[time-nuts] Re: Want to build your own optical clock?

[Ed Marciniak]

Sadly, my junk box doesn’t include a stabilized external cavity diode laser suitable for these sorts of adventures. I do like the idea of using tilt lock instead of Pound-Drever-Hall for frequency stabilization. The optical detector pair, with appropriate loop filters could directly drive a PZT positioner and modulate diode current, skipping the lock in amplifier and other parts. The pesky little problem for the intended use is some is my junk box lacks a few higher dollar items like a zero order littrow (gold coated master) grating blazed for the 780-800nm range, an etalon with coatings suitable for 800nm instead of 500nm, and an optical isolator.  I'm also lacking a small machine shop.

I'm going out on a limb guessing I'd want a really narrow linewidth, perhaps less than 10 kilohertz along with about 1-10mW output power, and reasonably low AM noise.

Am I close on the hard part?

Would the principal gains be boosting the signal to noise ratio(increasing accuracy), and reduced light shift(from running light source at rather than offset from resonance)? Would you also benefit from not needing a separate filter cell or accepting the performance loss of a single cell with a carefully chosen gas and isotope mix?

Might there be any advantages to two stabilized lasers and detecting the beat note directly at 6.8 or 9.7 GHz to discipline the 10MHz output with an inverse PLL?

Does the whole problem deserve another look using 405-412 nm blue laser diodes, or even a different alkali atom choice as a result of not needing a fortuitous natural isotopic filter being available?

Is a passive or active maser possible?


________________________________
From: Magnus Danielson via time-nuts <time-nuts at lists.febo.com>
Sent: Thursday, January 25, 2024 8:56:18 AM
To: paul swed <paulswedb at gmail.com>; Discussion of precise time and frequency measurement <time-nuts at lists.febo.com>
Cc: Magnus Danielson <magnus at rubidium.se>
Subject: [time-nuts] Re: Want to build your own optical clock?

Well, as I recall it, you don't see it in visual wavelengths, so you use
a scope to view it, and what happens is that a dot becomes formed "in
thin air" (vacuum with some rubidium gas). Since it is fairly small, you
don't really preceive shape but more intensity as more atoms stay at the
same place.

At least my eyes isn't sharp enough to make out single atoms, despite
being fairly near-sighted. :)

Cheers,
Magnus

On 2024-01-25 14:57, paul swed wrote:
> Very curious. If you can visually see the process of cooling, would
> the line or dot become more tightly bunched as it cooled?
> Regards
> Paul
>
> On Thu, Jan 25, 2024 at 8:05 AM Magnus Danielson via time-nuts
> <time-nuts at lists.febo.com> wrote:
>
>     Paul,
>
>     When I did the NIST Time and Frequency Seminars, one of the lab evens
>     was the MOT. It was really instructive and fun to get the cooling
>     going,
>     and we where able to cool both Rb-87 and Rb-85. Looking at a spot and
>     realize it is really really cool atoms is cool. Only time that was
>     surpassed, was when I saw the cool atoms in NPLs cesium fountain, as
>     these are among the few that contribute to actual TAI.
>
>     Having a MOT is tempting. However, I may need to consider my
>     priorities.
>     I mean, what gear get's thrown out???
>
>     Cheers,
>     Magnus
>
>     On 2024-01-25 00:13, paul swed via time-nuts wrote:
>     > Skip it was interesting to take a look at what you sent. Sure
>     looks like
>     > some students are going to have fun with cold atoms.
>     > In some of the earlier threads there was a discussion on failure
>     modes and
>     > the laser was most likely. This had me thinking about the old
>     cesiums I
>     > have.
>     > Assuming they are on average 40 years old, then the current life
>     of the
>     > units are some 350,000 hours. Granted I don't run them 24 X 7.
>     > But it at least sets a number to what we are used to at some level.
>     > Not very scientific but a fun fact.
>     > Regards
>     > Paul
>     > WB8TSL
>     >
>     > On Wed, Jan 24, 2024 at 5:47 PM Skip Withrow via time-nuts <
>     > time-nuts at lists.febo.com> wrote:
>     >
>     >> Infleqtion will have a booth at Photonics West in San Francisco
>     Jan. 27 -
>     >> Feb. 1.
>     >> They are working on all kinds of quantum physics stuff, but
>     have two things
>     >> of particular interest to time-nuts.
>     >> 1. Tiqker - A Rb cold atom clock in a 3U rack mount package.
>     Pretty
>     >> limited production at this point, but something to watch.  I
>     think there is
>     >> another company in CA working on a similar product.
>     >>
>     >> 2. desqtopMOT - geared towards educational institutions, but
>     suitable for
>     >> draining your bank account and free time. Explore, educate, and
>     empower the
>     >> workforce of tomorrow with the desqtopMOT cold atom platform —
>     Infleqtion
>     >> <https://urldefense.proofpoint.com/v2/url?u=https-3A__www.infleqtion.com_desqtopmot&d=DwIGaQ&c=euGZstcaTDllvimEN8b7jXrwqOf-v5A_CdpgnVfiiMM&r=JsDsKeR7cZC8wbZhIlxxBQ&m=grfujnn45v84byeOnHukN2pQgBehKvHgT8wBGuqiKrz-HEhCPg_vY-k8VrbT4Lcp&s=TQBJM5ZxhqfkHEPWKHPD3BZjb8YfYNiv_E4uMwDzveg&e=>
>     >>
>     >> Regards,
>     >> Skip Withrow
>     >> _______________________________________________
>     >> time-nuts mailing list -- time-nuts at lists.febo.com
>     >> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>     > _______________________________________________
>     > time-nuts mailing list -- time-nuts at lists.febo.com
>     > To unsubscribe send an email to time-nuts-leave at lists.febo.com
>     _______________________________________________
>     time-nuts mailing list -- time-nuts at lists.febo.com
>     To unsubscribe send an email to time-nuts-leave at lists.febo.com
>
_______________________________________________
time-nuts mailing list -- time-nuts at lists.febo.com
To unsubscribe send an email to time-nuts-leave at lists.febo.com