[GPS_Standard] 5MHz rubidium source

[Bertrand, VE2ZAZ]

Hi again Mark,

One limitation I see comes from the fact that the circuit does not use a
dedicated DAC for the tuning voltage, but instead a PWM which averaged DC
output level will vary as a function of PIC micro supply voltage. Remember
that the MC3487 output driver chip sits on the same supply rail as the PIC
micro.
This means that a change of loading on the reference outputs will affect the
supply line to the PIC. Not a lot, but enough to spoil your rubidium
stability.

For simplicity reasons, I elected not to split the supplies on the system.
If I had to make only one change to the PCB design, I would isolate the PIC
micro using a dedicated 78L05 regulator. This might be part of a future
board re-spin...who knows.

If you use an oscillator that has a bipolar tuning voltage (centered when at
0V DC, and this is the case with the HP oscillators), you can minimize the
above drawback by adjusting (if possible) your oscillator in such a way that
the sweet spot will sit around 0V. It can be mathematically demonstrated
that the supply variations cancel out at 0V DC.

Have fun experimenting...

Bert, VE2ZAZ


----- Original Message -----
From: "Mark GM4ISM" <gm4ism at blueyonder.co.uk>
To: <GPS_Standard at mailman.qth.net>
Sent: Thursday, November 23, 2006 10:46 AM
Subject: Re: [GPS_Standard] 5MHz rubidium source


Hi all,
Indeed a Rubidium standard is significantly better frequency stability than
any OCXO, but still they drift with age.
They are usually considerbly better than crystal based systems in phase
noise too.
The use of GPS to correct for aging has become commonplace in professional
applications. I have actually just got a quote for one such system for
work.. you dont want to know how much!

My standard has a quoted stability in 24hours of 2 x 10E-11   but only 1 x
10E-10 over 1 month
with very long avaraging on the GPS standard controller, and a documented
control range of  0.16Hz. it may be possible to correct for this long term
drift.
my understanding of the hardware limitations is somewhat stretched though,
but with a granularity of DAC step size calculated for this of just less
than 1 x10E-5 Hz, this means 2 x 10e-12  for a 5 Mhz source
so if I get the settings right i may be able to approach long term accuracy
similar to the short term stability.

Not that I need it HI
But it would be nice to know that I was  'on frequency'
redards
Mark GM4ISM





______________________________________________________________
GPS_Standard mailing list
Home: http://mailman.qth.net/mailman/listinfo/gps_standard
Help: http://mailman.qth.net/mmmain.htm
Post: mailto:GPS_Standard at mailman.qth.net