[time-nuts] Fundamental limits on performance
magnus at rubidium.dyndns.org
Fri Sep 18 08:07:01 UTC 2009
Chris Caudle wrote:
>> From: "Rick Karlquist" <richard at karlquist.com>
>> Is there any evidence that these crystals are actually
>> baked out at a higher temperature than cold weld
>> copper packaged ones?
> Did you find the second link that Bruce G. provided? At the page labeled
> 379 (I think actually page 11 of that document), the section "Future
> Research for Deep Space Mission Needs" describes the processing. Baking at
> over 230 deg. C to drive out impurities, and when doing that and drawing a
> vacuum, the copper softens and collapses if you try a standard package.
> I don't know if those fox all-quartz packages can maintain high vacuum or
> not. I guess I could ask them next time they are trying to get me to buy
> some oscillators.
In contrast, Oscilloquartz BVA use coldwelded package with a chimney,
bake it at 250 C and vacuum pump to 10^-8 mbar. They then seal it
through pinchoff. It is fairly well described in the 8601 datasheet.
They even feature a nice photo of the BVA package with measures and all.
Another interesting feature of the BVA design is a reduced sensitivity
to gravity. The 8601 datasheet even contains 360 degrees turn-graphs to
illustrate the sensitivity.
In one of their papers they even had the f^-3, f^-1 and white noise
The 8601 runs hotter than the 8600. 4,5 W against 2,5 W if beleiving the
Wish I had a 8607 to run along with them. Ah well.
Should make a rig to get the E1938A running and 10811A and 10811B as well.
More information about the time-nuts