[time-nuts] Measurement of frequency of HP 8720D option 1D5 oscillator after switch on

John Miles john at miles.io
Sat Oct 4 20:39:34 EDT 2014

> Out of curiosity, what's the difference between the two first traces?
> One looks more jump than the other.

These are all factory new OCXOs (in this case, they're TimePod spare parts undergoing incoming test before I put them on the shelf).   The 1303-series parts both exhibited a small jump at around 18 hours, which I thought was interesting, and the blue trace is a bit cleaner, but they're all within spec.  

The first two are from a lot that's more than 18 months older than the green one being tested now.  I have a few more of the new ones to test, at which point I'll have a better feel for how they compare to the older ones.

> Is it me, or do they look very cheaply made? I have never seen one for
> real, but the photo I see on eBay looked more like a $200 Chinese
> power scope.

In real life the SR620s aren't badly built, and they're excellent performers, but I don't know of any aspects other than frequency input range in which they outperform a 5370.  (What specs are you referring to, Magnus?)  

You can get them with an optional built-in rubidium standard, so those will have better stability and accuracy numbers than a 5370.  But you're better off without that particular option, because it has a stupidly-loud fan that runs as long as the counter is plugged in.

With the 5370 you get the classic HP front panel look and feel, but actually the front panel slide switches are one of the biggest problems with 5370s.  They're poorly shielded from the elements and rather fragile in operation.  Bob's right when he recommends owning a second unit for backup service and/or parts donation.  Arguably that's not as important with an SR620, since they're newer and have a good reputation for reliability.

-- john, KE5FX
Miles Design LLC

More information about the time-nuts mailing list