Bruce wrote: >Such blanket statements aren't a particularly useful guide unless >calibrated by measurements. >* * * >The input voltage noise @1Hz can't be used directly to estimate the >PN noise at 1Hz offset. My message referred readers to previous list messages for a more detailed explanation, and recommended a particular search term. If you had done that, you would have gone straight to a message in which I explained that one cannot directly correlate noise density measurements with PN because each amplifier implementation has its own specific mechanism of AM-PM conversion. That said, when one limits the discussion to inductorless, monolithic amplifiers, I have not observed anywhere near the quantitative differences between amplifiers (with respect to AM-PM conversion) that it would take to equalize the PN contributions of two amplifiers whose baseband noise voltage densities differ by more than 40dB (as is the case when comparing the MAX477/AD8055 with available low-noise parts). So, yes, when a part has an input voltage noise density *that* much worse than other available parts, one can conclude without detailed measurements that it is not a promising candidate for use as a low-PN amplifier. Best regards, Charles

My point was that the PN performance of a device (eg OPA355) with largish flicker noise can sometimes be a bit better than one might naively expect. The OPA355's flicker voltage noise is 30dB worse than some wideband bipolar opamps. Note: I don't recommend this particular fixed gain opamp for critical applications because its PN floor is relatively high (a bit below -160dBc/Hz with 10 dBm 10MHz input). I originally intended to use it to buffer the output of a ringing parallel LC circuit. However a good discrete enhanced CB stage may also suffice. The state of the art PN noise for an isolation amp as well as most of the MMIC devices in Wenzel's amplifier list appears to be around -150dBc/Hz. It would be useful to have PN plots available for various amplifiers including some with high flicker noise for comparison purposes. I can provide some of these. My instrumentation noise floor (@10MHz) is below -200dBc/Hz if I use the interferometer. The corresponding instrument PN floor is around -170dBc/Hz @1Hz (limited primarily by the splitter PN). Each measurement typically takes 12 hours or so to achieve this flicker PN performance. If I were to find the space for a couple of 10MHz Wilkinson splitters it may be possible to lower the instrumentation flicker PN to below -170dBc/Hz. Bruce On Tuesday, December 22, 2015 01:36:50 PM Charles Steinmetz wrote: > Bruce wrote: > >Such blanket statements aren't a particularly useful guide unless > >calibrated by measurements. > >* * * > >The input voltage noise @1Hz can't be used directly to estimate the > >PN noise at 1Hz offset. > > My message referred readers to previous list messages for a more > detailed explanation, and recommended a particular search term. If > you had done that, you would have gone straight to a message in which > I explained that one cannot directly correlate noise density > measurements with PN because each amplifier implementation has its > own specific mechanism of AM-PM conversion. > > That said, when one limits the discussion to inductorless, monolithic > amplifiers, I have not observed anywhere near the quantitative > differences between amplifiers (with respect to AM-PM conversion) > that it would take to equalize the PN contributions of two amplifiers > whose baseband noise voltage densities differ by more than 40dB (as > is the case when comparing the MAX477/AD8055 with available low-noise > parts). So, yes, when a part has an input voltage noise density > *that* much worse than other available parts, one can conclude > without detailed measurements that it is not a promising candidate > for use as a low-PN amplifier. > > Best regards, > > Charles > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the > instructions there.