[time-nuts] Speaking of HP5065 repair/improvement

Poul-Henning Kamp phk at phk.freebsd.dk
Mon Aug 3 17:54:20 EDT 2015

I recently bought myself a HP5065 and after I replaced the OCXO
with one from my defunct 5061, it seems to be humming along nicely:
4e-13 out to about tau=60k so far.

Next experiment is to calibrate/tune it according to the manual and
see if that helps anything.

My reason to buy it was three-fold.  I need a free-running
house-standard for the NTP stuff I'm doing, I wanted an electronics
project to justify having bought AOE3, and finally Corbys "Super"
modification and various peoples measurements show there is
unrealized potential in the physics package.

My unit is series 940, which means that it uses stuff like outdated
uA702 op-amps etc, that leaves quite a room for improvement.

At the far end of my journey, is an experiment where I drive the
physics package with entirely new electronics:  A DDS for the 60MHz,
an ADC on the "solar cell", a DAC on the C-field and an ARM CPU to
do steer it all.  (I might leave the temp-controller behind).

But to begin with I'm just going to do some of the upgrades I can
see in the newer version of the manual, for the sheer fun of it
and just to see how much they improve the performance.

For instance the A9 Integrator goes from a uA702 to a OPA111BM -
that's unlikely to hurt the performance...

This is not going to be a particularly fast project, but any insights,
wisdom, data og participation is most welcome.


PS: I want to stress that I am not going to compete with Corby: I
    won't fix or tune your 5065 for you.  If you want your 5065 in
    competent hands, send it to Corby.

Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

More information about the time-nuts mailing list