[time-nuts] R&S XSRM Rubidium

Magnus Danielson magnus at rubidium.dyndns.org
Sun Jan 9 17:44:00 UTC 2011

On 09/01/11 12:12, Magnus Danielson wrote:
> On 09/01/11 11:20, Bruce Griffiths wrote:
>> Magnus Danielson wrote:
>>> On 09/01/11 05:17, paul swed wrote:
>>>> Lamp is orange purple
>>> Thanks, this is what I recall but never bothered to check myself.
>>> Where is that box of gas-lamps to help identify alkali metals?
>>> Cheers,
>>> Magnus
>> Use your spectrometer.
> Don't have one. Imagine that, I am lacking an instrument... :S
>> Spectrum should be similar to that attached albeit with some background
>> and contribution from the filler gas if any.
> Xenon-gas is a buffer gas in these cells, but don't know if there is any
> other. The manual says that Xenon-gas can cause some stray light too.
> The manual doesn't say how to observe light, but it might be in the
> german version of the text. I think I have a fair idea anyway.

OK. Now time to report on the progress here...

I recalled the FRS lamp repair guide:

On page 3 it describes how to do "Rb lamp rejuvination". So out the lamp 
goes, looks into it to see all the rubidium splattered around as 
described, heat gun applied and I had it deposited on the wrong side of 
the lamp, putting it upside down and re-applying the heat gun I could 
see the rubidium evaporate and when it was clear I let the lamp assembly 
cool down. As I inspected it I now had all rubidium concentrated as a 
single droplet up in the reservoir corner of the back of the lamp. Nice 
and clean. I then inserted the lamp, had the temperature for the lamp 
assembly adjusted, trimmed the 100 MHz for maximum response and just 
waited. Low and behold, I finally got a purple light. After hooking in 
the SMBs to close the loop I saw how the lamp response grew up well into 
the green (7 on the 10 degree scale, just where it used to be). After a 
small session of jumping around the lab in joy I turned it all off, 
re-assembled the whole thing, had it replaced into the rack where it 
should be and turned it on. After some patience the lamp meter returned 
back into reading 7. It was interesting to notice that the reading went 
up to maybe about 5 before it fell down to 4 and then rose up to 7. It 
did this both of the times.

It then slowly acquired frequency lock. I have now a mid-term 
measurement running with my Thunderbolt as reference. Will make a 
long-term measurement starting later at the night to see if I need to 
adjust the OCXO and then the C-field.

Anyway, now it is back to life, some vacancy and recovering exercises is 
to be expected.

Once I knew what to look for it was fairly simple, but then again there 
is a learning curve for these things.

Now the bench is cleared so I can see what that Cesium is not doing 
right with power...

Would love to have a HP5065A to make it company...

Magnus - bringing the lab back into operation step by step

More information about the time-nuts mailing list