[time-nuts] OP-Amps for 10MHz distribution...?

Bob Camp lists at rtty.us
Wed Feb 29 17:25:36 UTC 2012


Hi

For a variety of reasons, you are trying to match a 50 ohm line in a
distribution amplifier. The simplest reason is that mismatch adds
reflections and can degrade timing. 

You *could* use transformers or filters to transform to higher impedance,
but they add phase shift. Phase shift is a problem if it changes (over time,
temperature, vibration, phase of the moon...). Change in phase = change in
time = signal is degraded.

Ideally you are only dealing with a single frequency. In practice you have
harmonics. Since filters are out, low distortion and clean signals is a
desirable feature.  

A reasonable question is always - how good do I need to be? If you are
plugging a cable into the back of your 5334 and setting the gate to 1 second
- not so good. If you want to check your Hydrogen Maser to see if it's doing
ok over night- you need to be very good. The difference is likely to > five
of magnitude. 

Muddying this all up is a duality between time measurement and frequency
measurement. Noise power spectral density makes a lot more sense to most of
us in the frequency domain. Many of our measures of stability have time
worked into them. They also take place in a region where frequency is a bit
awkward to use (10 micro Hz ...). It's also a region where wideband noise
floor is not as big an issue as flicker noise. 

Lots of fun.

Bob 

-----Original Message-----
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of Bill Fuqua
Sent: Wednesday, February 29, 2012 10:15 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] OP-Amps for 10MHz distribution...?

Low noise voltage at the cost of noise current which is around 1000 times
that
of low noise JFETs.

The discussion suggest that opamps contribute to the sideband phase noise
of the signal. I am interested in the mechanism that adds this phase noise.
It would have to be a small shift in either gain (changing Miller 
capacitance) or
an internal capacitance in the opamp.
   I am new to this group and have some catching up to do.

>I know of one op-amp that comes close to 1 nV/rtHz at 10 Hz and being
>capable of useful operation as a 10 MHz distribution amplifier -- the
>ADA4898 (1.2 nV/rtHz at 10 Hz, 4.3 nV/rtHz at 1 Hz).  These are
>wonderful parts, but the large signal frequency response with a 100
>ohm load is less than desired for a 10 MHz distribution amplifier.


73
Bill wa4lav




_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.





More information about the Time-nuts_lists.febo.com mailing list