No Script, No Fear, All Opinion
RSS icon Home icon
  • EEVblog #236 – FE-5680A Rubidium Standard Teardown

    Posted on January 14th, 2012 EEVblog 13 comments


    PART 1: http://www.youtube.com/watch?v=I55uLRRvLCU
    What’s inside the FE-5680A Rubidium frequency standard?
    Available on ebay for about $50 or so.

    Be Sociable, Share!
    • Phil Pemberton

      The disc-shaped thing on the top of the crystal is effectively a small crystal oven. It’s a high-power thermistor, which is rigged to maintain a constant temperature. If the temperature rises, the resistance rises too, and vice versa — thus, it will tend to reach a thermal equilibrium.

      If that’s under the foam — no doubt the foam is intended to act as thermal insulation to protect it from thermal shocks.

    • Andrew

      There’s a technical manual for the 5680A at http://www.ham-radio.com/wa6vhs/Test%20equipment/FREQUENCY%20STANDARDS/FE-5680A/5680%20TECH%20MANUAL.pdf

      On sheet 13 is the pinouts and commands used to talk to it.

    • Britt

      Ditto on the thermistor – I’ve got some around here that look just about exactly like that.

    • http://none Allan

      What are all the orange devices? Look like little plastic orange boxes.

      • Mike

        They’re tantalum capacitors.

    • Ross McKenzie

      Hi Dave,

      You may find the details in this link of use in your attempts to understand and control the unit.

      Cheers,

      Ross

      http://www.vk3um.com/Fe-5680A4.pdf

    • Nick

      Dave
      Just ordered 1 off ebay cost was $45 delivered.While searching around for info came across this site.
      http://www.qsl.net/zl1bpu/PROJ/Ruby4.htm

      Which appears to have the info needed to make the output variable from around 1K to 15M even has a DOS program for configuring the frequency.

      Cheers

    • http://kennethfinnegan.blogspot.com/ Kenneth Finnegan, W6KWF

      I’ve been doing some digging and poking around on my FE-5680A I just got in the mail. (Your stupid video cost me $50!)

      It is in fact serial programmable, but only 300Hz agile in each direction (with 32 bit resolution, none-the-less), so not useful for much more than calibration against a GPS locked source.

      The unpopulated switch-mode section is for one of the options that allows for the device to be powered from a single 20-35V supply instead of 5V,15V.

      The thing on top of the crystal looks to me like a PTC heater. These are barium titanate heaters that can be doped to severely lose conductivity at a specific temperature, so they’re essentially self-regulating heaters. They’re often used for QRSS oscillators as well, since it is a very easy way to oven-regulate a crystal.

      • http://www.eevblog.com EEVblog

        Thanks for the info. My videos cost time and money, at both ends of the stick!

    • Don

      This may be ignorance, but does the FE-5680A need to be sent to a calibration lab to be tracable? Or is it locked on frequency of the physics of the thing? Maybe just to verify the “locked” signal actually means the thing is locked and you’re not just seeing the OCXO output?

      • Bob Weiss, N2IXK

        The unit has a programmable frequency offset, set via serial command. I would think that if you want traceability, you would need to compare it to a primary standard (Cesium clock or GPS disciplined oscillator?), and tweak the offset to adjust the frequency as needed.

    • Bob Weiss, N2IXK

      Just got one of these in the mail yesterday. (Dave should be getting a commission on all the sales he spurred with these videos!)

      Thing powers right up when I apply +15 and +5, and the frequency seems to slowly sweep +/- 200Hz or so around 10MHz until the servo locks up, then (after about a minute from a cold start) it settles dead on to 10 MHz, at least as “dead on” s I can read with my frequency counter. :) There is a 1 PPS signal coming out on pin 6 of the DE-9 connector, in addition to the 10 MHz sine wave on pin 7. The 1 PPS is a TTL level positive going pulse, of ~1 uS duration.

      Next up is to try talking to it over the RS-232 port. According to what I can find on the ‘net, the units that need the additional +5 supply (and have only the 9 pin connector, no coaxial out) are only adjustable over a +/- 400 Hz range. Will have to take mine into the lab at work to get access to a higher resolution counter to see how close my unit is, and try tweaking the offset as needed.

      Now to build it into a nice box with a little distribution amp for multiple outputs, etc…

      • Todd Carney

        Great stuff, David. While you were showing the signal from the FE-5680A on your scope, it sure looked to me that the “sine wave” was really distorted, quite visibly distorted. It must be putting out a rich collection of harmonics, and who knows what else.

        Do I need to get my eyes checked, or have others noticed that same thing?