Bruce and Didier, > > The RF amp IC's generally have inferior noise figures and reverse > > isolation (20dB for RFIC, > 40dB for common base stage) than a well > > designed discrete common base amplifier. In fact by > stacking common base > > amplifiers the reverse isolation can be made very high, at > least at low > > frequencies like 5MHz or 10MHz, without a severe impact on > the noise > > figure. HP used a dc coupled common base + common gate + > common base > > cascade in their 8554 RF signal generator. > > NBS used cascaded common base stages in their high (120dB) reverse > > isolation amplifiers. > > If an OCXO buffer amplifier has poor reverse isolation then > a variable > > load impedance will have a measurable effect on the OCXO frequency.
i had the opportunity to measure the output to output as well as the output to input isolation of a diy MAX477 based distribution amplifier very similar to the TADD-1 (but not identical). The output to output isolation was in the order of 75 dB while the output to input isolation was in the order of 90 dB. This is not too bad for this simple design and perhaps more than adequate for most of us time nuts. I have not been able to measure the increase in terms of noise figure but this statement must be read as: The increase in noise figure was less than the noise floor of my equipment. Nevertheless my statement of beliefs is the same as Bruce's one: A single transistor stage may be more easily optimized than a multi-transistor-design. Best regards Ulrich Bangert > -----Ursprüngliche Nachricht----- > Von: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Im Auftrag von Didier Juges > Gesendet: Freitag, 2. Februar 2007 20:45 > An: Discussion of precise time and frequency measurement > Betreff: Re: [time-nuts] Stepping up the output of an OCXO > > > ---- Dr Bruce Griffiths <[EMAIL PROTECTED]> wrote: > > The RF amp IC's generally have inferior noise figures and reverse > > isolation (20dB for RFIC, > 40dB for common base stage) than a well > > designed discrete common base amplifier. In fact by > stacking common base > > amplifiers the reverse isolation can be made very high, at > least at low > > frequencies like 5MHz or 10MHz, without a severe impact on > the noise > > figure. HP used a dc coupled common base + common gate + > common base > > cascade in their 8554 RF signal generator. > > NBS used cascaded common base stages in their high (120dB) reverse > > isolation amplifiers. > > If an OCXO buffer amplifier has poor reverse isolation then > a variable > > load impedance will have a measurable effect on the OCXO frequency. > > > > Bruce > > Bruce, > > In cases where the output signal does not need to be a > sinewave, how would a common base amplifier compare to a fast > comparator and if necessary a digital buffer as necessary to > deliver the necessary power level? If this is designed to > drive a mixer, a square wave may actually be better than a > sine wave (lower conversion loss), and I would venture that > the isolation (at least load independence) in most > comparators is much better than that of a monolithic amplifier. > > Didier > > > > _______________________________________________ > time-nuts mailing list > [email protected] > https://www.febo.com/cgi-> bin/mailman/listinfo/time-nuts > _______________________________________________ time-nuts mailing list [email protected] https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
