C-Beam Usage
Hi All,
This question assumes the following:
- You are not doing "for money" cal work.
- There is a GPSDO and one or two Rb's running 24/7
- Your C-Beams are not nice new Agilent or Symmetricom's
So the question:
Do you run the C-Beam(s) 24/7 or only fire them up enough in advance of
when you want to check other things?
Obviously the question relates to the life expectancy of the tubes and the
cost involved in replacing one.
Thanks,
Had
In message 6.1.2.0.2.20050113190428.02c22eb0@mail.landmarcomm.com, Had writes
:
Hi All,
This question assumes the following:
- You are not doing "for money" cal work.
- There is a GPSDO and one or two Rb's running 24/7
- Your C-Beams are not nice new Agilent or Symmetricom's
So the question:
Do you run the C-Beam(s) 24/7 or only fire them up enough in advance of
when you want to check other things?
I only fire mine up when I need it.
--
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.
Up until I started working on my FTS4050, I would have said that they last
the same length of time whether you use them or not. That certainly was the
usual experience with HP5060, 5061, 5062 units of yor. However, seeing my
4050 show strong signs of activity in its 16 year old tube, which has just
spent the last 10 years untouched in storage, I am changing my tune.
With the FTS tubes, I think you can get away with just firing it up when you
need it. These tubes seem to be so well sealed that I am not sure that there
is any benefit in firing them up to keep the vacuum pumped down.
-Chuck
Poul-Henning Kamp wrote:
In message 6.1.2.0.2.20050113190428.02c22eb0@mail.landmarcomm.com, Had writes
:
Hi All,
This question assumes the following:
- You are not doing "for money" cal work.
- There is a GPSDO and one or two Rb's running 24/7
- Your C-Beams are not nice new Agilent or Symmetricom's
So the question:
Do you run the C-Beam(s) 24/7 or only fire them up enough in advance of
when you want to check other things?
I only fire mine up when I need it.
In message 41E7C234.8060106@erols.com, Chuck Harris writes:
Up until I started working on my FTS4050, I would have said that they last
the same length of time whether you use them or not. That certainly was the
usual experience with HP5060, 5061, 5062 units of yor.
There are multiple wear-out mechanisms, one is cesium depletion the
other major one is photocell wearout.
I don't know if turning off the Cs oven also turns off the photocell,
but you certainly do not use up your precious cesium when it is frozen
solid.
--
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.
The photocell (photomultiplier really) should last forever if it is
kept "dark", and in a vacuum. Discrete P-M tubes tolerate storage quite
well.
I believe that the traditional cause for C-beam tubes to fail in storage is
degradation of vacuum. These new tubes are unbelievably clean inside, and have
as near to perfect seals as is possible. I was absolutely amazed to find
my tube had made it through 10 years of careless storage with a good hard
vacuum. I suspect that mine is not unique in this regard.
-Chuck
Poul-Henning Kamp wrote:
I don't know if turning off the Cs oven also turns off the photocell,
but you certainly do not use up your precious cesium when it is frozen
solid.
Do you run the C-Beam(s) 24/7 or only fire them up enough in advance of
when you want to check other things?
If you're curious you'll keep the Cs on and monitor
its progress against the other standards.
If you're practical you'll use the minimum necessary
for your accuracy requirements. In many cases this
means that a Rb or GPSDO is all you ever need.
I've heard of some people who run off the OCXO in
their cesium standard (the 5061A "Cs Off" mode) and
turn on the Cs for a couple of hours once a month
simply to recalibrate the OCXO.
The OCXO inside Cs standards tend to be very high
quality so this makes a lot of sense. If the drift rate
is down to 1e-10/day your accuracy easily stays
within a few parts in 10^9th per month with almost
no drain on finite Cs resources. Or you can turn it
on once a week, etc.
The short-term stability of this solution is often better
than Rb or GPSDO.
Obviously the question relates to the life expectancy of the tubes and the
cost involved in replacing one.
I'm told Cs tubes reach end of life either when they
run out of fresh Cs metal in the oven on one end of
the tube or when the leading dynode of the electron
multiplier is saturated with used Cs on the other end
of the tube.
I also heard that high performance tubes (HP option
004) are more likely run out of Cs while standard tubes
are more likely to die of EM saturation in the long-term.
A Cs standard running in Cs Off mode is not subject
to either of these.
If someone has more reliable or inside information please
let us know.
/tvb
In message 000501c4fa6b$2b61f060$fe25f304@computer, "Tom Van Baak" writes:
I've heard of some people who run off the OCXO in
their cesium standard (the 5061A "Cs Off" mode) and
turn on the Cs for a couple of hours once a month
simply to recalibrate the OCXO.
That's more or less what I do :-)
--
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.