Re: Varactor ?

On Sat, 08 Nov 2008 06:37:18 -0600, Dave wrote:

"Andrew Holme" <ah@xxxxxxxxxxxx> wrote in message
news:VjeRk.17592$uq3.6030@xxxxxxxxxxxxxxxx

"Dave" <db5151@xxxxxxxxxxx> wrote in message
news:q92dnUSDJI9JzojUnZ2dnUVZ_sHinZ2d@xxxxxxxxxxxxxxxxxxxxxxxxx
Looking at the data*** for the NTE618, and wondering why it lists a
frequency (1MHz) after each entry for the interterminal capacitance.
Anyone have any ideas on this? I figured the applied voltage would be
DC. No?


They give the mean DC voltage, VR, as well. In operation, the diodes
are subjected to a DC reverse bias with a (hopefully) smaller AC signal
component superimposed. You have to apply an AC signal to measure
capacitance! 1 MHz is a good test frequency for an AM radio tuning
diode.


Hmmm. Well, thanks for the reply... Now I just have to try and
understand why an AC signal is necessary for measuring capacitance. Not
as smart as I thought I was. I don't remember covering varactors in
school (30 years ago!) but I thought I understood the principle. If you
would be so kind, why *is* an AC signal necessary for the measurement of
capacitance?

Thanks again,

Dave

First, because when you're dealing with a device whose capacitance varies
with voltage, the exact definition of capacitance can get a bit fuzzy.
Second, because the varactor doesn't act like a perfect capacitor, and
it's apparent capacitance can vary by frequency.

The "DC" definition would be C = q/V, so (assuming you could measure
charge accurately) that implies that you would measure the charge and
voltage, divide the measured charge by the voltage, and get the
capacitance value.

That would be nearly useless, because you don't care about the total
capacitance of the diode -- what you care about is it's incremental
capacitance about the given bias voltage:

C(Vb) = dq/dV |
| V = Vb

One very reasonable way to measure this is to impose a sinusoidal voltage
(or current) at some frequency, and measure the current (or voltage) that
results. If you do this, then you'll also take care of those parasitic
effects (mostly resistive losses at 1MHz, although at higher frequencies
the package inductance would become important). Because the parasitic
effects change the apparent capacitance, you want to measure the thing at
the frequency it's going to be used -- hence 1MHz for a diode that's
being marketed for AM radio use.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html
.

Quantcast