Stefan Wolfe wrote:
"Roy Lewallen" wrote in message
...
You've lost me. What is the meaning of a "time constant" in steady state?
What effect does it have? With a single frequency of constant amplitude,
how could you tell whether a circuit, resonant or not, has a "time
constant"? How could you measure it?

The meaning of a time constant is not dependent upon steady state sinusoids
or transient; it is merely a characteristic that dependent upon the
*physical* properties of the components in the circuit.

You know that of course.

Let us say we design a power supply for use in consumer appliances. The
power supply of course has a capacitor across line and neutral for EMI
filtering, along with a bleeder resistor in parallel with this capacitor.
Together, the bleeder resistor and phase/phase capacitor filter have a time
constant. Now the time constant is meaningless with respect to steady state
input voltage (other than wave shaping high f emi components) and with
respect to transients. It is meaningful with respect to safety. If the
consumer pulls out the plug, the bleeder resistor must discharge the phase
to phase cap safely to prevent the consumer from being shocked when touching
the L-N pins. The time constant RC in this case MUST meet certain
specifications, that is it must be less than 0.1sec. That is essentially
required legally (since standard compliance is compulsory). The time
constant exists is chosen for a worse case value, ie that the consumer
unplugs the power supply at the peak of the AC cycle.

Actually, it is the DC discharge characteristic that we care about here.
Transient suppression is not relevant nor is its ability to shape the
incoming sinusoid.

Sorry, you've completely baffled me again. I can't find any relationship
at all between what I asked and what you wrote. There's nothing more I
can contribute, so I'll return the readers to their regularly scheduled
programming.

Roy Lewallen, W7EL