Hi everyone,

I ran across an old rubidium frequency standup, model #: FE-5650A. The side panel is missing, so the insides are exposed. There is a label stating "Revision C, 0109657S02." There is a 20-pin slot on one side of the device.

A friend of mine told me that it could be used to make an atomic clock. Although the idea of having one hell of an accurate clock is interesting, I know nothing about this device. And I don't know anything about electronics. I only know basic stuff like how to solder and simple power systems, etc. Is it possible to make an atomic clock from this? If so, how would I do it?

 

AFAIK, Rubidium is the Slave part of the Atomic clock for Jitter Removal from a Master Cesium Clock.
The Cesium clock has High Accuracy, Very low Drift, but a lot of Jitter.
The Rubidium is Not a Master clock, it is a slave because it has Low Jitter, and it locks on to the Cesium clock and filters out the jitter.

I am not too sure if the Rubudium setup is used as a clock by itself.

 

A perfectly functioning rubidium frequency standard is part of "an atomic clock", but it seems that you may not have even that part in an operational condition. So the very first step is to discover what major part of the system you have is missing. It might be the part with all of the electronics. And while it might be possible, if you had an unlimited budget, it does not seem likely to succeed.

 

The clock part = easy.

Rubidium frequency standard = hard.

If the standard does not work, you got nothing. Focus on getting this to work first.
I imagine it's a bit of a power hog, how much is this idea worth in terms of wasted electricity?

 

The whole package could also be part of a retired rubidium frequency standard used for cellular systems back a few years, in the analog phone era. The portion that was on the missing circuit board that plugged into that 20 pin connector could be a serious show-stopper if it can not be found.

 

Ah, gotcha. Thanks for the insight! Looks like this item is just trash at this point.

 

Ah, gotcha. Thanks for the insight! Looks like this item is just trash at this point.

I suggest investigating the asembly because there is no clue a to what is still there. It could be a lot of interesting parts and pieces, and there could be other electronic portions useful for other applications.

 

Hi everyone,

I ran across an old rubidium frequency standup, model #: FE-5650A. The side panel is missing, so the insides are exposed. There is a label stating "Revision C, 0109657S02." There is a 20-pin slot on one side of the device.

A friend of mine told me that it could be used to make an atomic clock. Although the idea of having one hell of an accurate clock is interesting, I know nothing about this device. And I don't know anything about electronics. I only know basic stuff like how to solder and simple power systems, etc. Is it possible to make an atomic clock from this? If so, how would I do it?

Some of the FE-5650A were fixed frequencies, like 10Mhz, others were programmable. The programming was via a DB9 serial connector, though it was not a standard RS-232 interface. That is, there were 2 pins for RX/TX of serial data. Other pins gave things like 1 pulse per second. I don't recall the details.
The device had a rubidium "oven", that would heat up and produce a high frequency oscillation based on Rubidium resonance, somewhere up in the 6Ghz range I believe. From there, the electronics would divide it down to the desired output frequency, usually 10Mhz. The whole unit needed 15VDC (or a bit more) at a couple of amps to run.
Problem was that often the oven would eventually burn out, after all it was a filament that did the heating. Once that happened, the unit was junk. I had such a unit, and it did "burn out" eventually.
Rubidium clocks are fairly accurate, but not as accurate as a Cesium clock. Thus Rubidium clocks were often considered second tier atomic clocks. Still good for most uses, but not for high end scientific use.