SDRplay Community Forum

I have just bought a broadband noise generator to use in a sweep system. The output is about -30dB down from the level displayed on a normal spectrum analyzer. This is probably due to the detection system used in the RSP1A so my question is will it be possible to incorporate quasi-peak detection in future releases or is it a hardware issue which we'll just have to live with?

You will have to attenuate the -30 dBm noise level of the Noise Generator, otherwise the Spectrum Analyser sytem will likely overload & give distorted displays. Better go down to - 60 dBm. There is the peak measurement pushbutton on the console, or did I get you requirement wrong?

The Peak button you are referring to is a "Peak hold" I believe -- also called a Max hold . The analyzer shows the noise signal at -58dBm but on my old Texscan analyzer it is -20dBm (checked against a reference cw signal). The detectors in the analyzers is the only thing that can account for the 38dB difference. The difference in level between Raw/Peak/Avg in this case is only 2 or 3 dB.

[img][IMG]https://i345.photobucket.com/albums/p368/rb2k2/rawnoise_zpszho73mdw.jpg[/img][/img]

There is no logical reason why a Spectrum Analyser, with a random noise input, should display a difference of 38 dB in average to peak noise, unless we are measuring a very, very special kind of noise. Perhaps the problem could be with your old Texcan detector.....I suppose you are using the workhorse Noise Generator BG7TBL.

Quasi Peak would only really be useful if the spectrum analyzer has the appropiate EMC bandwidth etc etc

Otherwise you could just calculate the Quasi peak level from the Dbm level and then also to the same to adjust for bandwidths other than the reference EMC bandwidth.

Yes Quasi a quasi peak detector with the appropiate CISPR 3db bandwidth would be a nice feature for pre-compliance work.