Maker Pro
Maker Pro

New Cesium reference

K

Ken Smith

Jan 1, 1970
0
Just in case some of you missed the article about a tiny cesium based
reference clock:

http://www.commsdesign.com/showArticle.jhtml?articleID=46200241

That looks like it could really be the next generation of clock
reference, provided it becomes available at reasonable cost some day.

The tricky bit is the pumping laser. It has to be held to within about
100MHz on the line[1]. It has to be a single mode laser, at the right
frequency and quiet.

[1] It is dithered and servoed to keep the center of the dither lined up.

The very small cell is not really that big of a problem but the signal to
noise does suffer a bit. I have to assume that they are using a coated
cell. If not the Cs atoms would get scrambled by hitting the wall.
 
J

John Larkin

Jan 1, 1970
0
Just in case some of you missed the article about a tiny cesium based
reference clock:

http://www.commsdesign.com/showArticle.jhtml?articleID=46200241

That looks like it could really be the next generation of clock
reference, provided it becomes available at reasonable cost some day.

The tricky bit is the pumping laser. It has to be held to within about
100MHz on the line[1]. It has to be a single mode laser, at the right
frequency and quiet.

[1] It is dithered and servoed to keep the center of the dither lined up.

The very small cell is not really that big of a problem but the signal to
noise does suffer a bit. I have to assume that they are using a coated
cell. If not the Cs atoms would get scrambled by hitting the wall.

This is just like a rubidium, but they replaced the discharge lamp
with a laser... must have got lucky to find a decent laser that hits a
cesium line.

Presumably the cs cell must still be heated to vaporize the stuff, so
it doesn't actually sound like a good thing to integrate onto a chip.
The world *does* need some new, cheap physics for atomic-level clocks.

John
 
K

Ken Smith

Jan 1, 1970
0
John Larkin said:
This is just like a rubidium, but they replaced the discharge lamp
with a laser... must have got lucky to find a decent laser that hits a
cesium line.

It wan't luck. Part of the team that did the work was the guy who made
the lasers special for them. The yeld is fairly small so far. The laser
cavity is defined deposited layers that form a mirror at the desired
frequency range.

Presumably the cs cell must still be heated to vaporize the stuff,

Yes they run at about 65C in the cell. The laser has to be heated to a
different temperature so it, so far, can't be placed right on the cell.

The laser temperature controls the wavelength.
so
it doesn't actually sound like a good thing to integrate onto a chip.

They are using MEMS technology to make the physics very small. The small
cell is inside a slightly larger envelope which they pump the air out of.
This larger envelope is what would be set down on the chips surface in the
final version.

Their current thinking involves mirrors, but no smoke, to allow the laser
and the photodiode to be on the same side of the cell.
 
J

Joerg

Jan 1, 1970
0
Hi Tim,

Yes, that would be a high-tech solution. I always wondered how the guy can hold that. Even Arnold would probably work up a cramp with all that weight.

On the other hand I have seen waitresses at the Hofbraeuhaus in Munich carrying half a dozen steins of beer in each hand, clear across the whole large room.

Regards, Joerg
 
R

Robert Baer

Jan 1, 1970
0
John said:
Just in case some of you missed the article about a tiny cesium based
reference clock:

http://www.commsdesign.com/showArticle.jhtml?articleID=46200241

That looks like it could really be the next generation of clock
reference, provided it becomes available at reasonable cost some day.

The tricky bit is the pumping laser. It has to be held to within about
100MHz on the line[1]. It has to be a single mode laser, at the right
frequency and quiet.

[1] It is dithered and servoed to keep the center of the dither lined up.

The very small cell is not really that big of a problem but the signal to
noise does suffer a bit. I have to assume that they are using a coated
cell. If not the Cs atoms would get scrambled by hitting the wall.

This is just like a rubidium, but they replaced the discharge lamp
with a laser... must have got lucky to find a decent laser that hits a
cesium line.

Presumably the cs cell must still be heated to vaporize the stuff, so
it doesn't actually sound like a good thing to integrate onto a chip.
The world *does* need some new, cheap physics for atomic-level clocks.

John

Wild ideas..
How about using quantum dot electrons, interacting in some manner,
with Rubidium or Cesium?
Quantum entanglement?
Use Rubidium or Cesium as doping in a crystalline structure to get
around need for hot gas?
 
T

Tim Shoppa

Jan 1, 1970
0
Joerg said:
Yes, that would be a high-tech solution. I always wondered how the guy can
hold that. Even Arnold would probably work up a cramp with all that weight.

:).

What you posted is interesting if it can be made commercial...

Commercial Cesium standards have always been around
$20,000 and haven't really gotten a lot smaller since the
end of the 1960's. Of course, in that time they have gotten
way way more accurate and $20,000 today isn't nearly as much
money as it was 35 years ago.

Tim.
 
K

Ken Smith

Jan 1, 1970
0
Robert Baer said:
Wild ideas..
How about using quantum dot electrons, interacting in some manner,
with Rubidium or Cesium?
Quantum entanglement?

Shhhhh people might be listening
Use Rubidium or Cesium as doping in a crystalline structure to get
around need for hot gas?

That will cause too much Stark effect from the "cystal field". The line
would be very wide. Desides you'd have to trap Cesium atoms in the
crystal without causing them to react chemically.
 
J

John Larkin

Jan 1, 1970
0
Just in case some of you missed the article about a tiny cesium based
reference clock:

http://www.commsdesign.com/showArticle.jhtml?articleID=46200241

That looks like it could really be the next generation of clock
reference, provided it becomes available at reasonable cost some day.

The tricky bit is the pumping laser. It has to be held to within about
100MHz on the line[1]. It has to be a single mode laser, at the right
frequency and quiet.

[1] It is dithered and servoed to keep the center of the dither lined up.

The very small cell is not really that big of a problem but the signal to
noise does suffer a bit. I have to assume that they are using a coated
cell. If not the Cs atoms would get scrambled by hitting the wall.


Does anybody know why cesium and rubidium are the only things used in
commercial clocks? Obviously they are metals that have low vapor
pressures, but why metals, why not a gas or something? I think Townes
suggested an ammonia clock way back when.

And why the optical-microwave thing, and not just a microwave
resonance?

John
 
K

Ken Smith

Jan 1, 1970
0
John Larkin said:
Does anybody know why cesium and rubidium are the only things used in
commercial clocks? Obviously they are metals that have low vapor
pressures, but why metals, why not a gas or something? I think Townes
suggested an ammonia clock way back when.

You want one electron in the outer shell. More electrons means more
interactions which means more and wider lines. Narrow line width and big
SNR are key.
And why the optical-microwave thing, and not just a microwave
resonance?

The optical pumping gives you a signal that is not the signal you applied
and thus makes the crosstalk issues way less. The signal on the photocell
is the dither frequency and its second harmonic. Neither of these is
applied to the physics package so there is no chance for crosstalk.
 
R

Rich Grise

Jan 1, 1970
0
Ken said:
Shhhhh people might be listening


That will cause too much Stark effect from the "cystal field". The line
would be very wide. Desides you'd have to trap Cesium atoms in the
crystal without causing them to react chemically.
Buckyballs! With Buckytube waveguides for the output. :)

Cheers!
Rich
 
J

John Larkin

Jan 1, 1970
0
You want one electron in the outer shell. More electrons means more
interactions which means more and wider lines. Narrow line width and big
SNR are key.

Cool. Metals it is.
The optical pumping gives you a signal that is not the signal you applied
and thus makes the crosstalk issues way less. The signal on the photocell
is the dither frequency and its second harmonic. Neither of these is
applied to the physics package so there is no chance for crosstalk.

Good point. I did some summer work in microwave spectroscopy, and we
got some really good lines, but the Stark waveguide was 40 feet long.

I got a rubidium brick on ebay a while back. It came with a schematic,
and was surprisingly simple.

John
 
K

Ken Smith

Jan 1, 1970
0
Ken Smith wrote:
[...]Buckyballs! With Buckytube waveguides for the output. :)

Wax balls, Buckyballs with Hydrogen atoms as the inside surface,
would be needed. If a polarized Cs atom hits most atoms, it gets
scrambled. You need a surface that doesn't do this. Gas cells either
have a buffer gas to keep the Cs atoms from moving around or are coated.
 
J

John Woodgate

Jan 1, 1970
0
I read in sci.electronics.design that Ken Smith
You want one electron in the outer shell. More electrons means more
interactions which means more and wider lines. Narrow line width and
big SNR are key.

OK, one electron applies to ammonia. But its molecular weight is very
low compared with Rb and Cs. Luckily, there are ammonia analogues up the
Periodic Table column: PH3 (phosphine), AsH3 (arsine) and SbH3
(stibine). I'm not sure about BiH3, but that is MASSIVE if its stable
enough. It's a gas above 17 C.

Whether anyone can make a clock out of it is another matter.
 
J

Joerg

Jan 1, 1970
0
Hi Tim,
What you posted is interesting if it can be made commercial...

Commercial Cesium standards have always been around
$20,000 and haven't really gotten a lot smaller since the
end of the 1960's. Of course, in that time they have gotten
way way more accurate and $20,000 today isn't nearly as much
money as it was 35 years ago.
Well, in the 60's people bought houses for $20k. Today you can buy a car
with that kind of money, a house would be about 10dB to 15dB more
nowadays. 18dB for the Bay Area...

The NIST folks talk about low cost so it must be possible to build it
for a fraction. I had been involved in pressure sensors and it was the
same thing. Freaking expensive. I paid north of $200 for my parachuting
altimeter in the 80's. Then MEMS came and we could make sensors for a
few Dollars a pop, only 15 years later. But these were all on silicon,
no more mechanics. Here is an article about the same reference where
they state "low cost". Also has a picture.

http://yubanet.com/artman/publish/article_12814.shtml

Regards, Joerg
 
T

Thomas Magma

Jan 1, 1970
0
What are the best guesses on samples or finished product?
 
R

Robert Baer

Jan 1, 1970
0
Rich said:
Buckyballs! With Buckytube waveguides for the output. :)

Cheers!
Rich

....Use a "large" one, to hold Cesium or Rubidium gas; might get away
with less than 100 atoms, so it might act like a gas at 20C.
 
M

Mike Monett

Jan 1, 1970
0
[email protected] (Ken Smith) wrote in message
[... snip excellent info on cesium]

Ken, you seem to have a great deal of inside information on these
clocks. Do you work with any of the groups in development?

Also, how do they synthesize the required frequencies with sufficient
resolution and low enough jitter to find the cesium resonance? You'd
think this might be difficult to do in a 19" rack - but they shrink it
to very small dimensions.

Actually, their achievement is remarkable. But I would really like to
find out more about the synthesizer. One could think of a number of
other applications for a small low-power unit with specs like that:)

Regards,

Mike
 
J

John Larkin

Jan 1, 1970
0
[email protected] (Ken Smith) wrote in message
[... snip excellent info on cesium]

Ken, you seem to have a great deal of inside information on these
clocks. Do you work with any of the groups in development?

Also, how do they synthesize the required frequencies with sufficient
resolution and low enough jitter to find the cesium resonance? You'd
think this might be difficult to do in a 19" rack - but they shrink it
to very small dimensions.

Actually, their achievement is remarkable. But I would really like to
find out more about the synthesizer. One could think of a number of
other applications for a small low-power unit with specs like that:)

Regards,

Mike


My Efratom uses a 20 MHz VCXO. A divider chain creates 5 MHz and 312.5
KHz. They're mixed in an xor gate to make 5.3125 MHz. The 20 MHz is
also multiplied to 60 MHz. The 60 is combined linearly with the 5.3125
and drives an SRD in a resonant cavity. Somehow some line in this mess
hits the 6.834687 GHz rubidium resonance, 114th harmonic or something.
The power must be minute.

I don't quite get this, but that's what the manual says. The circuitry
is fairly simple, just HC logic and a few transistors. The whole thing
is about as big as a coffee mug.

John
 
Top