[time-nuts] HP OCXO's 10544A and 10811A

[Richard (Rick) Karlquist]

On 8/5/2019 7:25 AM, Walter Shawlee 2 wrote:
> I have hp counters with both OCXO's, they seem to be electrically and 
> mechanically interchangeable,
> and generally, they both have performed well.
> but it seems to me that the earlier 10544A runs hotter (to the touch), 
> and is not quite as long
> term stable as the 10811A. how have they worked for you?
> 
> is there a comparison of the key characteristics of these two 
> sub-assembles?
> 
> all the best,
> walter
> 

I worked with the 10811 designers when we were at HP, and I learned
as much from them as possible.  They were tasked with making an
completely new and different oscillator (including a new crystal cut),
but with the ability to be able to plug into any legacy 10544 socket.
The 10811 famously used the SC cut for fast warm up, vs the AT
cut used in the 10544.  I remember that the 10811 oven could
never draw more current than would be drawn by a 47 ohm resistor.
This requirement was derived from the 10544 characteristics.
It had to produce 10 MHz at +7 dBm into 50 ohms, etc.

Other than the can, the edge card connector, and the location of the
tuning screw hole, they started with a blank piece of paper in terms
of the oscillator design.  There is no similarity between the two
ovens.  Both oscillators may be offshoots of the Colpitts design, but
the 10811 oscillator has various innovations beyond the 10544.  It
has the patented phase noise reduction circuit, etc.

There was no requirement for 10811 sockets to support 10544's.

Rick N6RK