Main Page | Report this Page
 
   
Science Forum Index  »  Physics - Electromagnetic Forum  »  Magnetism and thermodynamics
Page 1 of 1    
Author Message
Rudy von Massow
Posted: Mon Jan 15, 2007 10:43 pm
Guest
I have been baffled for some time now. Could anybody answer my question?
When I move a piece of iron away from a permanent magnet, I have to do work
on the piece of iron. Does this work enter the first law of thermodynamics
dU = DQ + DW and increase the internal energy of the iron, or does it
increase the external energy, the potential energy?
Thanks in advance: Rudy
Jewish Cowboy
Posted: Mon Jan 15, 2007 11:08 pm
Guest
Hi Rudy

This is kind of a funny question. It's definitely first law, but it's a
little hard
to find where the energy goes.

In electrostatics, it's easy. You take two charged plate and double
their
separation. You have now the same field over twice the volume, so
there's
twice the field energy.

The magnetic case is strange because you have a geometry that looks
similar with a small gap between the iron and the magnet. But when you
double the separation, you doulbe the "magnetic resistance" of the
circuit,
and you get HALF the field strength over twice the volume. Since energy
goes as the square of the field strength, you have half the energy in
the
gap. (one quarter the field over twice the volume.)

The change in field energy does actually give you the correct value for
the work you do in moving the piece of iron...but the sign is wrong!

The explanation is not so easy. The electrostatic case is however of
some assistance. The analogy I made earlier of two plates with fixed
charge is wrong. The better analogy is two plates connected to a
battery
of fixed voltage. In this case, the work done to separate the plates is
the same, but you end up with half the charge on the voltage (because
you double the capacitance) and therefore half the energy (a quarter
the field energy over twice the volume) just like the magnetic case.
In this case the work done is in recharging the battery.

I think in the magnetic case the work done must be in dis-aligning the
internal poles of the iron. Paradoxically, it seems that a fully
magnetised
piece of iron must have less internal energy than a demagnetised piece,

Why doesn't a lump of iron then spontaneously magnetise? Because
of the extra cost of EXTERNAL magnetic energy. In fact, it does
spontaneously
magnetise on the grain level, so that there is no external field
energy. The alignment
of these magnetic domains must further decrease the internal energy.

That's the way I see it. I hope it makes sense. Do you remember me from
Pinawa, Rudy? I have a thread going in sci.physics.electromagnetis
about
quantum mechanics (The Case Against the Photon). You should check
it out.

Marty



Rudy von Massow wrote:
Quote:
I have been baffled for some time now. Could anybody answer my question?
When I move a piece of iron away from a permanent magnet, I have to do work
on the piece of iron. Does this work enter the first law of thermodynamics
dU = DQ + DW and increase the internal energy of the iron, or does it
increase the external energy, the potential energy?
Thanks in advance: Rudy
Jewish Cowboy
Posted: Mon Jan 15, 2007 11:08 pm
Guest
Hi Rudy

This is kind of a funny question. It's definitely first law, but it's a
little hard
to find where the energy goes.

In electrostatics, it's easy. You take two charged plate and double
their
separation. You have now the same field over twice the volume, so
there's
twice the field energy.

The magnetic case is strange because you have a geometry that looks
similar with a small gap between the iron and the magnet. But when you
double the separation, you doulbe the "magnetic resistance" of the
circuit,
and you get HALF the field strength over twice the volume. Since energy
goes as the square of the field strength, you have half the energy in
the
gap. (one quarter the field over twice the volume.)

The change in field energy does actually give you the correct value for
the work you do in moving the piece of iron...but the sign is wrong!

The explanation is not so easy. The electrostatic case is however of
some assistance. The analogy I made earlier of two plates with fixed
charge is wrong. The better analogy is two plates connected to a
battery
of fixed voltage. In this case, the work done to separate the plates is
the same, but you end up with half the charge on the voltage (because
you double the capacitance) and therefore half the energy (a quarter
the field energy over twice the volume) just like the magnetic case.
In this case the work done is in recharging the battery.

I think in the magnetic case the work done must be in dis-aligning the
internal poles of the iron. Paradoxically, it seems that a fully
magnetised
piece of iron must have less internal energy than a demagnetised piece,

Why doesn't a lump of iron then spontaneously magnetise? Because
of the extra cost of EXTERNAL magnetic energy. In fact, it does
spontaneously
magnetise on the grain level, so that there is no external field
energy. The alignment
of these magnetic domains must further decrease the internal energy.

That's the way I see it. I hope it makes sense. Do you remember me from
Pinawa, Rudy? I have a thread going in sci.physics.electromagnetis
about
quantum mechanics (The Case Against the Photon). You should check
it out.

Marty



Rudy von Massow wrote:
Quote:
I have been baffled for some time now. Could anybody answer my question?
When I move a piece of iron away from a permanent magnet, I have to do work
on the piece of iron. Does this work enter the first law of thermodynamics
dU = DQ + DW and increase the internal energy of the iron, or does it
increase the external energy, the potential energy?
Thanks in advance: Rudy
Don Kelly
Posted: Tue Jan 16, 2007 1:15 am
Guest
----------------------------
"Jewish Cowboy" <btestware@gmail.com> wrote in message
news:1168916888.930087.102900@11g2000cwr.googlegroups.com...
Quote:
Hi Rudy

This is kind of a funny question. It's definitely first law, but it's a
little hard
to find where the energy goes.

In electrostatics, it's easy. You take two charged plate and double
their
separation. You have now the same field over twice the volume, so
there's
twice the field energy.

The magnetic case is strange because you have a geometry that looks
similar with a small gap between the iron and the magnet. But when you
double the separation, you doulbe the "magnetic resistance" of the
circuit,
and you get HALF the field strength over twice the volume. Since energy
goes as the square of the field strength, you have half the energy in
the
gap. (one quarter the field over twice the volume.)

The change in field energy does actually give you the correct value for
the work you do in moving the piece of iron...but the sign is wrong!

The explanation is not so easy. The electrostatic case is however of
some assistance. The analogy I made earlier of two plates with fixed
charge is wrong. The better analogy is two plates connected to a
battery
of fixed voltage. In this case, the work done to separate the plates is
the same, but you end up with half the charge on the voltage (because
you double the capacitance) and therefore half the energy (a quarter
the field energy over twice the volume) just like the magnetic case.
In this case the work done is in recharging the battery.

I think in the magnetic case the work done must be in dis-aligning the
internal poles of the iron. Paradoxically, it seems that a fully
magnetised
piece of iron must have less internal energy than a demagnetised piece,

Why doesn't a lump of iron then spontaneously magnetise? Because
of the extra cost of EXTERNAL magnetic energy. In fact, it does
spontaneously
magnetise on the grain level, so that there is no external field
energy. The alignment
of these magnetic domains must further decrease the internal energy.

That's the way I see it. I hope it makes sense. Do you remember me from
Pinawa, Rudy? I have a thread going in sci.physics.electromagnetis
about
quantum mechanics (The Case Against the Photon). You should check
it out.

Marty



Rudy von Massow wrote:
I have been baffled for some time now. Could anybody answer my question?
When I move a piece of iron away from a permanent magnet, I have to do
work
on the piece of iron. Does this work enter the first law of
thermodynamics
dU = DQ + DW and increase the internal energy of the iron, or does it
increase the external energy, the potential energy?
Thanks in advance: Rudy

When you move a piece of iron away from the magnet, you are not changing its
magnetic moment (which can be expressed in terms of the effect of an
equivalent current carrying coil) nor are you making any but a small change
in its internal flux. What you are mainly doing is changing the total
external field and the total energy stored in the field so considerations of
the total external field give decent results. Note that doubling the gap
doesn't necessarily halve the field strength in the gap. With a permanent or
an electromagnet the field strength is at a maximum when the gap is 0 and
for a range of gaps near 0, the field strength is essentially constant and
uniform because the gap reluctance is small compared to total reluctance (I
hate the term reluctance as it implies a linearity that doesn't exist). Over
this range force is constant. At larger gaps, the field strength near the
magnet's pole is relatively constant but the field distribution beyond that
gets messier and definitely non-uniform so that one cannot say that the
field stored energy is reduced to 1/4 because of this. More to the point,
the work that you do is dependent on the integral of the force you apply
over the distance and there is a region where the work rises almost linearly
with distance before levelling out. When the iron is in contact with the
magnet, the total energy in the external field is at a minimum. Pull it away
and add (potential) energy to the field. You actually have a conservative
field just as you would with gravity. If you take an object away from earth,
you will find the same effect if you go far enough.
-
Gordon Slemon of U of Toronto handled permanent magnets quite nicely using
air gap lines and the magnetisation curve of the magnet (valid only for
uniform gaps where fringing is negligable). (this was about 20-30 years ago
in a text that he wrote). MIT online curriculum notes has some material-
brief- dealing with this.

The sign problem in the change of energy with position is dependent on
whether the magnet is doing the work or the work is being done by an
external source. If the magnet is doing the work the stored energy is
decreasing but if you are doing the work the field stored energy is
increasing.

By the way, as you mention Pinawa -do you remember Bob Hollies- (chemical)
who was there for many years.

--

Don Kelly dhky@shawcross.ca
remove the X to answer

>
Jewish Cowboy
Posted: Tue Jan 16, 2007 1:24 pm
Guest
Quote:
When you move a piece of iron away from the magnet, you are not changing its
magnetic moment (which can be expressed in terms of the effect of an
equivalent current carrying coil) nor are you making any but a small change
in its internal flux. What you are mainly doing is changing the total
external field and the total energy stored in the field so considerations of
the total external field give decent results. Note that doubling the gap
doesn't necessarily halve the field strength in the gap. With a permanent or
an electromagnet the field strength is at a maximum when the gap is 0 and
for a range of gaps near 0, the field strength is essentially constant and
uniform because the gap reluctance is small compared to total reluctance

Hi Don

You are analyzing the case where the field in the gap is constant. I
think
this case is deceptive, and I'll tell you why. For me to physically
create
a system where the field stays constant as you move the magnet away,
I have to make the "reluctance" constant around the circuit: which I
can
do by taking a pair of U-shaped magnets and joining one set of poses
together. So I have a big rectangular loop with a gap in it. If I put a
piece
of iron in that gap, I can move it the short distance between the poles
and the field in the gap stays the same, because the reluctance around
the loop is the same. BUT....the iron moves effortlessly from one pole
to another. When there is no change in field energy, there really is no
force.

I think the case I analyzed is more to the point: as you do work by
pulling
the iron away from the magnet, the energy stored in the field changes.
The problem is: the change is in the wrong direction! There is LESS
energy in the field after you do work on it.

Marty
J Thomas
Posted: Tue Jan 16, 2007 8:28 pm
Guest
Jewish Cowboy wrote:

Quote:
I think the case I analyzed is more to the point: as you do work by
pulling the iron away from the magnet, the energy stored in the field
changes. The problem is: the change is in the wrong direction! There
is LESS energy in the field after you do work on it.

I don't follow your reasoning.

Yes, after you pull the two things apart there is less force pulling
them together.

But if you were to put a string on the iron and have it do work when it
gets pulled closer to the magnet, by the time it gets as close as it
was before you pulled it away, you'll have done about as much work as
it took to pull them apart.

So the field has more energy after you pull them apart. it's just set
up so at the beginning you get that energy out slower.
Jewish Cowboy
Posted: Wed Jan 17, 2007 1:06 am
Guest
J Thomas wrote:
..
Quote:

But if you were to put a string on the iron and have it do work when it
gets pulled closer to the magnet, by the time it gets as close as it
was before you pulled it away, you'll have done about as much work as
it took to pull them apart.

So the field has more energy after you pull them apart. it's just set
up so at the beginning you get that energy out slower.

No, that's my point. It's not like a spring or an elastic. And
it's not even like two electically charged bodies. With magnets,
there is actually less field energy when you pull them apart. I've done
the calculation. Simplest case: two parralel currents. Do they
attract or do they repel?

It doesn't seem to make sense but that's how it is. The energy
is somewhere else.

Marty
Don Kelly
Posted: Wed Jan 17, 2007 2:31 am
Guest
"Jewish Cowboy" <btestware@gmail.com> wrote in message
news:1168968264.637946.161330@q2g2000cwa.googlegroups.com...
Quote:
When you move a piece of iron away from the magnet, you are not changing
its
magnetic moment (which can be expressed in terms of the effect of an
equivalent current carrying coil) nor are you making any but a small
change
in its internal flux. What you are mainly doing is changing the total
external field and the total energy stored in the field so considerations
of
the total external field give decent results. Note that doubling the gap
doesn't necessarily halve the field strength in the gap. With a permanent
or
an electromagnet the field strength is at a maximum when the gap is 0 and
for a range of gaps near 0, the field strength is essentially constant
and
uniform because the gap reluctance is small compared to total reluctance

Hi Don

You are analyzing the case where the field in the gap is constant. I
think
this case is deceptive, and I'll tell you why. For me to physically
create
a system where the field stays constant as you move the magnet away,
I have to make the "reluctance" constant around the circuit: which I
can
do by taking a pair of U-shaped magnets and joining one set of poses
together. So I have a big rectangular loop with a gap in it. If I put a
piece
of iron in that gap, I can move it the short distance between the poles
and the field in the gap stays the same, because the reluctance around
the loop is the same. BUT....the iron moves effortlessly from one pole
to another. When there is no change in field energy, there really is no
force.
--------------

I am not assuming that the field in the gap is constant. What I am saying is
that, for a permanent magnet (or electromagnet) with a iron armature there
will be a maximum flux density when the gap is 0. There will also be 0
stored energy in the gap. This limit is determined by the material and
excitation in the electromagnet (which is actually easier to analyse) or by
the magnet material and its magnetic moment in the permanent magnet. Now
consider that the gap is changed from 0 to, say 0.5 mm - the gap flux
doesn't change much over this range as the air gap mmf needed is small. If
the gap increases to 1mm, the flux doesn't drop to half of what it was at
0.5mm but may be only slightly lower. Now, as the gap increases more, you
will quickly approach the case where the flux density drops by half for
every doubling of distance because the air mmf becomes dominant. (ignoring
the effect of fringing and the drastic change in the field geometry that
would occur). In addition, at a small gap, the total field volume is smaller
than at a large gap.
Beyond the initial part of the movement, the force will vary inversely with
distance but the air gap stored energy is the integral of the incremental
force times distance. At 0 gap, it is 0. In the first part of the movement
where the flux is nearly constant, the energy increases almost linearly but
at larger gaps the energy increase tails off so total energy approaches some
limit. However, for all gaps x, the external energy will be larger at x+Dx
What is the energy within the magnet doing during all this? -That is
interesting as well. The flux doesn't change much and the volume is fixed
so? In fact it doesn't actually matter for purposes of calculating forces.
------------
I recall analysing magnetic structure-specifically looking at this
actor -basically a lifting magnet with an iron pole piece Using fmagnetic
=dW'/dx with loop mmf constant will give the typical force dependent on gap
squared where the iron mmf is 0. This implies that the force goes to
infinity at 0 gap. Looking at it from fmagnetic=-dW/dx with flux constant.
(W is the field energy and W' is the coenergy and the derivatives are
partials) it is necessary to calculate the flux at each position but the
result is the same ---except for small gaps--- where the flux comes to a
limit. The force is constant or close enough for government work, over a
short range where the gap reluctance is small compared to the magnet's
reluctance.
---------
Quote:
I think the case I analyzed is more to the point: as you do work by
pulling
the iron away from the magnet, the energy stored in the field changes.
The problem is: the change is in the wrong direction! There is LESS
energy in the field after you do work on it.
-----------

Note that the magnetic force is of the opposite sign to the rate of change
in the field stored energy with distance-opposite to the external force - so
an external force, opening the gap (x increasing), acts to increase stored
energy. I don't see a sign problem.

0 energy or ground state is when the gap is 0 (0 volume) Increasing the gap
increases the volume of the air gap field and even if the flux drops, there
is now energy stored in the gap. increasing the gap more results in less
force but more energy.The rate of change of energy with distance decreases
but is still positive. Your intuition that you are putting work into the
field so "what the hell goes with the sign?" certainly is right
--

Don Kelly dhky@shawcross.ca
remove the X to answer
----------------------------
J Thomas
Posted: Wed Jan 17, 2007 9:19 am
Guest
Jewish Cowboy wrote:
Quote:
J Thomas wrote:
.

But if you were to put a string on the iron and have it do work when it
gets pulled closer to the magnet, by the time it gets as close as it
was before you pulled it away, you'll have done about as much work as
it took to pull them apart.

So the field has more energy after you pull them apart. it's just set
up so at the beginning you get that energy out slower.

No, that's my point. It's not like a spring or an elastic. And
it's not even like two electically charged bodies. With magnets,
there is actually less field energy when you pull them apart. I've done
the calculation. Simplest case: two parralel currents. Do they
attract or do they repel?

It doesn't seem to make sense but that's how it is. The energy
is somewhere else.

Somehow I'm missing your point.

Start with two magnets that are 1 inch apart. You can get some work
done by harnessing the power they provide when they pull closer
together. Now start with the magnets 2 inches apart and harness the
power they release whenk they pull closer together first the one inch
to get to where you started last time, and then the second inch.

Which way do you get more work done? I think you get more work the
second way, where you get everything you got the first time, plus some
extra.

So why do you say the field has less power the second way?

I'm not saying you'r wrong, but if you're right I've misunderstood what
you're talking about.
Jewish Cowboy
Posted: Wed Jan 17, 2007 12:40 pm
Guest
Don, I think I can get to the gist of your argument by saying that
when you start with ZERO gap, and pull the iron away from the
magnet, you are creating a field where no field existed before, so
obviously the work done has gone into creating field energy.

I think the fallacy with your argument is that you are neglecting
the fact that there already WAS field energy...behind the iron, on
the back face. That's the point I was trying to illustrate with my
back-to-back U-shaped magnets. It's not obvious to me that
the TOTAL field energy (gap, external field, and internal field
in the iron) either decreases or increases in your example.

That's why I chose an example where I think it's clear that the
field energy decreases. Because even if your case DOES show
an increase in field energy (which I'm not prepared to accept based
on your arguments so far) I think you still have to explain the case
where the field energy clearly decreases.

That's why I cited the case of the two wires with parallel currents.
They attract each other and pull together...the field energy
obviously increases in this case, which is contrary to what you
would expect from your energy arguments.

marty
Jewish Cowboy
Posted: Wed Jan 17, 2007 12:52 pm
Guest
Quote:

Start with two magnets that are 1 inch apart. You can get some work
done by harnessing the power they provide when they pull closer
together. Now start with the magnets 2 inches apart and harness the
power they release whenk they pull closer together first the one inch
to get to where you started last time, and then the second inch.

Which way do you get more work done? I think you get more work the
second way, where you get everything you got the first time, plus some
extra.

So why do you say the field has less power the second way?

I'm not saying you'r wrong, but if you're right I've misunderstood what
you're talking about.

There is a calculation which no one has actually done explicitly so
far.
It's a specific energy calculation where you take the square of the
field
density and integrate it over all space. This is the energy we're
trying to
analyze.

My point is that as you pull the magnets farther apart, the value of
this
integral actually decreases, rather than increasing as you might think
it should.

We started off this discussion analyzing the case of a magnet and
a piece of iron. You have cited the case of two permanent magnets.
I'm not sure this is exactly the same case.

In fact, I'm not totally sure I was right even for the case of the
magnet
and the piece of iron.

But I'm pretty sure I'm right about the two parallel wires each carring
a constant current. As you pull them apart, the integral of the field
energy clearly decreases. So where does the work go?

If you make these wires into loops, it looks an awful lot like the case
of
two magnetic pole faces in proximity to each other.

Marty
Autymn D. C.
Posted: Wed Jan 17, 2007 2:53 pm
Guest
Take the magnitude. The sign is for the [vectoral] direction.
Don Kelly
Posted: Fri Jan 19, 2007 1:23 am
Guest
"Jewish Cowboy" <btestware@gmail.com> wrote in message
news:1169052035.310392.251370@q2g2000cwa.googlegroups.com...
Quote:
Don, I think I can get to the gist of your argument by saying that
when you start with ZERO gap, and pull the iron away from the
magnet, you are creating a field where no field existed before, so
obviously the work done has gone into creating field energy.

I think the fallacy with your argument is that you are neglecting
the fact that there already WAS field energy...behind the iron, on
the back face. That's the point I was trying to illustrate with my
back-to-back U-shaped magnets. It's not obvious to me that
the TOTAL field energy (gap, external field, and internal field
in the iron) either decreases or increases in your example.

That's why I chose an example where I think it's clear that the
field energy decreases. Because even if your case DOES show
an increase in field energy (which I'm not prepared to accept based
on your arguments so far) I think you still have to explain the case
where the field energy clearly decreases.

That's why I cited the case of the two wires with parallel currents.
They attract each other and pull together...the field energy
obviously increases in this case, which is contrary to what you
would expect from your energy arguments.


marty
------

Actually, I have been giving this some thought and did some initial
calculations (no guarantees attached at present) for a permanent magnet with
good electrical steel pole pieces to get effectively a C structure where
the back of the C is the magnet, the arms are steel with mu approaching
infinity and the opening being a gap. All flux is assumed to be in the gap
with fringing ignored. I also for simplicity assumed gap and magnet areas
are the same. The magnet is in saturation so then we have u(magnet) =muo
With these conditions, the result is that the stored energy in the gap
(which is what we are really concerned with with regard to force and with
regard to what we can actually observe) rises to a peak (in this case when
the gap and magnet length are equal) and then falls off (which is not what
I expected) - and the internal stored energy in the magnet also does the
same- I will check this but I do note that at B or H =0 the magnet's stored
energy is 0 and the maximum energy point is somewhere in between -in fact a
situation close to that at maximum power transfer in a resistive circuit
appears to exist. ( I can show you this-if you want- once I have
proof-read what I have done - I tend to back of envelope scribbling - it
would be easier with non-ascii math and pictures).

However the model fails at large gaps because there will be leakage around
the magnet (as well as fringing in the intended gap. At a gap length equal
to the magnet length, the total magnet flux will be shared nearly equally
between the gap and the fringing so the simplifying assumptions that I made
are then ratshit. Taking this into account, there will be a point where an
increase in gap has a negligable effect on the stored energy as the
effective gap becomes little more than the magnet length and the total
stored energy becomes constant.
In addition, there is a limit involved where irreversible demagnetisation
occurs and this is taken into consideration when designing a permanent
magnet. When the magnet is standing alone, the pole to pole gap must be such
that the flux density (and corresponding H) is above the knee of the
demagnetisation curve (see the second reference).

As for the two parallel wires. the same energy balance approach that I
mentioned earlier works for these as well. Note that now you have an
external current so that the energy balance involves the change in
electrical energy as well as the change in field and mechanical energy. I
agree that where the currents are in the same direction, the stored energy
in the field increases as the wires come together (as seen from the mutual
inductance) but the co-energy W' decreases.
You have a fixed mmf driving the system so that taking f(magnetic) =dW'/dx
with constant mmf (pain in the ass to try to solve using flux and ) leads
to a magnetic force pulling the wires together.

There are two factors of concern in trying to compare this situation to a
permanent magnet.

a)Consider current, in fact the same current, to simplify, in opposite
directions in the conductors. Now the magnetic force is one of repulsion.
You now have a solenoidal coil and the magnetisation of a permanent magnet
can be represented by a solenoid. This doesn't actually address your concern

b) That doesn't matter a hoot as the coil real or apparent is fixed and the
forces that exist are forces acting on the currents -not the magnetic
medium. The force of concern in any magnet are between parts of the magnetic
path. If one builds a solenoid, there are forces acting on the coils, trying
to expand them or shift them but these forces are, while a nuisance, not the
forces which are of importance- which are those due to changes in the
magnetic "medium"-e.g. the air gap of a lifting solenoid- where the coils
are fixed in space. The parallel conductor model deals with the case of
forces between conductors but not forces due to changes in the geometry of
the magnetic path which is the situation with a typical magnet.
The same energy balance approach can be used in either case (or a mixed
situation) but eventually, except in well defined situations, analysis gets
bloody messy.


References that I have recently found:

http://ocw.mit.edu/OcwWeb/Electrical-Engineering-and-Computer-Science/index.htm
look up course 6.685 Electromechanical Machines Fall 2006 (?) Chapter 6,
part 2

Actually I like the following better:
www.magnetweb.com which deals with both theory and practice.

--

Don Kelly
dhky@shawcross.ca
remove the X to answer
----------------------------
 
Page 1 of 1       All times are GMT - 5 Hours
The time now is Fri Dec 05, 2008 1:27 am