[time-nuts] HP 10811 stabilizing time

Bob kb8tq kb8tq at n1k.org
Mon Jul 22 12:27:28 UTC 2019


Hi

*Assuming* that we’re talking about 4 to 6x10^-10, you should be able to do better 
than that with the piston trimmer. I’m in no way suggesting that it is *easy* to do
better than that. 

The bigger question is: how much value is there in doing better than that? What is the
aging rate on the OCXO? How does it do in humidity / pressure? ( = when the weather 
changes). Do you leave the counter on all the time? 

Aging can anything from 4x10^-10 / day and down. Probably a 10 day number is a better
target and it’s easier to measure. 4x10^-10 over 10 days is pretty good on a 10811 after
a month or three. 

Humidity and pressure are the two gotcah’s in the 10811’s open box design. Humidity depends
on a lot of things. Dirt on the boards is not good. Component coatings that have cracked over 
the years also not good. Pressure is going to depend if you just had the eye of a hurricane 
pass over. You can get into parts in 10^-10. 

If the counter is power cycled (like to keep the LED’s from dying …) then warmup on the OCXO
comes into play. After a week off, you will be doing well to be within your 6x10^-10 after a couple
days on power. 

======

One somewhat goofy way around the super stable stuff on the EFC is to cut down on the
range / sensitivity of the port. If all you need is 1x10^-9 of range, that’s a lot less than 
what you have on the normal EFC. It’s massively less than what you have on a unit
pulled from the heater shell on one of the GPSDO’s. 

If you do a simple resistor L network to scale the voltage 100:1 (or more) that takes the voltage
at the EFC (with 0-5V drive) down to 0 to 0.05V. Yes, some though needs to go into what you do, but it 
gets it out of the “crazy” end of things and into something you probably can do from junk box
parts. 

For the brave (or crazy) there *is* another route - change the parts around the coarse trimmer
to reduce it’s range. Yes you have to do it “just right” to keep the oscillator on frequency. 
The “old school” approach was to hand grind single plate caps for the final part that went
in. Oddly enough NASA was fine with that approach. 

Lots of fun !!

Bob



> On Jul 21, 2019, at 9:53 PM, Perry Sandeen via time-nuts <time-nuts at lists.febo.com> wrote:
> 
> Yo Bubba Dudes!,
> I bought a HP 10811 from ebay for $36 and shipping.
> I installed it in my HP 5335a counter.  After running continuously for about 90 days it was down to about 4 to 6  parts in -10.  I couldn't adjust it for better accuracy as HP in their infinite and inscrutable wisdom provided a edge connector for the 10811 but didn't make provision for EFC one had only the course adjustment.  Sheesh!
> Now maybe I was very lucky in my ebay purchase.  On ebay prices seem to go from $125 to around $60.  One of the lower priced sellers advertise that their 10811's are *tested* but no details.
> Regards,
> Perrier
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.





More information about the Time-nuts_lists.febo.com mailing list