[NLRS] DB6NT PTC crystal heater

[Dr. Gerald N. Johnson, electrical engineer]

A classical crystal oven would surround the components with a copper
container and apply heat to that copper container with a winding of
heating wire covering most of that container. That makes the temperature
of the entire interior of the container quite uniform because the copper
conducts heat quite well. Then the insulation on the outside of the
container and that heater winding, reduces the heat losses and thus the
power requirements for maintaining the temperature. The thermal capacity
of the copper also helps slow temperature changes. But the long term
thermal stability depends on the oven control circuit. For really high
precision oscillators (like parts in 10^-9), the first layer of
insulation is often a glass dewer (like a thermos bottle), with a second
thermostat controlled heater outside of the thermos bottle and foam
insulation outside that. The double oven is a benefit.

With the copper inner oven enclosure (and no heat leaks past it like a
chunk of circuit board) its not too necessary to connect heat directly
to crystal. When the only components heated are crystal and thermistor,
surrounding them with insulation is fairly important. I think that the
other reactive components in the oscillator need as much attention as
the crystal if you really want to achieve stability. The tuning
capacitor or coil can probably move the frequency as far as the crystal
would over temperature, perhaps further. The modern NPO capacitor isn't
as good for temperature stability as the older temperature compensating
capacitors that are getting extremely hard to buy.

73, Jerry, K0CQ

-- 
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.