From: "R Wilcock"
Subject: Re: output impedance
Date: Fri, 8 Nov 2002 10:27:03 -0000
X-Newsreader: Microsoft Outlook Express 6.00.2462.0000
The current does not flow through the output inpedance of the current
source, that is the point, it should _ALL_ flow through the circuit you are
connecting to (hence an ideal source). More detailed explanation below:
It is somtimes tricky to get a grasp of the principles behind an ideal
component since we are so used to dealing with real life electronics, plus
most peoples view of infinity never helps either.
An ideal current source, if you view it as a block with one terminal (it
must have two in a simulator as otherwise the solution will not converge
because there is no 'dc' path to ground) will push out through that terminal
a specified current, whatever the voltage at the terminal happens to be. So
if you imagine a graph of current (y axis) and voltage (x axis) it will be a
horisontal line, at the specified current. This is a good way of reminding
yourself what the impedance is, because for any change in V the I does not
change, Ohms law says that V=IR, so R=V/I so R is infinite. What this means
really is that no current will ever flow through the current source itself,
instead all the current will flow through the circuit it is connected to.
We normally model a non-ideal current source by having an ideal current
source with a parallel connected output impedance which is purely resistive,
i.e. a resistor. The value of this resistance reflects how much current
flows through the current source itself, depending on the voltage at the
output node, V=IR again. So, going back to the case of the ideal current
mirror, if we use this same model we have to use a resistance equal to
infinity as the output resistance to make sure that the current source does
not steal and of the current. Then whatever the output voltage is that
resistor will never take any current and hence the source acts as ideal. Of
coure the bright engineers will, as some of you have been, say oh but at
infinite voltage there will be a current, and yes there will but although
the current source is ideal, the voltages around your circuit should be
representing what you would like in real life i.e. up to 5v for example.
Clearly with real life voltages the ideal current source will always act
Hope this helps.
"Mick McGrath" wrote in message
> I have learned that the output impedance of an ideal current amplifier
> should be infinite. How can any current flow through infinte