LED instead of discharge lamp for Rb vapor cell standards
Moin,
Has anyone ever tried using a LED instead of a discharge lamp for Rb standards?
There are quite a few and very cheap 780nm IR LEDs available. They usually
have a line width in the order of 20nm to 50nm (FWHM). This means that they
still need the Rb85 filter, but they would not produce all those spectral lines
that the discharge lamp has, thus one could get rid of the (not so cheap)
optical filter that Corby uses for the super-5065.
The one thing I am not sure about is the filter efficiency of Rb85.
Because now the light isn't two discrete lines from the Rb87 lamb
of which one needs to be masked, but a continuous and wide spectrum.
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
As a side-effect, you also get easier regulation of the light intensity
of the LED, thus potentially less instability due to light-shift variation.
Beside the LED operation being much simpler than that of a discharge lamp
and overall less heat dissipation.
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
Attila I haven't tried that and I suspect the lines are important because
that establishes the reference.
Now I would also suggest that an experiment like you suggest be tried on a
FRS-XXX cheapy Rb.
Easily dis-assembled and can be re-assembled back with the lamp. No harm.
Paul
WB8TSL
On Wed, Feb 21, 2018 at 12:49 PM, Attila Kinali attila@kinali.ch wrote:
Moin,
Has anyone ever tried using a LED instead of a discharge lamp for Rb
standards?
There are quite a few and very cheap 780nm IR LEDs available. They usually
have a line width in the order of 20nm to 50nm (FWHM). This means that they
still need the Rb85 filter, but they would not produce all those spectral
lines
that the discharge lamp has, thus one could get rid of the (not so cheap)
optical filter that Corby uses for the super-5065.
The one thing I am not sure about is the filter efficiency of Rb85.
Because now the light isn't two discrete lines from the Rb87 lamb
of which one needs to be masked, but a continuous and wide spectrum.
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
As a side-effect, you also get easier regulation of the light intensity
of the LED, thus potentially less instability due to light-shift variation.
Beside the LED operation being much simpler than that of a discharge lamp
and overall less heat dissipation.
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts
and follow the instructions there.
In message 20180221184928.1c9b4c3a4d2e762970b5b181@kinali.ch, Attila Kinali w
rites:
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
I have a hard time seeing how you can not get worse S/N that way.
If you want to do it with a LED, I think it needs to be a stabilized
LED-Laser.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
Hi
There are a number of papers on LED / Laser excitation of Rb cells (and
other gas cells). They go back quite a ways. The gotcha (as pointed out
in PHK’s post) is that you need to stabilize the LED source. Doing that is
a bit complex. Doing that so that the result beats a gas lamp is a bit more
involved than just getting it working. Part of the problem is that there are
a number of transitions you can hit and you only want the “right” one. The
next layer is the signal to noise once you get on the right transition
Based on the examples I have seen on my bench, the ADEV of a typical
laser based unit is not super duper compared to Corby’s modified units.
In defense of the designers, they normally are targeting small size rather
than super stability. That’s what the market wants to buy …. If you are
spending tens (or hundreds) of millions of dollars, you focus on a product
you can sell a lot of.
Bob
On Feb 21, 2018, at 12:49 PM, Attila Kinali attila@kinali.ch wrote:
Moin,
Has anyone ever tried using a LED instead of a discharge lamp for Rb standards?
There are quite a few and very cheap 780nm IR LEDs available. They usually
have a line width in the order of 20nm to 50nm (FWHM). This means that they
still need the Rb85 filter, but they would not produce all those spectral lines
that the discharge lamp has, thus one could get rid of the (not so cheap)
optical filter that Corby uses for the super-5065.
The one thing I am not sure about is the filter efficiency of Rb85.
Because now the light isn't two discrete lines from the Rb87 lamb
of which one needs to be masked, but a continuous and wide spectrum.
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
As a side-effect, you also get easier regulation of the light intensity
of the LED, thus potentially less instability due to light-shift variation.
Beside the LED operation being much simpler than that of a discharge lamp
and overall less heat dissipation.
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
On Wed, 21 Feb 2018 18:10:17 +0000
"Poul-Henning Kamp" phk@phk.freebsd.dk wrote:
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
I have a hard time seeing how you can not get worse S/N that way.
Well, the question is how much? Keep in mind that an Rb lamp is
anything but a clean light source.
If you want to do it with a LED, I think it needs to be a stabilized
LED-Laser.
The problem with a Laser is that you need to lock it to the
right line. This requires either some significant change to the
electronics of the Rb standard, or an additional vapor cell and
some optics to lock it to that cell. Neither is trivial (apparently,
according to the papers I've read, not difficult, but I don't trust
that until I have tried myself).
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
Hi
On Feb 22, 2018, at 8:22 AM, Attila Kinali attila@kinali.ch wrote:
On Wed, 21 Feb 2018 18:10:17 +0000
"Poul-Henning Kamp" phk@phk.freebsd.dk wrote:
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
I have a hard time seeing how you can not get worse S/N that way.
Well, the question is how much? Keep in mind that an Rb lamp is
anything but a clean light source.
The point is that it’s clean where it matters. It does not dump a bunch of
energy into the transitions that you want to avoid.
Bob
If you want to do it with a LED, I think it needs to be a stabilized
LED-Laser.
The problem with a Laser is that you need to lock it to the
right line. This requires either some significant change to the
electronics of the Rb standard, or an additional vapor cell and
some optics to lock it to that cell. Neither is trivial (apparently,
according to the papers I've read, not difficult, but I don't trust
that until I have tried myself).
Attila Kinali
--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi,
On 02/22/2018 02:53 PM, Bob kb8tq wrote:
Hi
On Feb 22, 2018, at 8:22 AM, Attila Kinali attila@kinali.ch wrote:
On Wed, 21 Feb 2018 18:10:17 +0000
"Poul-Henning Kamp" phk@phk.freebsd.dk wrote:
So the Rb85 "notch" filter might not get rid of all the unwanted light
and some of this might depopulate the excited state of the Rb87
that are in the RF cavity.
I have a hard time seeing how you can not get worse S/N that way.
Well, the question is how much? Keep in mind that an Rb lamp is
anything but a clean light source.
The point is that it’s clean where it matters. It does not dump a bunch of
energy into the transitions that you want to avoid.
780 nm vs 795 nm if I recall correctly. A dash of energy isn't too
critical, but it will de-pump the state and you will suffer with
somewhat worse S/N. For a rubidiumlamp the two "lines" are about the
same strength and fairly wide.
The one thing to care about is polarization, since that will be a
pulling through AC Stark so a quarter-wave window should be used.
Servo the LED to the detected lamp response should help with stabilization.
Cheers,
Magnus