[time-nuts] Re: Efratom FRK with "inoperative lamp"

[Julien Goodwin]

On 5/2/22 2:58 am, paul swed wrote:
> Julien
> I don't think I would read a lot into the failed R just replace it. They
> are in a nasty hot environment and simply can age and die. Get at leat 2
> resistors just incase there really is a shorted transistor or something
> else going on.
> Regards
> Paul

Oh absolutely, my rough guess is the thermal insulation foam degraded 
corroding the leads /eventually/. This unit is over 30 years old, and 
who knows how it was used, although being in a road case makes me 
suspect it was not the sort of life we might like.

I swapped in two 10-ohm 1-watt carbon film resistors today (being what I 
could get my hands on).

Sadly I couldn't just solder to the lead stubs still poking through the 
board, they were just too corroded. A significant disassembly and 
reassembly later the unit appears to be improved, but still not locking.

Should anyone run into this later, a clear bench, lots of lighting, and 
a way to ensure any screw you drop doesn't disappear never to be seen 
again are strongly recommended. Sadly my own bench is rather more of the 
Jim Williams[1] variety. A good soldering setup is also needed, but 
there I was rather luckier, even if my new desoldering iron did end up 
lifting a few pads.

I /think/ it's using more power than it did before (right around 20W at 
total cold, dipping to 10W after a while), although silly me didn't make 
a note of it, but even after 30 minutes it's not locking.

Given that the resistor I replaced is in a key point for the lamp 
excitation oscillator I won't be surprised if it's totally off and needs 
a recalibration, but that's a task for another day. When I stick a scope 
probe in the lamp cavity as the manual suggests I do see a a signal, so 
there's certainly hope, although I don't see the expected glow.

I did confirm the heater is working, and was getting above 90c when I 
poked a thermocouple at it just after power-off.

1: https://www.edn.com/honoring-jim-williams/

> On Fri, Feb 4, 2022 at 3:50 AM Julien Goodwin <time-nuts at studio442.com.au>
> wrote:
> 
>> Well that was unexpectedly easy (to diagnose).
>>
>> Thanks all for the replies, much appreciated, you have once again
>> confirmed my appreciation for this group.
>>
>> I sat down this evening to work through your various suggestions, and,
>> by chance, picked up the unit without my lab power supply (and more
>> importantly, its noisy fans) being on, where I heard a little rattle I
>> must have missed before.
>>
>> Removed the bottom panel, which I hadn't done yet, and out dropped the
>> remnants of a Vishay RW80U 20-ohm resistor. Based on the circuit this
>> pretty much has to be R2 on the lamp board (following the December 73
>> circuit I have, which has that at 10 ohms), which is directly in-circuit
>> for the lamp, so would explain exactly what I've seen.
>>
>> The lamp board in general seems to have a few holes missing components,
>> so I'm going to have to get it out which looks like a project in itself.
>>
>> If anyone has any pointers to decent guides on optimum process on
>> further disassembly, otherwise I guess it's copious photos, notes, and
>> hoping I don't stuff up anything too much.
>>
>> I'd also appreciate suggestions for replacement foam to use in the lamp
>> cavity as the original stuff has long since degraded.
>>
>> Paul, Peter Robert,
>> Thanks for reinforcing what I'd suspected with the lamp.
>>
>> Matthias,
>> I hadn't bothered with an actual temperature meter, luckily on the FRT
>> it's actually easy to confirm lamp temperature in use, and it certainly
>> wasn't anywhere near 100c. I think what heat there was likely leaked
>> through from the crystal oven, which was getting pretty hot.
>>
>> Bob,
>> Fortunately the onboard crystal does seem likely to still be close
>> enough, based on what I was seeing on a counter, so I feel like I have a
>> hope of recovering the unit if I do manage to get the lamp going again.
>>
>> On 1/2/22 8:17 pm, Julien Goodwin wrote:
>>> I picked up an Efratom FRT a while ago, and it finally arrived the other
>>> day, sadly it doesn't lock.
>>>
>>> The oscillator inside is an older Efratom FRK, and it's also clear
>>> someone had been inside already, although it does appear to be complete.
>>>
>>> The FRK manual does include a key description I wish I'd noticed a few
>>> days ago before full disassembly:
>>>
>>> "Terminal pin 7 provides an indication of proper operation of the
>>> rubidium lamp. For nominal operation the signal is 6 to 12V. An
>>> inoperative lamp is indicated by an signal of approximately 3V. The
>>> internal resistance of this circuit is approximately 6K."
>>>
>>> ... indeed I'm seeing 3v on that pin.
>>>
>>> Based on the circuit this appears to simply be the output of the first
>>> stage op-amp from the photo-cell, so is this effectively an indication
>>> that the (electrical part of the) lamp is dead? Certainly it wasn't
>>> anywhere near the temperature it should have been when I removed the
>>> bulb after running for over an hour.
>>>
>>> Just for completeness, the unit does draw up to ~0.8A when started up
>>> cold with 24V, dropping down to about ~0.5A after a while. The sealed
>>> sub-section of board A4 seems be where all the heat is, and I'm assuming
>>> that's an ovenised part of the crystal oscillator.
>>>
>>> I also popped the rubidium bulb out and confirmed there's no obvious
>>> issues there.
>>>
>>> I do have another FRK here, but as it's an FRK-LN, and it's also my
>>> working house standard I'm somewhat loath to open it up lest I end up
>>> with two dead units and zero working.
>>>
>>> Any suggestions?
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at lists.febo.com -- To unsubscribe
>>> send an email to time-nuts-leave at lists.febo.com
>>> To unsubscribe, go to and follow the instructions there.
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com -- To unsubscribe send
>> an email to time-nuts-leave at lists.febo.com
>> To unsubscribe, go to and follow the instructions there.
>>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com -- To unsubscribe send an email to time-nuts-leave at lists.febo.com
> To unsubscribe, go to and follow the instructions there.