receiver can have better performance than a general coverage receiver. regards Grant G8UBN > > Guys, what is your experience of the QSD response to a half of the > LO frequency? SDR to masses part 4 mentions about measurements of > the half of the LO frequency leaking through (RXing at 7MHz and injecting > signal at 3.5MHz) only -49dB. > > If this is true, low pass filters don't do enough to keep the output > of the QSD clean. > > I've used BPF on my earlier builds, but not sure this time since the > SDR to the masses part 4 is the only source that mentions about the > half LO frequency problem. > > Thanks + 73s > > de Janne, OH1GTF >
