5

If two conductors (A and B), separated by a distance R, are parallel for a length ℓ and conductor B has an AC current flowing through it (I), what would be the equation to calculate the EMF (voltage) induced in conductor A between points 1 and 2.

To help clarify the question, I will give the following example inputs:

  • Conductor B is AWG 4 (5 mm) diameter single core cable. (It is carrying an AC current of 10 A (RMS) @ 50 Hz)
  • Conductor A is an 8 mil wide 1 oz PCB copper trace
  • The length ℓ between points 1 and 2 is 50 mm
  • The gap between the centre of the conductors A and B (R) is 5 mm
  • A complete path (not shown in the diagram) exists such that current can flow in conductor A
  • The medium between the conductors is air

Diagram

Chris Brown
  • 53
  • 1
  • 4

4 Answers4

2

Consider first that the two wires (of length \$\ell\$) are placed very close to each other such that the distance R is very small (ostensibly zero). Consider also that the driven wire has self-inductance (L) and that this inductance has a reactance that develops a volt drop due to the current flowing: -

enter image description here

Because the two wires (driven and undriven) are very close, we can assume the field generated by the driven wire is wholly shared by the undriven wire and so we get 100% transformer action and, whatever voltage is across the length (\$\ell\$) of the driven wire will be (in fact has to be) across the receiving wire.

The voltage is \$I\cdot\omega L\$ and that voltage rapidly gets smaller as R increases. But, what is the inductance of a wire: -

enter image description here

However, you can use an online calculator for that to get the answer if you don't want a math/algebra solution: -

enter image description here

So that gives you the induced voltage when R is zero. To calculate the voltage when R is non-zero requires understanding of how the coupled flux falls away with an increasing distance R.

This is probably best understood by thinking about how the flux density falls away with distance from a long straight conductor (based on Biot Savart).

Flux density, B = \$\dfrac{\mu_0\cdot I}{2\pi d}\$

Where I is the current and d is the distance from the conductor. If this is integrated from d=0 to d=infinity, you get the total flux and this of course is used to calculate wire inductance as indicated at the top of the answer.

If you then integrate only as far as distance R you get the flux that isn't coupled to the receiving wire. The difference between the two is the flux that couples to the receiving wire and, as a proportion of the total flux that is the coupling factor, k: -

enter image description here

So, you calculate k (based on distance R) and that multiples by the voltage across the driven inductance/wire to give the induced emf.

Andy aka
  • 456,226
  • 28
  • 367
  • 807
  • Comments are not for extended discussion; this conversation has been moved to chat. – Dave Tweed Oct 01 '18 at 12:29
  • This is still not a correct answer unless wavelength or frequency and all the inductive phy. Specs (missing) are given. EMF=0 as given. WHere is the OP anyways? (Hiding) Give a measurable answer with all parameters if you disagree. Your answer computes flux but not voltage in a open conductor. YOur answer computes inductance but without current and Phy and frequency, there is no VOLTAGE. Or EMF – Tony Stewart EE75 Oct 01 '18 at 21:38
  • You don’t know what you are talking about tony. – Andy aka Oct 02 '18 at 06:16
  • @Andyaka, I still can't see how the total flux is not infinite (see chat). Can you explain where I am wrong? – Chris Brown Oct 02 '18 at 08:05
  • @ChrisBrown Maybe a different approach to finding the total flux is needed. For a single wire, as far as I remember, the inductance is defined as flux (not flux density) per amp. So re-arranging, total flux = inductance x current. This possibly solves the infinities in the equation and permits a sensible calculation of flux from close to the wire to R. I'm a bit rusty on the integration of the Biot Savart relationship so sorry I can't be more helpful at this time. Another possible route is to address this specific part of the question (calculating total flux) on the physics part of SE. – Andy aka Oct 02 '18 at 08:24
  • 1 Weber is equivalent to 1 Henry-Amp so I believe that formula holds as long as there is only a single loop. Thanks for your help. – Chris Brown Oct 02 '18 at 09:02
  • @ChrisBrown Give it a go and let me know what the numbers are for total flux and the total flux from driven wire to R. Interesting. – Andy aka Oct 02 '18 at 09:32
  • Would this answer still be valid if the inductance of the two conductors are not equal? (i.e. the mutual inductance does not equal the inductance of conductor B) – Chris Brown Oct 02 '18 at 14:03
  • The answer is only valid over the common length of the two wires. If wire B is longer than wire A then you must calculate the inductance of wire B over the common length. Did you post anything on physics.se? – Andy aka Oct 02 '18 at 14:17
0

Calculate the magnetic rate of flux change induced by the conductor B between the distance R and infinity.

The emf induced in conductor A = N d(phi)/dt. Here N =1, because there is only a single wire. phi is the magnetic flux enclosed by the conductor A.

The current you have provided is 10 Amperes. This is an RMS value, I assume. The peak value, assuming it is a sine wave is = sqrt(2)*10 = 14.41 Amps.

The equation of the current flow in A = 14.41* sin(wt), where w = 2*pi*f *t, f is in HZ (or cycles per second). You have not specified the frequency. Also, the medium in which the conductors are situated. You have to account for the magnetic permeability of the medium.

Here is a python program that calculates the voltage between points 1 and 2:

""" Python program to calculate the voltage induced in a wire by another parallel wire, both of finite length.

The formula given above appears to be wrong. It is possible for the self inductance. What we need here is a formula for the MUTUAL inductance. Formula for the mutual inductance is:

M = mu_0 * 2*l*[ ln[(l/d)+ sqrt{1 + (l^2/d^2) -{sqrt[1 + [d^2/l^2]] + d/l}

mu _{0}= 4π × 10−7 Comment:N·A−2 (Permeability of free space)

This formula can be obtained from:

  1. Professor Marc Anderson: http://www.thompsonrd.com/induct2.pdf

  2. Frederic W. Grover, Inductance Calculations, Dover Publications.

where l = length of the parallel wires, and d is the separation between the wires.

Here, we are assuming that both the parallel wires are of the same length.

In the problem above, d = R.

The current = 10 Amperes. Since the problem states, it is AC, we are going to assume it is sinusoidal and 60 hz.

The voltage induced between terminals 1 and 2 of conductor A is:

v = M di/dt volts.

The instantaneous current in conductor B is:

i = IMax sin(wt), where, w = 2*pi.f, f in hz, and t in seconds.

di/dt = IMax wcos(wt)

f =60 in the United States. IMax = sqrt(2) *10 di/dt = sqrt(2) * 10 *2.0*pi*60* cos(wt) The procedure we are going to adopt is to calculate the peak voltage first and then attach the cos(wt) to it.

""" import math

The Maximum value of I_prime = di/dt = math.sqrt(2)*10.0 *2.0*math.pi*60.0

I_prime_max = math.sqrt(2)*10.0 *2.0*math.pi*60.0

print(I_prime_max)

I_prime_max = 5331.459525790039 Amps = Approximately 5331.46 Amps

Let us calculate the mutual inductance M:

Since no values are given for L and D, the induced voltage between terminals 1 and 2 is : = M* 5331.459525790039* cos(wt), where M is the value of the mutual inductance given by M above. To attach some numerical values, let the length of the wires = 1 meter, the distance of separation be = 0.5 meters. M = mu_0*2*l*[ ln[(l/d)+ sqrt{1 + (l^2/d^2) -{sqrt[1 + [d^2/l^2]] + d/l}

mu_0 = 4* math.pi*10^(-7)

^ stands for exponentiation. * stands for multiplication. l = 1.0 d = 0.5 a = l/d print("a = ", a) b = d/l print("b =", b) M = mu_0*2.0* l *[math.log[a1 + math.sqrt[1 + (a1)^2]] - math.sqrt[1 + (a2)^2] + a2]

A = math.log(a + math.sqrt(1 + a^2))

print("A = ", A)

B = math.sqrt(1 + b^2)

print("B = ", B)

M = mu_0*(2.0 lA - B + b)

print(M, 'Henrys')

M = 2.851607267136502e-06 Henrys

V_peak = M * 5331.46

print("The peak voltage between points 1 and 2 =",V_peak )

print("The RMS Voltage between points 1 and 2 =" , V_peak/math.sqrt(2), "Volts")

The peak voltage between points 1 and 2 = 0.015203230080447576 The RMS Voltage between points 1 and 2 = 0.010750307085823781 Volts or, the RMS(Root Mean Square) Voltage between points 1 and 2 = 10.75 milli-volts.

  • hey Assume $\mu_R=\epsilon_R=1$ and 1.414 and for f, assume wavelength far greater than geometry shown and L>>R – Tony Stewart EE75 Sep 29 '18 at 14:01
  • To: Solar Mike: This is Faraday's law of induction. – user5292153 Sep 29 '18 at 14:03
  • Bandwidth or rise time is critical to the answer not yet displayed – Tony Stewart EE75 Sep 29 '18 at 14:09
  • or as Gauss once said "theorie der Strahlung" – Tony Stewart EE75 Sep 29 '18 at 14:15
  • There is no current flow in A and your answer doesn’t appear to take account of the distance between the two conductors nor their length. – Andy aka Sep 29 '18 at 14:47
  • 10 A was indeed intended to be an RMS current although this was just an example number to indicate the expected range. The expected frequency is 50/60 Hz. It should be assumed that the medium in which the conductors are situated is air. I presume the calculation of the magnetic flux density would take into account the distance between the conductors. And the length taken into account when calculating the flux contained within conductor A. – Chris Brown Sep 30 '18 at 16:49
  • @AndyAka: From the diagram, l = L. Also, the formula for B you have given is for the flux density, which is the flux per unit area. The formula you have given is for the flux density at a midway point (L/2 = l/2). To get the flux enclosed is B multiplied by area da. Please note that B varies at every point. Thus you cannot integrate B. you need to integrate B.(da). It is actually a double integral, parallel to the wire and perpendicular to the wire. The diagrams are very nice indeed! – user5292153 Oct 01 '18 at 20:06
  • I did mention that in a comment to the op but I shall check how clear I made that in my answer. I don't understand when you say " From the diagram, l = L"????? – Andy aka Oct 02 '18 at 06:56
0

Assume \$\mu_R=\epsilon_R=1\$ and there are no other conductors nearby ( which could shunt the mutual inductance.)

Use double integral Neumann formula to get Mutual Inductance

https://en.wikipedia.org/wiki/Inductance#Mutual_inductance_of_two_parallel_straight_wires

then V or EMF=Ldi/dt which say for 0.1 ns is 100% of applied EMF in crosstalk in practical terms where R < < L in geometry

The missing load R and wire inductance on each side control the rise time T+L/R for a step load current. So you can define dI/dt yourself, where 10~90% rise time dt=0.35/f(-3dB BW) approx.

I leave the re-search for your homework.

Remember

Wavelength is reduced with dt ( especially in ns range for short wires ) and antenna theory takes over for geometry > 5% wavelength for reflection coefficients, mismatched impedances and resulting effects on EMF or voltage crosstalk.

Tony Stewart EE75
  • 1
  • 3
  • 54
  • 185
  • 1
    You don’t need R on the load side to calculate the open circuit induced emf. – Andy aka Sep 29 '18 at 14:50
  • you dont have current with an open circuit unless using antenna theory and impedance of air – Tony Stewart EE75 Sep 29 '18 at 14:54
  • 2
    Precisely and the op isn’t asking for current. Read the question! – Andy aka Sep 29 '18 at 14:56
  • but that is ambiguous as there are no specs for current BW , frequency or impedance E Field coupling or EMF induced B fields matter. here But if you have the exact answer pls tell us. – Tony Stewart EE75 Sep 29 '18 at 14:59
  • @Andyaka tell me what is the EMF if current is 50 Hz or 3GHz, 0.1ns rise time? any difference? use any assumptions for L,R etc 0V? Read the question! – Tony Stewart EE75 Sep 29 '18 at 17:21
  • I don't have an exact answer because that does depend somewhat on skin-depth a la frequency but I do have an answer. – Andy aka Sep 30 '18 at 17:16
  • @Andyaka Would it not be 0V with 10A if the resistance is assumed to be 0 ohms as shown and the AC is like 50 or 60 Hz? – Tony Stewart EE75 Sep 30 '18 at 19:17
  • What has resistance got to do with induction? – Andy aka Sep 30 '18 at 20:32
  • Both produce a voltage ( term for EMF) but if the inductance is negligble (10nH @60Hz) and the resistance is similar < 1 milliohm What is your answer on EMF produced with 10A? ( essentially Zero) catch my drift? @Andyaka this question has no details so it is a trick question – Tony Stewart EE75 Sep 30 '18 at 21:04
  • If the wires are so short to give make 10 nH then the volt drop is virtually zero. No it isn’t a trick question; short wires induce very little at low frequencies. – Andy aka Sep 30 '18 at 21:24
  • Ok so you agree with my answer V=0, maybe the silent questioner means Magnetic force between conductors but then it is not EMF but the force from a magnetic field. – Tony Stewart EE75 Oct 01 '18 at 00:14
0

Assuming ZERO Ohm conductors and low-frequency AC current on the right then the voltage on the left, EMF=0 because of Resistance = 0 Ω from Ohms Law. ( an overlooked assumption)

Nobody measures by "EMF" in practice.

The international standards are in Volts.

This is a silly academic learning question.

Tony Stewart EE75
  • 1
  • 3
  • 54
  • 185
  • Assuming an 8 mil 1oz copper PCB trace (resistance < 1 ohm). Assuming the AC current is 50/60 Hz. I am looking for an answer in volts. I'm sorry the question is silly (I'm a little out of my depth) but this answer wasn't helpful at all. – Chris Brown Sep 30 '18 at 16:42
  • Conductor B would be a cable of diameter suitable to carry 10 A RMS – Chris Brown Sep 30 '18 at 17:03
  • yes and it is assumed to be very low resistance for low line loss so if no resistance is given, may we assume this dumb .. erhm I mean trick academic question that it is 0 Ohms @ChrisBrown given the gross lack of details – Tony Stewart EE75 Sep 30 '18 at 19:18