2

I am currently studying the textbook Practical Electronics for Inventors, Fourth Edition, by Scherz and Monk. In section 2.4.1 Applying a Voltage, the authors have written the following:

In the case of alternating current, the field reverses directions in a sinusoidal fashion, causing the drift velocity component of electrons to swish back and forth. If the alternating current has a frequency of 60 Hz, the velocity component would be vibrating back and forth 60 times a second. If our maximum drift velocity during an ac cycle is 0.002 mm/s, we could roughly determine that the distance between maximum swings in the drift distance would be about 0.00045 mm. Of course, this doesn’t mean that electrons are fixed in an oscillatory position. It means only that the drift displacement component of electrons is — if there is such a notion. Recall that an electron’s overall motion is quite random and its actual displacement quite large, due to the thermal effects.

enter image description here

enter image description here

I'm wondering how the authors concluded that the distance between maximum swings in the drift distance would be about 0.00045 mm? What is the calculation that was done here?

I would appreciate it if someone would please take the time to clarify this.

The Pointer
  • 1,261
  • 1
  • 14
  • 26
  • 1
    The information you have provided in your question is not sufficient to verify those results, in order to calculate the drift velocity in a wire you need to know the current in the wire, the diameter of the wire and the material of the wire, as the drift velocity is determined by the current in the wire in amps and the number of free electrons in the wire per length unit, which can be calculated if you know the diameter of the wire and the material. All these are parameters that you have not given in your question or in the picture. –  Jan 01 '20 at 22:26
  • @Vinzent My apologies for any confusion. As I said, it isn't clear how the authors calculated that value, so I didn't realise that I would need to include the information from the diagram on a previous page. I have now edited my post. – The Pointer Jan 01 '20 at 22:30
  • 1
    @ThePointer You can see some thoughts here. Just FYI. Not an answer. – jonk Jan 02 '20 at 06:53
  • 2
    According to my calculations for 0.002 mm/s maximum velocity at 60Hz the amplitude should be 0.002 / 2pi*60 = 0.0000053mm. I have no idea how they got a value 85 times higher. – Bruce Abbott Jan 02 '20 at 10:17
  • @jonk Thanks for that. Can the calculation you used to find $$\frac{1:\text{A}}{1.3476\times 10^{4}:\frac{\text{Coulomb}}{\text{cc}}:\cdot: 0.51887: \text{mm}^2}\approx 143 :\frac{\mu m}{s}$$ also be used to find the distance between maximum swings? If so, then what is the calculation to find the cross-section of the copper wire, as you did in your answer to that question? – The Pointer Jan 02 '20 at 12:09
  • @BruceAbbott Hmm, that's strange. So the general formula to find the distance between maximum swings is $$\dfrac{\text{maximum drift velocity}}{2\pi \times \text{AC frequency}} = \text{distance between maximum swings}$$? And you're sure that all of the units are correct? It could be an error on the authors' part. – The Pointer Jan 02 '20 at 12:17
  • I used the formula for Simple Harmonic Motion. I suppose it's possible that electron drift in a wire don't follow this, but the only example I could find used the same formula that I did. – Bruce Abbott Jan 02 '20 at 21:03
  • @BruceAbbott I figured it out (see my answer). – The Pointer Jan 05 '20 at 14:40
  • @jonk In an unrelated comment, how does one format mathematics inline on this website (using single $ on either side instead of $$ on either side)? – The Pointer Jan 05 '20 at 14:42
  • 1
    Use $ for inline mathjax https://electronics.meta.stackexchange.com/questions/5565/most-common-mathjax-uses-in-electrical-engineering (they want us to use mathjax, but this is all the help they give us...) – Bruce Abbott Jan 05 '20 at 16:03
  • @BruceAbbott Ahh, thanks for that! – The Pointer Jan 05 '20 at 16:33
  • 1
    @ThePointer I'm glad Bruce got you to an answer. It's worth learning. If you have ever used Latex, it's like a "some stuff removed to make it safer" version. You can use the simpler portions of Latex with some confidence. – jonk Jan 05 '20 at 20:24
  • @BruceAbbott User Philip Wood from physics.stackexchange just answered that he thinks that your answer is the correct one physics.stackexchange.com/a/523322/141502 – The Pointer Jan 05 '20 at 23:42

2 Answers2

2

Recall that displacement \$d\$ is the area under the velocity curve. For a sinusoidal drift velocity \$v_d\$ having radian frequency \$\omega=2\pi f\$ where \$f=60\,\text{Hz}\$, the magnitude of maximum displacement over one half cycle can be calculated as the integral of \$v_d\$ with respect to time, during the time interval \$(0 \le t \le \pi/\omega)\,\text{s}\$:

$$ \begin{align*} d &= \int_{0}^{\pi/\omega}v_d\,dt,\;\;v_d(t) = J(t) / (\rho_e\,e)\\ &= \frac{1}{\rho_e\,e}\int_{0}^{\pi/\omega}J(t)\,dt,\;\;J(t) = I(t)/A\\ &= \frac{1}{\rho_e\,e\,A}\int_{0}^{\pi/\omega}I(t)\,dt,\;\;I(t) = k\,\sin (\omega t)\\ &= \frac{k}{\rho_e\,e\,A}\int_{0}^{\pi/\omega}\sin(\omega t)\,dt\\ &= \frac{2\,k}{\rho_e\,e\,A\,\omega} \end{align*} $$

where \$k=0.1\,\text{A}\$ (as specified in the book example).

For what it's worth, when I crunch the numbers with MATLAB (see Listing 1 and Figure 1 below) the calculated displacement—i.e., drift distance—is approximately 12 nm; so I'm not sure how the authors arrived at the value 450 nm for the drift distance.

See also:


Listing 1. MATLAB source code

%% Housekeeping
clc
clear

%% Givens
d = 2.05e-3;            % wire diameter, m
r = d/2;                % wire radius, m
A = pi*(r^2);           % wire cross-sectional area, m^2

q = 1.602e-19;          % electron charage, C
                        % (NB: This is 'e' in the equation above).

n = 8.46e28;            % estimate of the number of charge-conducting 
                        % electrons per cubic meter in solid copper
                        % (NB: This is 'rho_e' in the equation above).

k = 0.1;                % Sinusoidal current amplitude, peak
f = 60;                 % Sinusoidal current frequency, Hz
w = 2 * pi * f;         % Sinusoidal current frequency, rad/sec

%% Equations
% Current in the wire, C/s
I = @(t)  k * sin(w*t);

% Current density in the wire at time t, C s^-1 m^-2
% J = I/A = k*sin(w*t)/A = k/A * sin(w*t)
% Let k2 = k/A
k2 = k/A;
J = @(t)  k2 * sin(w*t);

% Average electron drift velocity at time t, m/s
% vd = J/n/q = I/n/q/A = k*sin(w*t)/n/q/A
% Let k3 = k/n/q/A
k3 = k/n/q/A;
vd = @(t)  k3 * sin(w*t);

% Average electron displacement at time t, m
% displacement = k/n/q/A/w * (1 - cos(w*t))
% Let k4 = k/n/q/A/w
k4 = k/n/q/A/w;
displacement = @(t)  k4 * (1 - cos(w*t));

%% Solutions
% For sin(w*t), max drift velocity occurs at w*t == pi/2 -> t = pi/2/w
vd_max = vd( pi/2/w )
    % 2.2355e-06 -> ~2.2 um/s

% Maximum average displacement of an electron during 1/2 cycle of 60 Hz 
% can be calculated as the area under the drift velocity curve during 
% the time interval (0 <= t <= pi/w) sec
% NB: For sin(w*t), 1/2 cycle occurs at w*t == pi -> t = pi/w
displacement_max = integral(vd, 0, pi/w )
    % 1.1860e-08 -> ~12 nm


%% Plot the velocity and displacement curves vs time
clf('reset')

% NB: For sin(w*t), 1/2 cycle occurs at w*t == pi -> t = pi/w
t_ = linspace( 0, pi/w );

% drift velocity in micrometers/sec at time t
vd_t = vd(t_) * 1e6;
yyaxis left
plot(t_, vd_t)

% displacement in nanometers at time t
displacement_t = displacement(t_) * 1e9;
yyaxis right
plot(t_, displacement_t)

yyaxis left
title('Velocity and Displacement vs time')
xlabel('Time (sec)')
ylabel('Velocity (um/s)')
yyaxis right
ylabel('Displacement (nm)')
grid on

MATLAB plot of electron velocity and displacement vs. time

Figure 1. MATLAB plot of electron velocity and displacement vs. time.

Jim Fischer
  • 3,074
  • 11
  • 15
  • Thanks for the answer, Jim. Since no one seems to be able to get the same result as the textbook authors, unless the physics guys come up with something, I think it's likely that it is an error. – The Pointer Jan 05 '20 at 23:05
  • User Philip Wood from physics.stackexchange just answered that he thinks that Bruce Abbott's answer is the correct one https://physics.stackexchange.com/a/523322/141502 – The Pointer Jan 05 '20 at 23:40
  • Jim, Phillip says that you seem to be working from a peak current, $k$, of $0.1A$, which doesn't seem to be relevant to the problem in hand. – The Pointer Jan 06 '20 at 00:19
  • The current value is relevant. Current is a function of voltage, and it's the electric field through the wire (caused by the potential difference across the wire) that determines the force that's applied to the electrons, and therefore the drift velocity of the electrons, and therefore the displacement of the electrons. – Jim Fischer Jan 06 '20 at 02:14
  • Phillip said the following: "The stuff in the shaded box headed "12 gauge copper.." in the page that you reproduce is about a wire that carries a constant current. It doesn't seem relevant to your question." – The Pointer Jan 06 '20 at 02:17
  • And for what it's worth, this solution does not take into account real world factors such as skin effect, temperature effects, and so on. – Jim Fischer Jan 06 '20 at 02:32
  • Understood. Thanks for the clarification, – The Pointer Jan 06 '20 at 02:33
  • Jim, Phillip replied with the following: "No it isn't! I supported the use of $x_{max}=v_{max}/\omega$, applied to a particle moving back and forwards in a straight line with simple harmonic motion (displacement a sinusoidal function of time). This follows from the standard equation for such motion, $x=x_{max} \sin(\omega t)$, in which $\omega=2\pi f$. [If a particle is moving in a circle at a constant speed the projection of its motion on to a diameter is simple harmonic, but the simple harmonic motion itself is in a straight line!]" – The Pointer Jan 07 '20 at 18:40
  • My mistake. I must've misread Phillip's post. I've deleted that comment. – Jim Fischer Jan 08 '20 at 03:56
  • Thanks for the clarification. So whose answer should I be viewing as the correct one? Yours or Phillip’s? – The Pointer Jan 08 '20 at 04:20
-1

I just saw the answer to this question by user freecharly. They claim that the mean drift velocity of electrons is $$v = \dfrac{j}{n e},$$

where

$$j = I/A$$

is the current density for current \$I\$ and cross-sectional area \$A\$, and the electron density is \$n\$. If I'm not mistaken, applying this gives us

$$\dfrac{\frac{3.02 \times 10^4 \text{A}}{\text{m}^2}}{\frac{8.5 \times 10^{28}}{\text{m}^3} \times 1.602 \times 10^{-19} \text{coulombs}} = 0.00000221781\text{m/s},$$

since \$1 \text{A} = 1 \ \frac{\text{coulomb}}{\text{s}}\$

EDIT: Ok, I just noticed that this calculation is for the maximum drift velocity, and for a direct current cycle, whereas what we're trying to find is the distance between maximum swings in the drift distance for alternating current. I seem to have confused myself and forgot precisely which result it was that I was trying to find. My apologies for any confusion caused.

The Pointer
  • 1,261
  • 1
  • 14
  • 26
  • That's the drift velocity for DC current, right? How does 'distance between maximum swings in the drift distance' of an AC signal relate to that? – Bruce Abbott Jan 05 '20 at 16:15
  • @BruceAbbott Uhh, is it? Haha, I honestly have no idea; after all, I'm the one who asked the question. But it produces the same value, right? So either we're misunderstanding something, or the author made the incorrect calculation. Given how confusing this has been, I'm going to ask the people on physics.stackexchange for clarification on this. – The Pointer Jan 05 '20 at 16:20
  • I think the downvote is unreasonable. The result is the same value as that of the textbook authors'. – The Pointer Jan 05 '20 at 16:36
  • Perhaps someone downvoted because you didn't answer the question? (wasn't me!). Unfortunately some people do it without any explanation, even when the reason is not obvious. – Bruce Abbott Jan 05 '20 at 16:37
  • @BruceAbbott Hmm, what do you mean? It seems/seemed to me that this mostly answers the question -- although, as you said, apparently this might be the calculation for electron drift for DC, rather than AC. Even so, it's a critical result in answering the question, since we were unable to make sense of this at all with our prior efforts, meaning that the question was probably going to be left unanswered indefinitely. – The Pointer Jan 05 '20 at 16:40
  • 1
    It maybe half-answers the question, but not the most important part. I couldn't explain how they got the 0.00045 mm, and I looked up the source to check that there wasn't something missing or taken out of context. But my physics knowledge is not deep so I didn't risk putting up a wrong answer. Hopefully the physics guys will be able to clarify... – Bruce Abbott Jan 05 '20 at 17:01
  • @BruceAbbott Yes, I have asked here https://physics.stackexchange.com/q/523227/141502 . Thank you for all of your help, Bruce; I greatly appreciate it! – The Pointer Jan 05 '20 at 17:02
  • @BruceAbbott Ok, I just noticed that this calculation is for the maximum drift velocity during an alternating current cycle, whereas what we're trying to find is the distance between maximum swings in the drift distance for alternating current. I seem to have confused myself and forgot precisely which result it was that I was trying to find. Now I understand what you were referring to here in the comments. My apologies for any confusion this caused! – The Pointer Jan 05 '20 at 23:04
  • 1
    Ok, according to Phillip Wood's answer https://physics.stackexchange.com/a/523322/141502, it seems that my calculation of $0.00000221781m/s$ is actually for direct current, which is precisely what Bruce said, and not relevant to the calculation of the distance between maximum swings in the drift distance for alternating current. – The Pointer Jan 05 '20 at 23:45
  • Furthermore, Phillip says "The stuff in the shaded box headed "12 gauge copper.." in the page that you reproduce is about a wire that carries a constant current. It doesn't seem relevant to your question." I wonder if perhaps this is where the authors confused themselves and made an error? – The Pointer Jan 05 '20 at 23:49