Impedance

[stan001]

I am a beginner in electronics and have a hard time trying to understand the concept of impedance..

The below statement :-
Arduino (Atmega) pins configured as INPUT with pinMode() are said to be in a high-impedance state.

Is high impedance state equal to high resistance ??

What is the relationship between impedance and resistance ?

How do I know if a sensor or load is high or low impedance ??

Hope someone can help to clarify this or give some examples...

[tyblu]

impedance: Z ; admittance: Y = 1/Z
resistance: R ; conductance G = 1/R
reactance: X ; susceptance B = 1/X

Z = R + jX ; Y = G + jB

Impedance takes into account frequency-dependent (frequency reactant) nature of a load. It's probably a good idea to write the above on the ceiling above your bed, or somewhere common. A resistor has impedance of Z = R; a capacitor has impedance of Z = Inf. (open-circuit) at DC, Z = -j/wC with AC; an inductor has impedance of Z = 0 (short-circuit) at DC, Z = jwL with AC. It is more useful to define impedance instead of resistance, as it tells the reader how it behaves with different frequency electrical signals.

[Time]

Impedance is how something resists change in voltage or current.  If its not changing than there is nothing to impede.  This is due to the field effects of changing voltages and currents.  Fundamentally its all based around a set of equations known as Maxwell's equations which turn out to be quite elegant and beautiful in their own right.  Its a kind of difficult concept at first but if you are a student you will eventually take electromagnetics for engineers and this will shed light onto the concept.  

Something that is purely a resistance will resist the flow of electrons at some constant regardless of how its varying with time.  If you want to physically conceptualize this think of electrons flowing through something and bumping into the lattice of the bonded atoms.  The more they bump into things the slower they will flow which produces heat as a direct result.

As a whole an impedance is both how much electrons bump into the structure (resistance) and how much the fields from the changing current effect the flow of the electrons themselves.

Now, nothing is purely resistive or purely reactive (reactive being the portion of the impedance that accounts for only the field effects).  In real life everything has some finite resistance, inductance, and capacitance.  

So, with that being said when talking about devices you can do a little analysis and find that you essentially want infinite impedance looking into the input of the device and 0 output impedance looking into the output of the device.  This is something you will also learn about in school.  The reasoning for this is related how much energy is transferred from the output of something into the input of another thing.

If you want to know about the impedance of the input/output of a sensor you can usually find it in the manual.  If it does not say in the manual its usually safe to assume the input is high-Z or high impedance.

[williefleete]

impedance is indeed similar to resistance, in fact in a dictionary they would have basically the same meaning
however as tyblu said resistance is a DC property impedance is more an AC property
a high impedance would be open circuit to DC and AC signals eg almost nil capacitance (picofarads at worst) and infinite resistance, inputs on micros and CMOS logic have this property hence the need for pull down resistance as the pin, case and its ground/positive acts as a parasitic capacitor, charges can be stored and trigger a false detection, a pull down merely discharges the parasitic capacitor to ground relatively quickly, lower the value the faster it discharges after a pulse

[stan001]

Hi,

Thanks for all the replies..

I think my question is related to DC circuit instead of AC circuit.

Here is my understanding of impedance using the voltmeter example...

Voltmeter have high impedance so that it would not alter the voltage or current of the circuit it is measuring ...

Is the above statement a good example to explain about "high impedance" ??

Thanks

[GeoffS]

A good site to checkout the fundamentals is All About Circuits.

[tyblu]

Voltmeter have high impedance so that it would not alter the voltage or current of the circuit it is measuring ...
Is the above statement a good example to explain about "high impedance" ??

Yep, this is good. Impedance is equal to resistance when restricted to DC circuitry. High resistance circuits impede current flow.

[Mechatrommer]

Voltmeter have high impedance so that it would not alter the voltage or current of the circuit it is measuring ...
Is the above statement a good example to explain about "high impedance" ??

its good, but not good enough. if you want to restrict your mind into DC, then "impedance" word is useless (or redundant, or will mislead imo), if you just want a dc, stick to "resistance".

The below statement :-
Arduino (Atmega) pins configured as INPUT with pinMode() are said to be in a high-impedance state.
Is high impedance state equal to high resistance ??

impedance is more general term (dependent on freq), where resistance only applicable in dc. arduino input is said to be high "impedance", is to ensure users that it will remains high in any frequency given by users/external circuit. someone will be worry to feed the arduino input with higher freq if its only be said to be "high resistance". and for you, with the dc feed, you simply can read the impedance=resistance.

is my explanation not good enough?

[alm]

impedance is more general term (dependent on freq), where resistance only applicable in dc. arduino input is said to be high "impedance", is to ensure users that it will remains high in any frequency given by users/external circuit.

This statement is misleading, there's no such thing as a high-impedance input at 1GHz+ frequencies. For a 10kOhm impedance (not really high) at 1GHz, the parasitic capacitance should be below 16fF (0.016pF).

Atmel lists the input capacitance of an I/O pin as 10pF (for I2C), any capacitance between traces / ground plane will add to this. The minimum rise time (for I2C) is 20ns, which in the frequency domain translates to about 20MHz. At this frequency, even a bare input without any PCB trace connected will only have an impedance of 800ohm. So it's not even high impedance for these frequencies.

someone will be worry to feed the arduino input with higher freq if its only be said to be "high resistance". and for you, with the dc feed, you simply can read the impedance=resistance.

Something can have a high AC impedance but low DC resistance, think inductors. So I would be careful with these kinds of assumptions. But at the low frequencies the ATmega is dealing with, I wouldn't worry about significant inductances, so in this case it's correct.

Impedance is usually specified at a certain frequency of interest (or as function of the frequency), so they probably mean that they consider the impedance high for the frequencies that they expect. If you connect it to a faster bus, you could get into trouble, even with the pin in Hi-Z state. Even the impedance at the highest frequency supported by the ATmega is pretty low, about an order of magnitude or more below most pull-up/down resistors.

[Time]

Voltmeter have high impedance so that it would not alter the voltage or current of the circuit it is measuring ...

Is the above statement a good example to explain about "high impedance" ??

Ok, you are close.  If you are measuring voltage you probe the component of interest by placing the test leads PARALLEL.  You want to minimize the impact of the meter itself, so if you think about it enough you can see that for voltage measurements you want a very high impedance. (Placing a much larger impedance parallel to the component will cause very little current to flow out of the circuit and into the meter)

So now it is the opposite when measuring current with a meter.  To impact the functioning of the circuit as little as possible, you want as close to zero impedance as possible since the measurement is taken in SERIES with the component. (Placing a much smaller impedance in series to the component will cause very little change in the current flow of the circuit)

EDIT: On second thought, your statement is correct if you are only using a voltmeter and are not talking about measuring current.

[stan001]

impedance is more general term (dependent on freq), where resistance only applicable in dc. arduino input is said to be high "impedance", is to ensure users that it will remains high in any frequency given by users/external circuit. someone will be worry to feed the arduino input with higher freq if its only be said to be "high resistance". and for you, with the dc feed, you simply can read the impedance=resistance.

is my explanation not good enough?

Yes, thank you for yr detail explanation...

Time, yes, I understand it much better when reading on how to take measurements using voltmeter and ampmeter as an example for DC circuits..

http://www.ibiblio.org/kuphaldt/electricCircuits/DC/DC_8.html

Thank you all for the replies..

[Simon]

I think on the whole people use the word impedance instead of resistance for unknown devices or basically anything not known to be a resistor as it covers everything, it will account for pure resistance plus inductive and capacitive reactance

[Zero999]

I think on the whole people use the word impedance instead of resistance for unknown devices or basically anything not known to be a resistor as it covers everything, it will account for pure resistance plus inductive and capacitive reactance

And for weird non-linear loads who's impedance varies depending on the current/voltage i.e. diodes, neon lamps, incandescent lamps etc.

[scrat]

If the question was about impedance as the third "logical state" (since it comes from an MCU datasheet)...

High-impedance is considered as a kind of third logical state, in the sense a so called "3-state" logical pin can be 1, 0 or not driving the output, which is "high-impedance". This way you can use a logical pin as an output or an input (for some other circuitry),  otherwise there could be a conflict between the voltage the pin is trying to output and what an external device attached is trying to drive.
In the silly image I attached, just think that an output pin has a set of controlled switches (mostly MOSFETs, nowadays). A 1 output level is obtained by turning ON the only high switch, a 0 turning on only the low switch, high-impedance ("Z") by leaving both OFF.

Electrically, you can assume that the impedance that an external device sees on a logical pin in the high-impedance state is high enough not to be read as a logical 1 nor a 0 for the frequency interval the device is meant to work.