[time-nuts] Cesium "wrist-watch-lite"

[SAIDJACK at aol.com]

H jwalck,
 
the availability of the CSAC in quantities and with short lead-times is not 
 an issue anymore. Jackson Labs Technologies can sell CSACs and can ship 
from  stock right now.
 
The main advantages of why the relatively higher price of the CSAC is not  
an issue for many projects is as follows:
 
* only 0.5 inches height. A comparable DOCXO has 0.8 inches typically. This 
 allows fitting the CSAC into VME chassis which are widely used, where the 
DOCXO  solution or Rb solution does not fit. Seems like a minor issue, but 
this makes  or breaks many a project
 
* Very fast turn-on. Atomic lock in typically 60 seconds, max 120 seconds.  
Very little retrace. No OCXO can even touch this
 
* Extremely low g-sensitivity. I tried to measure it and it was below the  
noise floor of my 53132A counter. No OCXO comes close, and most larger Rb''s 
are  also not nearly as good
 
* Power. The SRS PRS-10 unit consumes over 50 Watts according to spec  
during warmup. The CSAC's 0.12W are probably less than what the PRS-10 unit  
loses on it's power supply cable due to IR loss! The PRS-10 also needs  cooling 
of one form or another
 
* Extended temp range. -40C to +85C temp range is not matched by any  
commercial Rb, and very hard to find in DOCXO's
 
* On the ADEV stability, see the attached typical plot.
 
The drawbacks of the CSAC are not many, but also not trivial:
 
* Phase noise and spurs are pretty bad. The PRS-10 has excellent phase  
noise typically
 
* ADEV below 100s is not that great, but quite good >100s
 
* Thermal sensitivity is higher than the best state-of-the-art DOCXOs
 
But for many projects the above advantages far outweigh the slightly higher 
 price and slightly lower PN/ADEV performance.
 
bye,
Said
 
 
 
 
In a message dated 5/2/2013 10:23:27 Pacific Daylight Time,  
jwalck at netnod.se writes:

On  2013-05-02 14:18, Poul-Henning Kamp wrote:
> In message  <5182325E.4020003 at t-online.de>, Volker Esper writes:
>> Rummor  has it that the single piece price in the US is $1475 for just 
the
>>  CSAC. Weather that's with or without the demo board 
> 
> And for  that price a SRS PRS10 is a better buy, unless you need it
> to be  compact and not use too much power.

Yes, exactly. But the CSAC is  marketed for a small niche, both given its
price and that at least until  fairly recently they couldn't (and I'm
assuming still today can't) make  that many.

Small like (or smaller than) an OCXO, but with better  stability and much
lower power use. For a higher price. If one plotted  ADEV/W I recon it
would turn out a great performer. ADEV/USD and it  wouldn't fare as well.


I did go as far as asking for a quote from a  Symmetricom rep. last year
with a wrist watch project in mind. Put that  project on the shelf when I
got the quote back.:) Great to see someone with  deeper pockets and more
dedication made it happen! Is there any  competition/alternatives for a
pocketable time-nutty-stable  movement?

// jwalck

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ADEV_CSAC_Final.pdf
Type: application/pdf
Size: 23532 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20130502/34475084/attachment.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ADEV_Free_Run.pdf
Type: application/pdf
Size: 21549 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20130502/34475084/attachment-0001.pdf>