[time-nuts] Pictures of Sulzer, PRS10 and FE-405

[Attila Kinali]

Hi,

Last week, I met Steve Krull and got a big package with a PRS10 from
Bob Martin and Sulzer 5.088MHz (no, that's not a typo) from John Ackermann.
Unfortunately, I didn't have yet the time to put them into use, but
instead I did what all good engineers would do, when they receive
new toys: I took them appart. And of course, I took picture:
http://time.kinali.ch/OCXO/Sulzer/
http://time.kinali.ch/Rb/PRS10/

A few months ago, I also received a FE-405 from Corby, which I, of course,
also took appart:
http://time.kinali.ch/OCXO/FE-405/

The construction of the Sulzer is quite interesting. It's been a long
time since I last saw a non-printed circuit board, and quite a bit
longer since I had the opportunity to hold one in my hands. From the
construction of the double oven, it's quite obvious that the engineers
at Sulzer didn't have our tools in simulating ovens and there are a
few points that lead to sub-optimal heat-flow. But the high thermal
isolation (the devar flask which covers the oven part is not shown)
including the outer can, which effectively makes the whole electronics
and metal frame part of the oven (kind of an outer, third "oven" with
lots of thermal mass), should lead to a high thermal gain, nonetheless.

I refrained from taking the inner oven apart, as I wasn't confident
I could put it back together without damage.

A small side note on the odd frequency: This Sulzer was apparently
ment to be used with NBS-5, the fifth atomic clock the then-NBS-now-NIST
built. I don't know whether it was actually used for NBS-5 or whether
it was just a spare part. As explained in [1], the oscillator was directly
used for the interrogation of the Cs beam using a multiplication chain
to 9.1GHz.

The PRS10 has, of all the Rb vapor cell standards I have seen, the most
interesting construction. Unlike most Rb's where there is a PCB on one
side and the physics package mounted on top of it or to the side, the
PRS10 has PCBs on all 6 sides around the physics package, similar to
the FRK. As PCB material is a pretty bad thermal conductor, I guess this
leads to additional thermal insulation, keeping the physics package more
stable. I don't know how much, but I guess the PCBs also result in a bit
higher RF shielding. Unlike, eg. the LPFRS there is also a pretty wide
gap between the lamp and the vapor cell, potentially leading to a higher
isolation and thus lower temperature gradient on the cell. This also
includes that the cell and the lamp are attached on the bottom PCB using
screws instead of a metal frame, again leading to higher thermal insulation.
I am not entirely sure what the bracket around the lamp is for, but my guess
would be some form of thermal or RF shielding.

Another interesting tidbit is the POS-400 (200-380MHz) VCO, which
is used to generate 359.72MHz for the SRD in the cavity. This is
much higher than the ~60MHz or ~90MHz most other designs use. Ie the
cavity Q can be a bit lower and still achieve high suppression of
the nearby harmonics.

The 10MHz OCXO of the PRS10 is on the output/connector board
and built by hand instead of using an of the shelf OCXO. Probably
because they didn't need the high stability of a commercial one
and could instead concentrate on the phase noise of the oscillator.
The package itself is pretty heavy and has large thermal mass,
so (short term) temp stability should be not an issue with this design.

The overall mechanical design and how the PCBs fit together suggests
that someone with good understanding of mechanics and assembly of
mechanical parts was involved from the very beginning. 

In contrast to the PRS10 and the Sulzer, the FE-405 is pretty boring
design. A PCB to host all the electronics and a small oven for the
5.0033MHz crystal. Onto the outside of the oven, the main crystal of
the FE-405 is attached, giving it some temperature control as well.
While the PCB is neatly done, I'm a bit disappointed at the haphazard
construction of the oven. It looks like they went the cheap way and
just let the soldering ladies glue the parts of the oven together.
What surprises me a bit is how small the oven is, suggesting that
the crystal inside uses a pretty small blank as well. Which goes
contrary to the wisdom that a larger blank leads to lower aging.

Last but not least, I'd like to thank Bob, Corby, and John who
made these nice frequency sources available to me and especially
Steve who hand-carried the Sulzer and the PRS10 over the big pond.


			Attila Kinali

[1] "Accuracy Evaluation and Stability of the NBS Primary Frequency Standards",
by Glaze, Hellwig, Allan, Jarvis and Wainwright, 1974
http://tf.nist.gov/general/pdf/65.pdf

-- 
It is upon moral qualities that a society is ultimately founded. All 
the prosperity and technological sophistication in the world is of no 
use without that foundation.
                 -- Miss Matheson, The Diamond Age, Neal Stephenson