Apologies for the rolling shutter effects, and the excess jitter introduced by converting this to a GIF.
(That WAS an animated GIF, but the forum seems to have chewed it up. Anyone know how to work around that, other than uploading it elsewhere?)
That my friends is the output of my prototype (read bailing wire and string) GPSDO being counted by the HP 53132A, now that the latter has had overnight and part of the day powered up for its OCXO to settle. The short term stability for the CTI OCXOs that several of us bought is specified at 0.05 ppb/s and that is what you can see there along with unavoidable measurement uncertainties - the last digit is 0.01 ppb. (The GIF has müllered the timing, the counter is on a 1 second gate time, so those readings are changing once per second).
Looks like I'm out of excuses for not making a proper PCB and putting the whole thing in a proper box with a proper UI. OK, I've one excuse left, I need to get an LCD connected up and roughly working before I commit to a PCB. That's a simple SPI interface so it's not exactly challenging or likely to turn into a roadblock unless it starts spraying digital noise into all my nice quiet analogue circuitry.
So, aim of buying the 53132A achieved. I am no longer marking my own homework and the GPSDO is looking as good as it claimed it was when it was checking its timing itself against GPS (Which could have been disastrously wrong if I'd made any mistakes).
Now, anybody got a source for a good cheap cæsium standard or a hydrogen maser?