[time-nuts] Possible HP 10811 instability clue Cont’d.

Bruce Griffiths bruce.griffiths at xtra.co.nz
Fri Oct 8 00:50:11 EDT 2010

Perry Sandeen wrote:
> List,
> Wrote<Some on this list mentioned having had less-than-optimal performance with their HP 10811 series oscillators.  James Miller G3RUH made a slight mention of optimizing performance by readjusting the set point temperature in one of his phase detector articles.  He said though time consuming the results were worth the effort.
> Replied:<Regardless of isolated anecdotal data on one oscillator, it is probably not advisable to change the set point.
> Answer: Agree.  My comments were made about someone with a “wonky” oscillator that that this MIGHT be part of the problem.
> Replied: The majority of 10811 crystals do NOT have a turnover, only a region of low tempco around 82 degrees.
> Answer:  I respectfully disagree.  First isn’t a region of low tempco another way of saying turning point?
Not so.
A turning point of the frequency vs temperature curve is a location 
where the the tempco is zero.
It may be located at a maximum, at a minimum or at a point of inflection.
>   Semantics aside, I believe that they do have a turning point that is selected.  This is why.   HP selects 41 separate resistance values to obtain 41 different oven temperatures between 80.0C to 84.0 C in .1 degree increments.  Now if that isn’t finding a turning point I don’t know what is.
Not so, it merely reflects that there is a temperature, for a particular 
10811A crystal, at which the tempco is minimised somewhere in the (80C, 
84C) range.
It also indicates that an accuracy of around 0.1 C or better in locating 
this minimum is desirable.
Thermistor tolerances will increase the apparent temperature range somewhat.

> I believe that is why they say if the oscillator fails it needs to be repaired at the factory.
> Wrote:<Instead of that, change the circuit to B-mode and optimize the heat between the two heater resistors for maximum thermal gain (you should be able to get 1000).  Then change the circuit back to normal.
> Answer:  I have no idea B-mode is.  Would you please expand on it and how to do it?  It could be very useful.
The B mode is a mode of oscillation of the crystal which has a 
relatively large tempco and thus it can be used to measure changes in 
the crystal temperature.
The normal mode of operation is the C mode which has a much lower tempco.
To allow oscillation in the B mode the mode suppression components in 
the 10811A oscillator need to be disconnected.

For some further detail on this read the HP journal article on the 10811A.
> Regards,
> Perrier

More information about the time-nuts mailing list