On Wed, 26 Nov 2003 04:59:56 GMT, "Rick Karlquist N6RK"
wrote:


"J M Noeding" wrote in message
...
, he also mentions
using two MOSFETs to form a variable capacitor with maximum
capacitance of 1500pF

Why bother with MOSFETs? You can get varactor diodes used for tuning
AM broadcast receivers with that much capacitance.

What is the capacitance tolerance for such varactors and what is the
temperature coefficient ? These can be quite a problem in any
oscillator circuits.

To get such huge capacitance, the control voltage must be quite low.
In LF receiver front ends, the total RF voltage from any decent sized
antenna can be quite large, modulating the varactor voltage and hence
capacitance. This sounds like a varactor multiplier :-). Such
behaviour would not be so nice from intermodulation point of view.

In the worst case, the varactor could even be driven into conduction
due to a large applied total RF voltage, so a pair of back to back
varactors would have to be used. But with two capacitors in series in
the resonant circuit, the capacitance drops to one half of a single
varactor capacitance, thus, you would have to use two pairs of back to
back (4) varactors of 1500 pF each to get a total 1500 pF capacitance.

Even a larger number of varactors would be preferable, so a higher
minimum voltage could be used, with less risk for RF generated tuning
voltage modulation, but then the minimum capacitance would also
increase due to the limited maximum tuning voltage characteristics of
the varactor maximum available tuning voltage supply. I guess that the
larger varactors are quite expensive, so this would also limit the
number of units.

Switching in some fixed capacitors (e.g. in a binary sequence) avoids
some of these problems, but it would be quite hard to operate the
circuit with a large number of fixed capacitors.

Paul OH3LWR