[time-nuts] Oscillator temperature compensation

[Magnus Danielson]

On 06/21/2013 06:59 AM, Joseph Gray wrote:
>> Can you show some pictures of the oscillator?
>
> The wiring is point-to-point, so I don't think a picture is going to tell
> you much.
>
>> Is there a tunable inductor in the oscillator circuit?
>
> As I mentioned, nothing tunable there.
>
>> Who makes the unit?
>
> It is an LPB RC-6A carrier current AM transmitter. It was used at the local
> university many years ago. I was told they used to have several. I am
> rescuing it from oblivion.

The closest to a schematic I find is this:

http://i30.photobucket.com/albums/c343/1073Dave/Schematics/LPBRC-5ATransmitter.jpg

Hooking a trim-pot to either of the caps next to the crystal should 
allow you to trim it, should be a good start.

If you really need temperature compensation, a first degree compensation 
should help. I haven't seen anything matching the X or Y cut crystal you 
most probably have. The only plot I have for an X-cut shows a mostly 
linear shift. A simple trimable first degree compensation should not be 
too hard.

Otherwise you might just as well lock it up instead. Divide 10 MHz down 
to 20 kHz (divide by 500) and 660 kHz to 20 kHz (divide by 33) and then 
a phase-comparator of choice. The varicap is just inserted under the 
foot of one the caps around the crystal oscillator. Use a PI-active loop.

Cheers,
Magnus