Hi Getting back to the TCXO ….. Assuming all the data is close to correct, you should be able to do better than the noise level that TCXO is providing. The “sort of -130 something” dbc / Hz noise level is not all that great, even for a cheap TCXO. Since the OCXO reads way lower than the TCXO, it’s reasonable to guess that the test isn’t lying to you. Bob > On Jul 15, 2022, at 7:09 AM, Erik Kaashoek via time-nuts <time-nuts@lists.febo.com> wrote: > > After adding an additional 20dB gain stage the measurements became a bit cleaner and the noise floor without a DUT connected became flatter. Due an extra capacitor in the audio path there is some increased drop off below 5Hz. > > As it was impossible to change the scale of the FFT program used you have to subtract 40dB from all measured noise levels!!!!! > > Noise level of the PNA without DUT connected : http://athome.kaashoek.com/time-nuts/PNA/PN_baseline.JPG > This is NOT the noise of the internal DOCXO as without a DUT connected the mixer can not produce any output. Its just the noise of the first mixer output opamp, which happen to be at least 10dB better than any measured DUT. > > CTI OCXO: http://athome.kaashoek.com/time-nuts/PNA/PN_OCXO.JPG > AR60 Rubidium: http://athome.kaashoek.com/time-nuts/PNA/PN_Rb.JPG > > 10MHz reference output of a Rigol SG: http://athome.kaashoek.com/time-nuts/PNA/PN_Rigol_10MHz.JPG > 10MHz generated by the same Rigol: http://athome.kaashoek.com/time-nuts/PNA/PN_Rigol.JPG > The elevated noise floor due to the DDS is clearly visible. > > And the reason why I build the PNA: > Cheap Chinese TCXO: http://athome.kaashoek.com/time-nuts/PNA/PN_TCXO.JPG > > Next step is conversion to triple supply > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com