[time-nuts] time-nuts Digest, Vol 21, Issue 22

[Normand Martel]

Help me! I need explanation!

Robert Lutwak wrote:

>With that said, the Perkin-Elmer rubidium on the IIR
>satellites is an outstanding clock, unquestionably
the
>best rubidium ever manufactured for any application,
>with short-term stability an order-of-magnitude
better
>than the IIF cesiums.  The drift is also remarkably
>low for a rubidium oscillator, compared to commercial
>rubidium, and, with daily updates, they will
>outperform the cesiums, though the cesiums will beat
>them outside of a week or so.

Well, i've got the PDF brochure for the Perkin-Elmer,
and it's indiscutably an outstandingly superb Rubidium
frequency standard (458 000 hours MTBF!!!), but
there's something that ticks me here, and i've seen it
on other rubidium standards. Let me explain:

On the brochure, the output frequency is 13.40134393
MHz and in the RF section of the block schematic, we
can see that the 13.40134393 MHz OCXO is multiplied
510  times (classic x6 followed by an SRD x85) before
driving the physics. If my calculator is OK, that
gives me 6834685404.3 Hz. However, the hyperfine
transition of Rubidium is at 6834682612.8 Hz! A
difference of 2791.5 Hz (0.40843 PPM)!!!! I don't
understand! That's not a Lada, that's a "Ferrari"
grade standard!

As i wrote earlier, i've seen a similar difference on
another (Efratom, if i remember) Rb standard! Can
someone explain me? I don't understand!

Thanks for your attention!

Normand Martel
Montreal, Canada

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com