Stephan, I made me a DIY distribution amplifier based on AD8007s. Mine achieves a typical >90 dB output to output isolation and a typical >115 dB (any) output to input isolation @ 10 MHz. The phase noise starts abt. -140 dBc @ 1 Hz. The wideband phase noise floor is abt. -160 dbc and is reached at abt 4 kHz.
There are some spurs at multiples of 100 Hz of which the biggest is -114 dBc @ 100 Hz. The power supply uses a switcher @ 80 kHz followed by a "voltage refiner circuit". I would have expected to see the switcher frequency in the phase noise diagram but I don't notice anything. Instead I see these 100 Hz based spurs. I have currently no idea whether the 100 Hz is really in the output voltage or whether it couples magnetically from the switcher's primary circuit into the T1-1T transformers that I use at the input and the outputs of the amp. Phase noise Measuerements were done with a TimPod 5330A and level measurements were done with an ANRITSU MS2601B spectrum analyser. High quality double shielded cables were used in the measurements. Best regards Ulrich > -----Ursprungliche Nachricht----- > Von: [email protected] > [mailto:[email protected]] Im Auftrag von Stephan Sandenbergh > Gesendet: Donnerstag, 21. November 2013 14:47 > An: Discussion of precise time and frequency measurement > Betreff: [time-nuts] Isolation achieved by opamp based isoamp? > > > Hi, > > I'm curious about the level of isolation that is achieved by > an opamp based isoamp. I'm referring to ones described here > on Bruce Griffiths' page: > http://www.ko4bb.com/~bruce/IsolationAmplifiers.html > > Anyone has a number for this? > > I've tried googling it, but the results are mostly filled > with the other kind of iso amplifier where isolation refers > to galvanic isolation. > > Regards, > > Stephan. > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
