[time-nuts] Any reason not to use one power amplifierand splitter for distribution amplifier?

Bill Hawkins bill at iaxs.net
Sun Jan 4 23:58:30 EST 2015

Thanks, Bruce. That does clear it up, although pulling an oscillator
through a FET gate to a 50 ohm cable seems a stretch. If things are not
that simple, e.g., a wiring harness to a front panel selector switch,
then maybe. I'm assuming the source oscillator is well buffered against
the world outside the oven can.

I should have said 10base2, not T, meaning coaxial cable with BNC
connectors and T connectors at the receivers, terminated at the far end.

The allusion to audiophiles had to do with people who pay hundreds of
dollars for a line (mains) cord that has special properties to make the
sound from their amplifier somehow more pleasant. They do this because
marketing told them so, ignoring what goes on in the house wiring to the
wall outlet.

There are people who need to handle time distribution very carefully
(lest they get FTL neutrinos), but most of the list seems to buy their
equipment from eBay.

Bill Hawkins

-----Original Message-----
From: time-nuts [mailto:time-nuts-bounces at febo.com] On Behalf Of Bruce
Sent: Sunday, January 04, 2015 3:39 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Any reason not to use one power amplifierand
splitter for distribution amplifier?

Almost all frequency counters have an internal source which is a
potential means of injection locking an external reference if the
isolation between the internal source and the external source is
inadequate. High impedance taps on a single terminated line ensure that
the isolation between such internal sources and the shared line is
limited by the isolation afforded by the internal source selection
gating/switching of each device.adding or removing a tap invariably
changes the phase shift between the source and each of the other
receivers.The minimum isolation required can be estimated from the
maximum acceptable frequency shift, the resonator Q and internal reverse
isolation between the source output and the resonator Q.
Frequency distribution systems like the Spectracom 8140 with wide range
ADC tend to degrade the source phase noise significantly with respect to
non agc distribution systems.


     On Sunday, 4 January 2015 9:41 PM, Bill Hawkins <bill at iaxs.net>

 Friends in Time,

There's been a large amount of discussion about distribution amps on
this list.
People may be using them just because that's what's done. So I ask you:

What are we trying to isolate? The destination devices do not generate
an interfering signal, n'est ce pas?

The receiving devices do not need to have 50 ohms input impedance if the
source cable is properly terminated, no?

If I use high impedance receivers tapped off a terminated line, how is
this different from 10 base T?
Yes, there will be cable delay between receivers, but how were you going
to avoid that with your distribution amp?
Put another way, why do counters like the Racal 1992 allow you to choose
50 ohm or high impedance at the input?

Please, no "take it on faith" audiophile answers.

Bill Hawkins

More information about the time-nuts mailing list