[R-390] Silver Mica versus C0G/NPO ceramic ?
Charles P. Steinmetz
charles_steinmetz at lavabit.com
Mon Jan 9 22:19:56 EST 2012
Bill wrote:
>A quick review of the specs appear to be as good or better than silver mica
> - temp comp equivalent or better
> - voltage stable, ie. Value doesn't vary with voltage (only C0G)
>
>So if a silver mica was used in the old gear why not replace it with a C0G
>ceramic?
> -e.g. oscillators, tuned circuits, IF transformers.
The two parameters you mention -- tempco (temperature coefficient of
capacitance) and voltage coefficient of capacitance -- are only two
of a myriad of capacitor characteristics. Both are frequently of
little import (lots of circuits wouldn't show any perceptible
difference if the capacitance varied by a factor of 5 up or
down). Two that are often of great importance are dielectric
absorption and dissipation factor. The choice of capacitor depends
on which of these many parameters are most important in the
particular application.
For general tuned circuit duty where temperature compensation is not
required, C0Gs are fine capacitors. (Although C0Gs have a near-zero
tempco, inductors and crystals most assuredly do not -- so,
ironically, by using the "better" near-zero-tempco part, you are
guaranteed of having greater temperature sensitivity in the circuit
than if a capacitor with an appropriate non-zero tempco were used to
compensate the overall circuit tempco.) For oscillators, you will
almost always want capacitors with a particular tempco, to offset the
tempcos of other components. There, you would choose from the "N"
and "P" series of ceramics (for example, the N750 series has a
negative temperature coefficient of nominally 750 ppm/degree C).
Some choices are driven by practical considerations, such as what
manufacturers produce. One advantage of silver micas is that they
are readily available in tight tolerances (1%, 0.5%, 0.25%). Not so
with ceramics of any kind, including C0Gs. There is also not as good
a selection of C0Gs at 500 V and above as there is with micas.
Best regards,
Charles
More information about the R-390
mailing list