[time-nuts] Re: DIY Low offset Phase Noise Analyzer (Erik Kaashoek)

[Bob kb8tq]

Hi



> On Jul 8, 2022, at 10:35 AM, Mike Monett via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> Bob, you wrote:
> 
>> Mike. One concern I have with active components as mixer is noise.
>> For an SA I designed only a passive DB diode mixer had  low enough
>> output noise.  Would a PF detector as being  an  active component,
>> not create more noise as output? Erik
> 
>> Yes, you are correct. The only thing with a low enough noise floor
>> for good  phase noise measurements (via the  quadrature technique)
>> is some sort of mixer. Normal digital phase detectors have  way to
>> high a noise floor.
> 
>> Bob
> 
> You are talking about old technology. Old tecnology PFD's were built with
> discrete circuits and probably suffered from crosstalk, deadband, ground
> bounce, VCC noise, and noisy input oscillator signals.
> 
> Modern PFD's have very low noise. For example, the Hittite HMC984LP4E
> digital phase-frequency detector has -231 dBc/Hz of noise and goes up to
> 350MHz:
> https://www.analog.com/media/en/technical-documentation/data-sheets/hmc984.pdf

Hi

The noise floor of the double balanced mixer (used as a phase detector
at 100 MHz) is in the -165 go -170 dbc / Hz range. I’ve used the parts you
are talking about. Their floor is *way* higher.

Bob

> 
> Too bad the price jumped enormously when Analog bought Hittite.
> 
> The MC100EP140 Phase-Frequency Detector has 200 femtoseconds of jitter and
> goes up to 2GHz. That is not going to match the HMC984LP4E but will be
> adequate in many applications:
> https://www.onsemi.com/pdf/datasheet/mc100ep140-d.pdf
> 
> Modern synthesizer IC's have PFD's as the frequency detector and offer very
> low noise.
> 
> You also forget that double-balanced mixers are also very noisy. For
> example, most receivers need a good low noise preamp in front of the mixer
> to get an acceptable noise figure. I am told that part of the reason for
> the high DBM noise is multiple harmonics are generated by the internal
> signals, which combine as part of the output signal.
> 
> Mike
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com