AnsweredAssumed Answered

E4426B, poor AM spectrum

Question asked by solidstate on Oct 2, 2013
Latest reply on Dec 2, 2013 by tabbott
I am using a E4426B and I am in general very happy with it. Signal has good quality. I now was going to measure some amplifiers with AM modulation. 
I have 20MHz and modulate with 1kHz and 90% (values actually dont really matter, its same behaviour with different values)

To find the AM distortion (THD+N) I am using either FAM Modulation analyzer or i give signal to demodulator and then analyze the audio. Anyway, in both cases i get a distortion of around 1%, which seems ok. If i connect the generator to a spectrum analyzer, i have only 19dB distance between carrier+1kHz and carrier+2kHz. This would lead to a distortion of around 10%. I dont understand where the bad spectrum comes from.

Looking at a different signal generator, i have more than 30dB difference between those two frequencies and the distortion is almost the same. 

Is there a phase modulation or phase noise on the signal? This could be seen in spectrum analyzer but not at the demodulated signal. Did anyone observe the same bad AM spectrum with those generators?

Thanks for hints and suggestions.