[time-nuts] HP5061 Cesium ion pump question

[Bob kb8tq]

Hi

Having watched a number of Cs tubes die ….. 

Measuring ADEV on your Cs is a good idea. As the tube goes downhill the
signal to noise often gets worse. This shows up even at relatively short tau
ADEV ( like 100 to 1000 seconds).  On a 5061 era device, this means doing
ADEV at the 1x10^-11 to 1x10^-12 level in the 100 to 1000 second range. 

A good (as in eBay for < $100) OCXO *might* cover this range adequately. 
A Telecom Rb probably is a better bet if you want to do a bit less “shopping” 
for your reference. 

If you have a counter that does 2x10^-11 at 1 second, it *should* get you
into the right range at 100 to 1,000 seconds. No need for ultra fancy setups. 
We’re not trying to show how *good* a working one is with this test. We’re
only trying to catch the “end of life” of the tube. 

Why bother? The compelling reason for playing with a Cs is accuracy. If
the SNR has the stability degraded by 10X (or more ….) it will take you a 
*very* long time ( = very long observation times) to get to that accuracy.
Indeed, the same things that degrade the SNR *may* degrade the accuracy
as well. 

Does this mean that all old tubes are junk? No it most certainly does not
mean that. There isn’t a lot of correlation between this and that and the
SNR (as the term is arbitrarily used above). After pumping down for a 
while, it may be doing fine. It also may not be. If, even after extended 
this or that, it’s still very noisy, you might want to factor that into what
you decide to do next.

Bob



> On Dec 4, 2020, at 9:16 AM, paul swed <paulswedb at gmail.com> wrote:
> 
> Exactly filaments cold. KE5FX has great details on doing this for a Cs
> tube. Google his details.
> Regards
> Paul.
> 
> On Fri, Dec 4, 2020 at 3:24 AM Dana Whitlow <k8yumdoober at gmail.com> wrote:
> 
>> I wonder if it might be good to power up the ion pump with the heaters
>> cold, then
>> after the ion current trails off to nil, begin operating the filaments at
>> very low voltage
>> (and all other voltages off except for the ion pump) and gradually ramp the
>> the filament
>> voltages up towards normal operating voltage, doing so slowly enough that
>> the ion
>> pump never trips off.  Then, once things have settled down at normal
>> filament voltage,
>> set up and fire up the whole instrument.  I think this might be easier on
>> both the filaments
>> and the ion pump.
>> 
>> Comments?
>> 
>> Dana
>> 
>> 
>> On Thu, Dec 3, 2020 at 8:34 PM <vilgotch1 at gmail.com> wrote:
>> 
>>> Thanks very much for this info Corby. It explains the behaviour of my
>> spare
>>> 5061A perfectly.
>>> 
>>> It also raises the question of what I could do with the spare tube I
>> have.
>>> I
>>> can connect a +3500V supply to the ion pump but that won't do anything
>>> about
>>> any gas molecules adsorbed onto the filaments as you described. Is it
>> worth
>>> powering up the filaments to get rid of them or can that be left for some
>>> indefinite time in the future when the tube could be reinstalled in a
>>> 5061A?
>>> 
>>> Morris
>>> 
>>> -------------------------------------------
>>> Date: Wed, 2 Dec 2020 11:06:22 -0800
>>> From: <cdelect at juno.com>
>>> To: time-nuts at lists.febo.com
>>> Subject: [time-nuts] HP5061 Cesium ion pump question
>>> 
>>> The gettering in the tube is only for capturing any stray Cesium atoms
>>> that don't get caught in the main gettering patch. If the gettering fails
>>> or gets too loaded up then the Cesium background level will get too high
>>> causing poor SN.
>>> 
>>> The ion pump is for any gases.
>>> 
>>> When a tube is off for extended times any gas atoms lingering or leaking
>>> slowly into the tube than happen to impinge on either the mass
>>> Spectrometer filament or the Cs oven filaments get capture by the
>>> filaments. They function as excellent getters!  (this even if the ion
>>> pump is on)
>>> 
>>> Now this is not by design but results in the filaments being "loaded"
>>> with the gas atoms.
>>> 
>>> Then when you turn the tube on the filaments light up and expell a burst
>>> of gas.
>>> 
>>> This of coarse causes the ion pump current to rise and trips off the
>>> filaments,
>>> 
>>> Once the ion pump removes the burst the cycle repeats until the filaments
>>> have expelled the trapped gases.
>>> 
>>> Then the ion pump can handle the load and pump the tube down completely.
>>> 
>>> Cheers,
>>> 
>>> Corby
>>> 
>>> 
>>> *************************
>>> 
>>> 
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>> To unsubscribe, go to
>>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>>> and follow the instructions there.
>>> 
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe, go to
>> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
>> and follow the instructions there.
>> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.