All,
Capacitors exhibit reactance when excited by alternating current.
This reactance opposes changes in voltage. If the capacitance
value remains constant, and frequency is increased, the amount of
this reactance goes down. Likewise, the opposite is true, that
if the capacitance remains constant, and frequency is decreased,
reactance goes up. This shows that Capacitive Reactance is
Inversely Proportional to Frequency. Now, lets say we have chosen
our capacitor value to give us a specific reactance value at a given
frequency. And this frequency is the lowest we have to deal with.
In this case, as frequency goes up, reactance goes down, and the
impedance gets lower. In bypassing, this means that as frequency
goes up, the capacitor acts as more of a short than at lower
frequencies.
So, the statement
"This suggests that lower capacitance a priori works better at
higher speeds."
is somewhat correct. What it really means is that you need less
capacitance at higher frequencies to get the same value of reactance
that you get with a larger capacitor at lower frequencies.
The statement:
My understanding of the specifications suggests the opposite: for
any make of capacitor, higher capacitance always works better at
any frequency.
also is somewhat true. From a capacitive reactance point of view,
the higher the frequency, the less the capacitive reactance of any
capacitor. Therefore the above statement is true. But, there are
other losses that are also frequency dependent. The most obvious is
INDUCTANCE. This is directly proportional to frequency, and opposes
a change in current. So, at some point, what you gain in lower
capacitive reactance will be offset some by an increase in Inductive
reactance. Therefore, you have the potential for any degree of
inefficiency. So in this respect, the above statement is not true.
The statement:
What might be true is that certain types of capacitors are only
available in small denominations, and these types have lower ESR
and ESI so they work better at higher frequencies.
Is basically true. ESR and ESI have the most effect on Q, which
translates to effeciency. Of course this means overall effeciency,
not just at a given frequency or range of frequencies. It reduces
the amount of heat generated inside tha cap which means that less
energy is consumed by the cap and transformed into heat, which
indicates higher effeciency.
Just my 2 cents worth. Sorry for butting in uninvited.
Regards,
Jim
On Mon, 05 June 2000, "Robert A. LaBudde" wrote:
{Quote hidden}>
> At 10:49 AM 6/2/00 -0700, Dave wrote:
> > >Has anyone found/experimented with the optimum size capacitance
> > >value and/or physical size for use in/around a microprocessor?
> >
> >You want the caps to absorb well at the third harmonic of the clock.
> >0.1uF does well at 3 MHz, 0.01 at 30. 0.001 at 300.
> >It's a broad response, so dont think that there's one specific value.
> >However, if you use 0.1uF on a 20 MHz part, you won't get the supression
> >that you could if you used 0.047uF.
>
> This suggests that lower capacitance a priori works better at higher speeds.
>
> My understanding of the specifications suggests the opposite: for any make
> of capacitor, higher capacitance always works better at any frequency.
>
> What might be true is that certain types of capacitors are only available
> in small denominations, and these types have lower ESR and ESI so they work
> better at higher frequencies.
>
> What's important is that the impedance of the capacitor be only 1 ohm or so
> at the frequency of interest to reduce ripple at the current draw required.
spam_OUTjimTakeThisOuTjpes.com