Are there torus shaped permanent magnets ?

[jaymo...@hotmail.com]

Ive tried making a torus shaped (circular or donut shaped)
magnetic field by arranging a circle of bar magnets.
Unfortunately all I get is a series of north south poles
In a circle.
My question is.. are torus shaped permanent magnets
possible to make and buy. Ive noticed horseshoe shaped
Fields are available so thats half way there.
Thanks.

[benj]

Two horseshoe shaped magnets stuck together does it. Or you could
probably find two semi-circular half torus magnets. And one could easily
make such a magnet by starting with a torus of magnet material and then
winding a coil on it and pulsing it to charge the magnet.

The question would be, however, why you'd want to do this as the field is
for the most part enclosed inside the magnet. Unless the torus has a slot
or split somewhere you really couldn't access that internal field.

[Salmon Egg]

In article <d46495b6-4e4a-4781...@googlegroups.com>,

Of course. There are zillions of toroidal cores and transformers out in
the world. Most of such cores are made of soft magnetic material for
transformers rather than permanent magnets. Most permanent magnets using
toroidal cores are magnetized normally to the circular plane of the
core. These find wide use in magnetrons, loud speakers, and holding
magnets.

If you want to have a toroidal field, most of the applications have used
air cores. Your setup with nose to tail bar magnes would do the job if
it were not for the air gap. That could be filled in with closely fitted
pieces of soft magnetic material.

If you really wanted flux moving through a core with little outside, it
is possible to take one of the hard speaker magnets and reorient the
magnetic domains with a high enough pulse of current through the hole.

--

Sam

Conservatives are against Darwinism but for natural selection.
Liberals are for Darwinism but totally against any selection.

[jaymo...@hotmail.com]

Thanks for both of your replies..What I wanted
to do was see what direction a compass needle
points when brought up to the outside of
a circular/toroidal magnetic field from a
permanent magnet. That is placed tangentally
to the circumference of the torus. What direction
is the field? Ie. is it north clockwise or south. I cant imagine
what direction it would be.Because when one brings
a compass needle to the curved top end of a horseshoe
magnet I assume the field will have a north south
tangental direction.Yet if one could close off the two horse
Shoe ends into a toroid presumably the compass needle
wouldnt show any specific direction any more, otherwise some
basic laws of physics would be violated.
(By horseshoe Im referring to the ones that have the open ends
as one north and one south, like a bent bar magnet)

[Salmon Egg]

In article <42a29cc0-d20a-4abb...@googlegroups.com>,

My guess is that it would be very difficult to make a toroidal magnet so
uniform that no flux will be exterior to the magnet.

The only application I know of was an attempt to see if the vector
potential inside such a magnet could be seen outside the magnet itself.
The attempt would be for a non-quantum test of the Aharonov�Bohm effect.

[benj]

On Wed, 22 May 2013 19:47:14 -0700, Salmon Egg wrote:

> In article <42a29cc0-d20a-4abb...@googlegroups.com>,
> jaymo...@hotmail.com wrote:
>
>> Thanks for both of your replies..What I wanted to do was see what
>> direction a compass needle points when brought up to the outside of a
>> circular/toroidal magnetic field from a permanent magnet. That is
>> placed tangentally to the circumference of the torus. What direction
>> is the field? Ie. is it north clockwise or south. I cant imagine what
>> direction it would be.Because when one brings a compass needle to the
>> curved top end of a horseshoe magnet I assume the field will have a
>> north south tangental direction.Yet if one could close off the two
>> horse Shoe ends into a toroid presumably the compass needle
>> wouldnt show any specific direction any more, otherwise some
>> basic laws of physics would be violated.
>> (By horseshoe Im referring to the ones that have the open ends as one
>> north and one south, like a bent bar magnet)
>
> My guess is that it would be very difficult to make a toroidal magnet so
> uniform that no flux will be exterior to the magnet.

This would be my guess too. Basically the compass experiment would really
show only the field leakage around the magnet and not anything
fundamental.

> The only application I know of was an attempt to see if the vector
> potential inside such a magnet could be seen outside the magnet itself.
> The attempt would be for a non-quantum test of the Aharonov­Bohm effect.

That would be about it.

[jaymo...@hotmail.com]

I suppose that getting a perfect toroid field
like the one Im trying for would actually
be impossible now youve described how difficult
it would be technically to achieve.
Because a perfectly circular field
would have no dipole arrangement. That is,
there would be no north or south pole in
a circular field. My mistake for even thinking
it was possible.

[Timo Nieminen]

On Thursday, 23 May 2013 22:31:44 UTC+10, jaymo...@hotmail.com wrote:
>
> Because a perfectly circular field
> would have no dipole arrangement. That is,
> there would be no north or south pole in
> a circular field. My mistake for even thinking
> it was possible.

Why would that be needed?

As Sam noted above, it's done routinely with electromagnets (e.g., transformer cores). The practical difficulty of making one perfect enough for some particular application, and the lack of common useful applications resulting in little effort being put in to overcome such practical difficulties doesn't mean that it isn't possible.

Yes, there would be no north or south pole in a circular field. So?

How good a toroidal magnet do you need? Might an electromagnet be good enough? Are you cancelling the Earth's field? How completely? That affects how good "good enough" is. Why do you think you would have a magnetic field outside such a toroidal permanent magnet?

[Salmon Egg]

In article <1707c96a-8633-44a6...@googlegroups.com>,

IT IS POSSIBLE in principle,but what is the point? A friend of mine did
do something of that nature. He was thinking of a classical experiment
to see if a magnetic vector potential could be detected where B=0.

He also showed that you could have a significant magnetic retention
field in a soft magnetic material if there were not air gaps in the
magnetic circuit.

[jaymo...@hotmail.com]

On Thursday, 23 May 2013 23:01:40 UTC+1, Salmon Egg wrote:
> it was possible. IT IS POSSIBLE in principle,but what is the point?

Just really interested to see if a perfectly contained circular field
would make a compass needle point tangentally to the field.
Kind of like the way a compass needle points tangentally to the field
around the circular field around a conducting wire.
> A friend of mine did do something of that nature. He was thinking of a classical experiment to see if a magnetic vector potential could be detected where B=0. Interesting. Did he detect any magnetic vector potential?
At a guess Im would say he didnt , provided the there werent any gaps
in the field. But that makes me wonder. How do you you know if there
is a field in a permanent magnet with supposed a circular field, if
you cant detect a field?

> He also showed that you could have a significant magnetic retention field in a soft magnetic material if there were not air gaps in the magnetic circuit.

When you say no air gaps was he making the circular field out of seperate
sections of permanent magnet with soft magnetic material inbetween?
The reason I ask is...
I was looking at a magnet supply website and they sold no magnets with
circular fields. Nor did they supply sections like 45 degree arcs that
had the field in the direction of the section. Rather the field was
through the arc piece .North on the outside (convex face)
south on the inside of the arc.

[jaymo...@hotmail.com]

On Thursday, 23 May 2013 22:36:16 UTC+1, Timo Nieminen wrote:
>> there would be no north or south pole in
>> a circular field. My mistake for even thinking
>> it was possible.
>> Why would that be needed? As Sam noted above, it's done routinely with electromagnets (e.g., transformer cores). The practical difficulty of making one perfect enough for some particular application, and the lack of common useful applications resulting in little effort being put in to overcome such practical difficulties doesn't mean that it isn't possible. Yes, there would be no north or south pole in a circular field. So?

Every *permanent* magnet field description always has a north pole and
south pole included inside the magnetic material. THis suggests to
me that if the material of the magnet doesnt have these two poles
within its confines then its not a magnet. And If I imagine a perfectly circular field,
Where are the poles?
I know a electromagnet can have a circular field without any `poles`
but thats not the same as a permanenet magnet.

> How good a toroidal magnet do you need? Might an electromagnet be good enough?

Im interested in particular at the moment about permanent magnets

> Are you cancelling the Earth's field? How completely?

I havent really thought about this

> That affects how good "good enough" is. Why do you think you would have a magnetic field outside such a toroidal permanent magnet?

THats why I was trying to get hold of a circular peremant magnet.
To find out if theres a field outside. From all the responses ,
my guess is not. But asd I said to Salmon,
..if one can make a permanent magnet with a perfectly circular field
It sounds a bit like schrodingers cat. How do you know theres
a field if you cant measure anything?
Another seperate point that seems interesting to me is.
From what Ive seen its impossible to create a replica of field
lines of a bar magnet with an electromagnet. Take the illustration
of iron filings around a bar magnet field on wiki page `magnetic field`
And then compare its lines to the lines around the solenoid in
the illustration of the electromagnetic field at the top
of the same page. Notice how electromagnetic field lines always
only point into the pole from the top of the electromagnet.
Whereas in the permanent magnetic example field lines point
to each pole from the top of the bar and the sides of the top.
Completely different from the electromagnetic field lines.
So a peremant magnet always has its poles inside the material.
Wheras a electromagnet can only have its poles outside the wire
in the air space. And the electromagnet cannot have field
lines pointing to the north pole from all directions. Only
from the top. Odd

[jaymo...@hotmail.com]

Sean said in previous post..."from what Ive seen its impossible
to create a replica of field lines from a bar magnet with
an electromagnetic field"
I just thought of one exception. Position a wire from
the neg terminal of a battery vertically near to a vertical
wire from the pos terminal. Both either with ends
dipping in an oil filled petrie dish with filings,or both
peircing a sheet of paper with iron filings. What I think
should be observed is the filings arranging themselves
into two nodal points about each end of each wire. With
lines radiating away from each node except for those
filings inbetween each node(pole). Here they would
bend in arcs towards each node. Identical to the pattern
seen in the iron filing image of a bar magnets field part
way down the page fron wiki titled magnetic field.

[Don Kelly]

An electromagnet (DC) that is a closed torus isn't of much use- all that
you would get would be the leakage flux. For AC toroidal coils on a
ferromagnetic core are useful for inductors or transformers.
As for the comparison of electromagnets and bar magnets- The Wiki page
that you mention gives the same field distribution for both (top 2
pictures)- with field lines in the same direction for both cases. If
both were ideal, all flux would be from the ends. Neither are ideal so
in both there is leakage flux along the sides such as shown in the iron
filing picture -which really doesn't indicate strength of the field.
However, an "iron core" solenoid will concentrate more flux at the
ends-because iron is a far better "magnetic conductor" than air and
there is far less leakage.
--
Don Kelly
remove the cross to reply

[Salmon Egg]

In article <DRvot.53504$cD1....@newsfe29.iad>,

Don Kelly <dh...@shaw.ca> wrote:

> An electromagnet (DC) that is a closed torus isn't of much use- all that
> you would get would be the leakage flux. For AC toroidal coils on a
> ferromagnetic core are useful for inductors or transformers.
> As for the comparison of electromagnets and bar magnets- The Wiki page
> that you mention gives the same field distribution for both (top 2
> pictures)- with field lines in the same direction for both cases. If
> both were ideal, all flux would be from the ends. Neither are ideal so
> in both there is leakage flux along the sides such as shown in the iron
> filing picture -which really doesn't indicate strength of the field.
> However, an "iron core" solenoid will concentrate more flux at the
> ends-because iron is a far better "magnetic conductor" than air and
> there is far less leakage.

There was a toroidal air-core tank inductor that I came up with. The
application required an rf drive for a laser. The space allocated for it
was too small for a normal tank tank coil. When placed in the allocated
space, the tank circuit detuned from how it was aligned outside.
Moreover, the rf heated the metal surfaces near the coil. More space was
not an option.

The way I solved it was to take a fairly thick piece of glass epoxy
board. A series of holes were drilled throug the board in a pattern that
allowed hand threading of magnet wire through the holes to form a
toroidal coil, This kept almost all the field inside the coil. There was
no significant detuning and heating, but it was difficult to manufacture.

The lab manager was going to help improve the situation for a
non-toroidal coil with his own hands-on attention. When he burned his
finger on the induction heated metal, it suddenly became possible to
find some impossible to find extra space for the coil.

[amdx]

Speaker magnets
Magnets inside a magnetron

Is that to simple? Maybe I don't understand.

See;
http://tinyurl.com/ksn67um




Mikek

[benj]

The only problem is speaker magnets are magnetized the "wrong" direction.
To get what you want you'd have to take one and wrap a manetizing coil
around it like a standard toroid inductor and charge it that way. In such
a case the field would be circular inside it.

[Jos Bergervoet]

On 8/4/2013 6:04 PM, benj wrote:
> On Sun, 04 Aug 2013 10:30:56 -0500, amdx wrote:
>> On 5/22/2013 10:53 AM, jaymo...@hotmail.com wrote:

>>> My question is.. are torus shaped permanent magnets possible to make

>> ..
>> http://tinyurl.com/ksn67um

>
> The only problem is speaker magnets are magnetized the "wrong" direction.
> To get what you want you'd have to take one and wrap a manetizing coil
> around it like a standard toroid inductor and charge it that way. In such
> a case the field would be circular inside it.

And outside it there would be no field at all.
But still the electrons in a current through its
hole would get a phase shift, only that would be
difficult to see since you can't easily switch
off a pemanent magnet. And also because the phase
of electrons won't be preserved for more then
their free path length.

So it would not be easy to find out if you really
bought such a magnet, or just some ring of dark
colored material that is not magnetic at all.
(Just as a warning. Caveat emptor!)

--
Jos

[Don Kelly]

Such cores have been made - magnetic memory cores.

http://en.wikipedia.org/wiki/Magnetic-core_memory

This is about the only use for such toroidal cores that you appear to
suggest. This type of computer memory is now in museums.

The magnetic material used is essentially permanent magnet material
with a "square" hysteresis loop -that is it is magnetized in one
direction or the other. This is related to some extent to what Jos is
saying- but sensing a flip from one direction to the other is easy.

Failing such an operation- a permanent magnet toroid with the magnetic
field as you suggest is rather useless (Jos is right on).
What is needed is an "air gap" such that you can access the field. A
horseshoe magnet is simply a distorted torus with an air gap- the
actual shape of the magnet is a non- issue. I have seen a large magnet
with a small gap. This had a fat iron core with small permanent magnets
on each side of a gap. Cheaper than using a whole core of PM material

The magnets Benj addresses have a radial field or a field like a disc
magnet with a hole in the middle. Here the external field is what is wanted.

[Salmon Egg]

In article <51fe86da$0$15922$e4fe...@news2.news.xs4all.nl>,

This indeed is interesting quantum physics! Although there would be no
field outside the torus, in principle, there will a magnetic vector
potential. The quantum interference introduced by the curl-less vector
potential demonstrates its existence. I know of no classical way to
demonstrate the existence of a curl-less magnetic vector potential.

It may be possible do demonstrate the existence of a field inside a
toroidal permanent magnet using optical methods. It would require some
transparency. Yttrium iron garnet, YIG might do it.

A toroid could probably be magnetized by discharging a large capacitor
through a wire threaded through the toroid hole. It should probably be
critically damped so as not to demagnetize the magnet after creating it
with consequent current oscillation.

[Don Kelly]

Addendum- the core material for core memory may be similar to permanent
magnet material in having a "square" hysteresis loop but in general it
had a lower coercive force -so that extremely high currents weren't
needed to reverse the magnetic polarity.

[benj]

On Mon, 05 Aug 2013 16:52:41 -0700, Salmon Egg wrote:

> A toroid could probably be magnetized by discharging a large capacitor
> through a wire threaded through the toroid hole. It should probably be
> critically damped so as not to demagnetize the magnet after creating it
> with consequent current oscillation.

"could probably"? Nice attempt to invent the wheel! See article below for
standard magnetizer.

http://www.laboratorio.elettrofisico.com/pdf/Magnetizzazione/
mag_01_pulse_magnetizers_and_coils.pdf

Note the cute circuit that functions as BOTH a magnetizer (with a single
pulse) and demagnetizer with a ringing decaying waveform. The magnetizer
waveform is not critically damped but is rather has the reverse trimmed
off with a damping diode. These devices are common and usually have a big
soft copper rod that you arrange to do the magnetizing like turns on a
toroid. You don't critically damp because the goal is to get as much
current in the pulse as possible and critical damping creates too much
loss.

[Jos Bergervoet]

On 8/6/2013 1:36 AM, Don Kelly wrote:
> On 04/08/2013 8:30 AM, amdx wrote:

...
>> ...
>> http://tinyurl.com/ksn67um

>
> Such cores have been made - magnetic memory cores.
>
> http://en.wikipedia.org/wiki/Magnetic-core_memory
>
> This is about the only use for such toroidal cores that you appear to
> suggest. This type of computer memory is now in museums.
>
> The magnetic material used is essentially permanent magnet material with
> a "square" hysteresis loop -that is it is magnetized in one direction or
> the other. This is related to some extent to what Jos is saying- but
> sensing a flip from one direction to the other is easy.

Yes, but producing the flip with a current pulse
trough another wire leads to the extra difficulty
of minimizing the coupling between the flipping wire
and the sense wire. Obviously this is why they are
at right angles in memory cores, but there will
always be some coupling, so you will always see
some signal in the sense wire. Even with a non-
magnetized ferrite ring you just have a 1:1
transformer, after all. So you need to calculate
carefully in advance what is the signal shape that
would indicate a flipping field, as opposed to just
the signal transfer through a transformer!

There is another way, without an extra current
pulse: heating up the magnet beyound the Curie-
temperature will reduce its magnetization to zero.
The sense wire is then the only wire needed and
a current flow will indicate the transistion of
the field.

Both methods (flipping the field and removing the
field) will leave you with a magnet that no longer
meets the original specification, I'm afraid..
But if you buy a whole bunch you could just test
one of them.

--
Jos

[Don Kelly]

The thermal approach demagnetizes the core. The flip method reverses the
direction of the flux in the core -both a 1 and a 0 are fully magnetized
states. It is true that reading a 1 will result in a 0 so it is
necessary to follow reading of a one by a restore signal to flip back to
a 1.
In that way, there is no net state change.
Read the Wiki article.

[benj]

On Tue, 06 Aug 2013 12:28:22 -0700, Don Kelly wrote:

> The thermal approach demagnetizes the core. The flip method reverses the
> direction of the flux in the core -both a 1 and a 0 are fully magnetized
> states. It is true that reading a 1 will result in a 0 so it is
> necessary to follow reading of a one by a restore signal to flip back to
> a 1.
> In that way, there is no net state change.
> Read the Wiki article.

So what is wrong with "core" memory, and how could you "fix" that today
to make it more viable?

[Don Kelly]

Probably not possible. 3 reasons, cost,size, speed.

How small can one make a magnetic core with 3 wires through it?
The reason we don't see them now is the same reason that we do not now
store information on clay tablets (baked after writing so not
re-writable)-technology aims at faster, smaller and cheaper (unlike
modern Olympics, faster, bigger and more expensive).
We still use magnetic storage- e.g. hard drives but now we are dealing
with small domains of magnetic material which can be read without
destruction, or written over. Such a hard drive is far superior to core
memory. We also have such things as flash drives which pack a lot of
memory into a small package at a reasonable price.

Neanderthals painted pertinent information on cave walls. Excellent
retention of information. Egyptians, Mayas and others preferred to
carve or paint rocks- also good. Persians wrote on clay tablets and
baked them- somewhat more fragile but portable.

I expect that, at some time in the future, the technology of today will
be considered as rather primitive- or, depending on the future, writing
on rocks will again be cutting edge technology carried out by cockroaches.

[benj]

========

I'm not worried about computer archeology, but you are thinking along the
lines I was.

Core memory was from the early days of slow transistors. That has
obviously changed in the interim. Second, cores in those days were really
ferrite doughnuts. Huge by say the standards of domains packed into a
typical hard disk storing gigs on a tiny platter.

I totally disagree that hard disks are superior. First off is Lenz's law:
If it rotates it sucks! (That would be Dave Lenz a guy I went to school
with :) Disks are mechanical and that makes them DEFECTIVE right out of
the box. Secondly they are basically 2D (except for the stack of
platters). And thirdly there is an access time problem both from rotation
speed and track to track access.

Now imagine if you will, some kind of ferromagnetic photoformed
Integrated chip with a huge array of domains accessed by high speed
drivers and sensing. Now imagine those ICs stacked into a cube. Some
serious memory density starts to look possible. And access speeds are
electronic.

And having tried to use flash drives for long term data storage I can
assure you they are FAR less than perfect. Failure is something that is
always on the horizon. But they DO show the advantages over mechanical
devices. Oddly, in trying to store large amounts of data over time, I've
have best luck with floppy disks (though capacity is now too low). Hard
drives tend to have precision bearing failure over time especially if not
run periodically and obviously wear out if they run constantly. Optical
record-able disks tend to fail when stored (I just lost a big batch of
those. Commercial disks (which are made differently) were OK but the
recorded onces developed lots of errors. This is a kind of hint that
magnetic data storage may not be so "old hat".

So how one stores long-term High-density data is a problem. Cuneiform
tablets DO have proven reliability. Personally, I have some high hopes
for the Nitride (stone) Blueray record-ables, but only time will tell.
But they are still mechanical.

Currently it seems that the only really SAFE way to back up data is with
multiple redundant servers with a program of regular hard disk
replacement. Not quite my idea of a cuneiform tablet.

So I suggest that a modernized miniaturized "core" memory may not be as
Neanderthal as you might think.

[extremesou...@yahoo.com]

On Wednesday, May 22, 2013 8:53:38 AM UTC-7, jaymo...@hotmail.com wrote:
> Ive tried making a torus shaped (circular or donut shaped)
>
> magnetic field by arranging a circle of bar magnets.
>
> Unfortunately all I get is a series of north south poles
>
> In a circle.
>
> My question is.. are torus shaped permanent magnets
>
> possible to make and buy. Ive noticed horseshoe shaped
>
> Fields are available so thats half way there.
>
> Thanks.

unfortunately since a toroid is a manifold the geometry is all wrong for polarization until we find way to create a monopole

[extremesou...@yahoo.com]

On Wednesday, May 22, 2013 8:53:38 AM UTC-7, jaymo...@hotmail.com wrote:
> Ive tried making a torus shaped (circular or donut shaped)
>
> magnetic field by arranging a circle of bar magnets.
>
> Unfortunately all I get is a series of north south poles
>
> In a circle.
>
> My question is.. are torus shaped permanent magnets
>
> possible to make and buy. Ive noticed horseshoe shaped
>
> Fields are available so thats half way there.
>
> Thanks.

since a toroid is a manifold the geometry is all wrong to accept any type of molecular polarization without the field lines and flux canceling themselves out, unless some day we discover how to create a monopole magnet.

[Don Kelly]

---------------
I'm not saying that some form of nanomagnetic storage isn't possible-
but the technology of ferrite cores per se is passe.
Your concept does eliminate mechanical limitations but may be subject to
other limitations
-i.e. if the dimensions of the domains are small enough-then it becomes
difficult
to be sure to not affect adjacent domains when writing or reading?
I note that there is some research in organic solar cells where electron
spin orientation
can have a beneficial effect- Can we get down to this ideal magnetic
orientation and.
if we can- how reliable will it be?
Certainly- whatever we use will have failures -and it may be better
to concentrate on reliability of long term storage rather than
higher densities for such storage.

Much of the problem with hard drives is the push to higher densities and
speeds
higher rotational speeds do mean higher mechanical problems, and higher
density requires
smaller domains and read heads. Some of these problems can be reduced
but then the price
per unit of memory stored will cause users (other than governments) to
scream. I recall buying an early hard drive for an early IBM PC (a step
up from using 2 5+inch floppies) and I vaguely recall something of the
order of $100/MB.
I have been lucky, not having had hard disc failures-so far- but
definitely have had floppies that failed. My backup HD is one that
survived a power supply and CPU failure. It is, however, a small HD
running at a slower speed than newer ones. My present main HD is only
500GB ( I don't need more as I clean up my unused garbage every now and
then -a habit formed when memory was expensive or non-existent (main
frame, punched cards in, paper out, so that the original IBM PC was a
blessing with a floppy drive (and optional math co-processor or second
floppy). I am an ancient artifact -first program about 1961- written in
"MAD"(real programmers used assembly language).

[benj]

I"m thinking that just because "core" is passe, that there may be
something there to re-think. Personally I'm astounded at the magnetic
density that is being achieved on hard disks these days. If only they
weren't mechanical.

I've tried saving data lots of ways and most of them were less than
optimum. I recall an adventure buying brand new hard disks and filling
them and shoving them on a shelf. What could possibly go wrong, right?
Both I and a friend who did the same thing discovered that unused drives
allowed to sit for years just have the bearings seize for some reason and
they won't spin. You really have to run them periodically to keep them
limber or something. And that's a pain.

I've been using flash drives for a while for back up as well. Small,
cheap, nice capacity and easy to load and read. Unfortunately I've also
got a box full of dead ones. The big hint came when I bought a bunch of
HP thumbs which I figured would be pretty high quality. Only they had
this notice right on the package that they were in no way guaranteed to
store any data and asking them to do that was probably asking too much!

So Right now I've got a server with 9 2 terabyte drives, with some in
RAID configuration, Plus a second server with a terabyte of important
things, plus a drawer of thumb drives as redundancy, Plus I've got stacks
of DVDs that are slowly sinking into the sunset and then I've been
putting important things on blueray Nitride disks for which I have high
hopes. Pretty close to carved on stone!

Long term safe storage of lots of data is not as simple as you think. I
really think that there is a possibility of nanomagnetic core solving
some of this hassle.

Unlike you I don't go back to '61. In those days I was really anti-
computer and while everyone else was gaga and signing up to run the
(tube) Univac, I turned my nose up at the whole idea. Later, my first
efforts were in SCATRAN a private version of Fortran that is now totally
a dead language. So tell me how am I ever going to get that space back in
my brain that I used up with the Scatran manual? Or for that matter how
do I get the space back used up by the DOS manual?

[Don Kelly]

On 09/08/2013 11:07 PM, benj wrote:
> On Fri, 09 Aug 2013 16:21:21 -0700, Don Kelly wrote:
>
>> On 08/08/2013 12:15 AM, benj wrote:

--------snip of past for brevity---------------

>
> So Right now I've got a server with 9 2 terabyte drives, with some in
> RAID configuration, Plus a second server with a terabyte of important
> things, plus a drawer of thumb drives as redundancy, Plus I've got stacks
> of DVDs that are slowly sinking into the sunset and then I've been
> putting important things on blueray Nitride disks for which I have high
> hopes. Pretty close to carved on stone!
>
> Long term safe storage of lots of data is not as simple as you think. I
> really think that there is a possibility of nanomagnetic core solving
> some of this hassle.
>
> Unlike you I don't go back to '61. In those days I was really anti-
> computer and while everyone else was gaga and signing up to run the
> (tube) Univac, I turned my nose up at the whole idea. Later, my first
> efforts were in SCATRAN a private version of Fortran that is now totally
> a dead language. So tell me how am I ever going to get that space back in
> my brain that I used up with the Scatran manual? Or for that matter how
> do I get the space back used up by the DOS manual?
>

I do not have your storage needs and intend to keep it that way.
I have some photos and other data duplicated on a hard drive and flash
drives and DVDs (tax stuff is also stored on paper). I recognize the
need to refresh these storages but my world won't fall apart if I lost
them.

I did not imply that long term storage of lots of data was simple. What
I intended to say was that possibly more effort should be made to
improve long term storage. Nanomagnetic storage is a possibility but
should it be "core" if there are other approaches that can give equal or
better storage densities with equal or better retention-recognizing that
as sizes shrink, the greater chance exists of a random wandering photon
or cosmic particle buggering things up.
Not my area -I am used to Watts, Volts and Amperes to positive powers.

Mad was related to Fortran but more powerful than the Fortrans available
at the time. Of course, at that time, the priests in the air
conditioned computer room fed cards into the machine and later in the
day a printout of results was available -usually with an error message
"This is Mad" along with an Ascii picture of A E Newman.
Do I remember MAD (Michigan Algorithic Decoder) which was better than
the Fortran of its days and introduced concepts later taken up by
Fortran, Fortran (various versions), Pascal and C++ (ugh). Turbo Basic
was good for its time (Fortran on steroids) as it was user friendly.
APL was and is a dream -allowing concentration on the problem rather
than having to make programming decisions that the idiot box could do
for itself but is an interpreter. Presently I am dealing spasmodically
with J which is APL on steroids as an interesting exercise and I am
still learning- spasmodic and short term memory problems don't fit
together very well.
However when one can type +/%# y to get the arithmetic mean of a list of
y values (and can assign this operation to a variable name) or %.X is
the inverse of x (real or complex) and the actual operation involved is
at the machine language level or a (bloated vs assembly) C++ compiled
level. My objective for programming is getting the desired job done-and
I do not want piddly details, even though they are necessary) to get in
the way of this.
Naturally APL and J are not beloved of computer scientists in general.

[benj]

On Sat, 10 Aug 2013 18:28:46 -0700, Don Kelly wrote:

> On 09/08/2013 11:07 PM, benj wrote:
>> On Fri, 09 Aug 2013 16:21:21 -0700, Don Kelly wrote:

> Mad was related to Fortran but more powerful than the Fortrans available
> at the time. Of course, at that time, the priests in the air
> conditioned computer room fed cards into the machine and later in the
> day a printout of results was available -usually with an error message
> "This is Mad" along with an Ascii picture of A E Newman.

> Do I remember MAD (Michigan Algorithic Decoder) which was better than
> the Fortran of its days and introduced concepts later taken up by
> Fortran, Fortran (various versions), Pascal and C++ (ugh).

Yeah, that's the era. Like a laundry. Turn in cards pick up results
later. Apparently at that time all the schools were engaged in proving
how cool they were by "improving" Fortran. Obviously MAD was the Michigan
version of the SCATRAN (forget what it stood for but it was the Ohio
State "improved" fortran). Happily all that incompatible nonsense soon
came to an end.

> APL was and is a dream -allowing concentration on the problem rather
> than having to make programming decisions that the idiot box could do
> for itself but is an interpreter. Presently I am dealing spasmodically
> with J which is APL on steroids as an interesting exercise and I am
> still learning- spasmodic and short term memory problems don't fit
> together very well.

APL is still around? TELL ME MORE! I was always a HUGE fan of APL as an
idea going the right direction (for certain programming tasks), but I
thought it was as dead as Aramaic! I can understand the problems of the
host of oddball special symbols, but I always figured with modern
computers with graphics and all something could be worked out so the huge
power of that approach could be utilized. Is "J" that improvement? I'll
have to look into this! I still have my APL manuals that I kept with the
idea of one day "fixing" APL for modern use. It would be great if someone
else already did that job!

> Naturally APL and J are not beloved of computer scientists in general.

You said it! So much better to have modern bloatware filling beaucoup
gigabytes just to read some file or display it.

[Jos Bergervoet]

On 8/10/2013 8:07 AM, benj wrote:
...

> Unlike you I don't go back to '61. In those days I was really anti-

"If you can remember the 60s, then you weren’t really there."

> computer and while everyone else was gaga and signing up to run the
> (tube) Univac, I turned my nose up at the whole idea. Later, my first
> efforts were in SCATRAN a private version of Fortran that is now totally
> a dead language. So tell me how am I ever going to get that space back in
> my brain that I used up with the Scatran manual? Or for that matter how
> do I get the space back used up by the DOS manual?

But benj, for that you just use INT 13, ah=07.
Didn't you know that?!

--
Jos

[Jos Bergervoet]

On 8/11/2013 3:28 AM, Don Kelly wrote:
> On 09/08/2013 11:07 PM, benj wrote:

>> ...

>
> I do not have your storage needs and intend to keep it that way.
> I have some photos and other data duplicated on a hard drive and flash
> drives and DVDs (tax stuff is also stored on paper). I recognize the
> need to refresh these storages but my world won't fall apart if I lost
> them.

Bringing us back to those old civilizations that carved
the data in stone.. Their world fell apart anyway!

--
Jos

[Alastair Black]

You're not really going to leave the
drive parameter as CURDRV are you?
Gadzooks, man! Governments have toppled
over less...

Alastair

[Don Kelly]

APL is still alive and there are commercial versions as well as free
versions. I have NARS2000-64 (64bit free version for windows- I used to
have the old Dos aplpc version 11 but a 64bit machine doesn't like DOS )
and, as with some equation editors, it does allow selection of symbols
from a table.
comp.language.apl exists still and Sam Stirlin (sp??) gives
updates-although there aren't many.

J was developed by Iverson and uses scripts (ie. workspaces) or direct
input output as with APL. I have version 602 but there are 702 and now
802 for a wide range of platforms including tablets. The learning curve
is steep but there is a great support community and lots of literature
(none of which quite approach Gillman and Rose).
There is a great deal of flexibility in that one can write explicit and
tacit "verbs" (which may be a complex function or a simple naming of an
operation such as sort=: /: { which sorts a list or mean=:+/ % #
(essentially sum % count)

As with APL there are many ways to deal with a problem and there is a
tendency to develop one-liners which are hard to decipher.
Look into it at:

http://www.jsoftware.com/

The price is right- $0.00

Have fun

[Don Kelly]

So radiation resistant cockroaches will reign supreme?

[benj]

Say! How much data can be encoded into cockroaches? I think we may be
onto something here...

[benj]

Yeah, but "born again" is not for me. The key is getting rid of the bad
stuff while keeping the good stuff...you know... just like everything
else in life!

[extremesou...@yahoo.com]

On Sunday, August 4, 2013 8:30:56 AM UTC-7, amdx wrote:
> On 5/22/2013 10:53 AM, jaymo...@hotmail.com wrote:
>
> > Ive tried making a torus shaped (circular or donut shaped)
>
> > magnetic field by arranging a circle of bar magnets.
>
> > Unfortunately all I get is a series of north south poles
>
> > In a circle.
>
> > My question is.. are torus shaped permanent magnets
>

Tokamak Fusion Test Reactor - Wikipedia, the free encyclopedia



en.wikipedia.org/wiki/Tokamak_Fusion_Test_Reactor







The Tokamak Fusion Test Reactor (TFTR) was an experimental tokamak built at Princeton Plasma Physics Laboratory (in Princeton, New Jersey) circa 1980.

Images for tokamak fusion

[extremesou...@yahoo.com]

On Wednesday, May 22, 2013 8:53:38 AM UTC-7, jaymo...@hotmail.com wrote:
> Ive tried making a torus shaped (circular or donut shaped)
>
> magnetic field by arranging a circle of bar magnets.
>
> Unfortunately all I get is a series of north south poles
>
> In a circle.
>
> My question is.. are torus shaped permanent magnets
>

> possible to make and buy. Ive noticed horseshoe shaped
>

A tokamak is a device using a magnetic field to confine a plasma in the shape of a torus

[extremesou...@yahoo.com]

On Wednesday, May 22, 2013 9:09:39 AM UTC-7, benj wrote:

> On Wed, 22 May 2013 08:53:38 -0700, jaymoseley wrote:
>
>
>
> > Ive tried making a torus shaped (circular or donut shaped)
>
> > magnetic field by arranging a circle of bar magnets. Unfortunately all I
>
> > get is a series of north south poles In a circle.
>
> > My question is.. are torus shaped permanent magnets possible to make and

> A tokamak is a device using a magnetic field to confine a plasma in the shape of a torus

> > buy. Ive noticed horseshoe shaped Fields are available so thats half way
>
> > there.
>
> > Thanks.
>
>
>
> Two horseshoe shaped magnets stuck together does it. Or you could
>
> probably find two semi-circular half torus magnets. And one could easily
>
> make such a magnet by starting with a torus of magnet material and then
>
> winding a coil on it and pulsing it to charge the magnet.
>
>
>
> The question would be, however, why you'd want to do this as the field is
>
> for the most part enclosed inside the magnet. Unless the torus has a slot
>
> or split somewhere you really couldn't access that internal field.

[extremesou...@yahoo.com]

On Monday, August 5, 2013 6:03:15 PM UTC-7, Don Kelly wrote:
> On 05/08/2013 4:36 PM, Don Kelly wrote:
>
> > On 04/08/2013 8:30 AM, amdx wrote:
>
> >> On 5/22/2013 10:53 AM, jaymo...@hotmail.com wrote:
>
> >>> Ive tried making a torus shaped (circular or donut shaped)
>
> >>> magnetic field by arranging a circle of bar magnets.
>
> >>> Unfortunately all I get is a series of north south poles
>

A tokamak is a device using a magnetic field to confine a plasma in the shape of a torus

Such a helical field can be generated by adding a toroidal field (traveling around the torus in circles) and a poloidal field (traveling in circles orthogonal to the toroidal field

[jaymo...@hotmail.com]

Don Kelly wrote 06 08 2013...

>Such cores have been made - magnetic memory cores.

http://en.wikipedia.org/wiki/Magnetic-core_memory

>This is about the only use for such toroidal cores that you
>appear to suggest. This type of computer memory is now in museums.

Sorry to revisit an old thread. But this toroid you have cited is sort
of what I was thinking. Because I wondered: if you could get a circular
field around a current carrying wire, would one get a current induced
in a wire that threads through a ' necklace' of toroidal magnets. As in
the toroidal magnets in your cited wiki page. And assuming something like this
is happening in the old RAM memories with the tiny toroidal magnets, how long
will these magnets induce a current? Presumably not long as otherwise
it would be a new source of free energy.
Or is this how the magnets are read in the old RAM memory chips?
That is..loop a conducting wire through the toroidal magnet,...then complete
the circuit. Which induces a current in the wire . That is until the toroid
"loses" or discharges its magnetism?

[Wond]

During a memory read cycle, the cores of the word of interest were
fed a current pulse sufficient to switch them to the opposite state. The
cores that actually switched states produced a relatively large pulse in
their corresponding sense line, and the interpreted results appeared in
the memory register. The (now destroyed) word was then written back to
the cores to complete the cycle.



--
/home/wond/Documents/sig.txt

[Don Kelly]

If you had a toroidal permanent magnet-it would have constant flux
-essentially completely in the magnet -so there would be no changing
flux and no induced voltage to produce a current.
Permanent magnets will not supply any free energy. They work a bit like
pumping water up a hill to store energy- and then running a turbine to
withdraw some, never all, of this energy from the falling water. No
free lunch. (Yes, I know of pumped storage- benefit is strictly
economics -buy low-sell high).

"Wond" dealt with the operation of the old memory toroids.

[jaymo...@hotmail.com]

Don Kelly wrote...

>If you had a toroidal permanent magnet-it would have constant flux -essentially
> completely in the magnet -so there would be no changing flux and no induced
>voltage to produce a current. Permanent magnets will not supply any free energy.
>They work a bit like pumping water up a hill to store energy- and then running
> a turbine to withdraw some,

If I understand this correctly the initial pulse through the conducting wire
gives the toroid its circular magnetic field . But to retrieve that , how does one
get the energy back out of the toroids magnetic field? If there is no movement of
the magnet relative to the wire.
Or maybe its as wond describes where the field just gets reversed by consecutive
pulses of current in opposite directions. So its either clockwise for 0 or ccw for 1?

[Don Kelly]

See

http://en.wikipedia.org/wiki/Magnetic-core_memory

THe magnetic field in the toroid is produced by the application of two
currents whose sum is sufficient to drive the core into saturation-in
one direction (call it a 1). Energy is put into the core. Reading the
core consists of trying to reverse the magnetism and the core will end
up with the same stored energy after this happens-but the direction of
the flux will be reversed(call it a 0). While this reversal is taking
place-the core energy will drop to 0 then rise again. In the process
there is an induced voltage, depending on the rate of change of the
magnetic flux, in both the sense and read windings. During this cycle,
the device actually acts as a transformer.
If the core is initially a 0, nothing happens.

This is a bit rough and handwaving -diagrams needed

[jaymo...@hotmail.com]

Don wrote april16...
Thanks Don. I think somebody mentioned earlier that a torus
magnet doesnt have an associated external field that can be,
for instance, detected by a compass. In the same way as a compass
can detect a circular field around the conducting wire. Which seems
odd to me. All permanent magnets have external fields. Parts of which
run parrallel to the magnet( half way between poles in a bar magnet
for instance)
I dont have any convenient toroid magnets to check with, but what
do you think would happen to a toroid magnet near another external
north southfield? Would the toroid, having a circular field (cw or ccw
like the field around a conductor) experience a force pushing it left or
right in the same way as the conducting wire does?

[Don Kelly]

An ideal torus would have no external field -any field that is external
is due to defects in the material of the core (e.g a nick or some
inclusion that distorts the internal field. As far as the outside world
is concerned it isn't a magnet. If you could cut a very small slice in
the torus- you could detect the field but it would spread at the cut-how
much depends on the width of the cut. I recall trying to measure the
field external to a rectangular transformer core- it really was
negligible except at the inner side of the corners.

Now, with the external field- I need to think about this-but I suspect
that there would be only the normal force between a field and an
unmagnetized torus. From an energy point of view-I think that this would
be it. However, I haven't put much thought into this case.
Note that a toroidal magnet would be somewhat like a solenoid for which
the ends are joined -and the "current" in the solenoid winding would
be modeled as being uniformly distributed around the toroid-not as an
equivalent current at the center or the toroid.

Good question and I haven't at this time a definitive answer.

[szczepan bialek]

"Don Kelly" <dh...@shaw.ca> napisal w wiadomosci
news:XWh5v.58420$Pk4....@fx28.iad...

> On 21/04/2014 9:21 AM, jaymo...@hotmail.com wrote:
>> All permanent magnets have external fields.
>>

> An ideal torus would have no external field -
>

> Good question and I haven't at this time a definitive answer.

The definitive answer is in Maxwell's model:
http://en.wikisource.org/wiki/On_Physical_Lines_of_Force

" a circular field around a current carrying wire" is in Heaviside's model.
In Maxwell's the field is along the wire.
The magnetic lines of force are like the smoke ring (or rotating flexible
shaft).
S*

[Jos Bergervoet]

On 4/22/2014 1:35 AM, Don Kelly wrote:
> On 21/04/2014 9:21 AM, jaymo...@hotmail.com wrote:

...

>> I dont have any convenient toroid magnets to check with, but what
>> do you think would happen to a toroid magnet near another external
>> north southfield? Would the toroid, having a circular field (cw or ccw
>> like the field around a conductor) experience a force pushing it left or
>> right in the same way as the conducting wire does?

....

> Now, with the external field- I need to think about this-but I suspect
> that there would be only the normal force between a field and an
> unmagnetized torus. From an energy point of view-I think that this would
> be it. However, I haven't put much thought into this case.

...

Anapole magnetic moment:
http://phys.org/print290089874.html

--
Jos

[jaymo...@hotmail.com]

Don Kelly wrote...

>Now, with the external field- I need to think about this-but I suspect that
>there would be only the normal force between a field and an unmagnetized torus.

I assume by this you mean the torus's internal field would not interact with
any external sourced field. Seems an odd contradiction to me. Because on the one hand
the torus field cannot interact with any field from an external source. Yet it was created
from and by the field from an external source . And can be destroyed or at least
changed and manipulated by the field from an external source.
Although Im tempted to agree with you. Only because if it did respond
to the external field by moving left or right. Then it would suggest that one could then
make a homopolar motor with a torus magnet replacing the conducting wire.
Which leads to the impossibility of free energy

I havent the experimental knowledge and equipment to pulse charge and magnetize
a torus and check myself .( Ive only just bought neodium and bar magnets, a voltmeter
and some copper wire and thats expensive enough for me). But maybe theres a reader here
who can do this and test if it has a response to an external field.

[benj]

Magnetic fields do not act on each other. Forces are from magnetic
fields on charges. So if there is an external magnetic field upon a
toroidal magnet one has to ask what is the "equivalent" current that is
generating that internal field. Well as Don already noted, the
equivalent current is a solenoid bent till the ends meet and make a toroid.

So then the real question would be if you make a toroidal coil like that
carrying a current, what would be the force generated if you apply an
external field. And then you have to ask how that external field varies
in space over the coil. Seems to me it might or might not create a force
or possibly a torque.

[jaymo...@hotmail.com]

Benj wrote...

>So then the real question would be if you make a toroidal coil like that

>carrying a current, what would be the force generated if you apply an external field?

I believe you and Don have equated the toroid magnet to a toroid coil.
Can I assume that an external field can effect a force on a toroid magnet by virtue
of observing the change or creation of the toroid magnets field by the current
carrying wire in the memory writing process? If so, then I would say that a
quantifiable force will be generated on a toroidal coil that carries a current if an
external field is applied.

[Wond]

On Tue, 22 Apr 2014 20:22:10 -0400, benj wrote:

> On 04/21/2014 12:21 PM, jaymo...@hotmail.com wrote:
>> Don wrote april16...
>> Thanks Don. I think somebody mentioned earlier that a torus magnet
>> doesnt have an associated external field that can be, for instance,
>> detected by a compass. In the same way as a compass can detect a
>> circular field around the conducting wire. Which seems odd to me. All
>> permanent magnets have external fields. Parts of which run parrallel to
>> the magnet( half way between poles in a bar magnet for instance)
>> I dont have any convenient toroid magnets to check with, but what do
>> you think would happen to a toroid magnet near another external north
>> southfield? Would the toroid, having a circular field (cw or ccw like
>> the field around a conductor) experience a force pushing it left or
>> right in the same way as the conducting wire does?
>
> Magnetic fields do not act on each other. Forces are from magnetic

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

> fields on charges. So if there is an external magnetic field upon a
> toroidal magnet one has to ask what is the "equivalent" current that is
> generating that internal field. Well as Don already noted, the
> equivalent current is a solenoid bent till the ends meet and make a
> toroid.
>
> So then the real question would be if you make a toroidal coil like that
> carrying a current, what would be the force generated if you apply an
> external field. And then you have to ask how that external field varies
> in space over the coil. Seems to me it might or might not create a force
> or possibly a torque.

WADR, please clarify!

--
/home/wond/Documents/sig.txt

[benj]

What is your question? Are you thinking that when you have two magnets
and they repel it is the magnetic fields that are repelling each other?

[Don Kelly]

Wrap a wire around a torus and energize it. the current is around the
core and magnetize the torus- while there may be be forces acting on
this coil -they always add up to sweet toot. The coil doesn't move.
Now you are looking at something different- apply an external field in
the region of the torus or the toroidal solenoid. There may be forces
but there is no net force acting on the toroidal magnet or the toroidal
solenoid as there is cancellation of forces.

Here is something that does happen- if you have a "squarish" coil such
as in a transformer and pass a high current (i.e. lightning discharge)
through it- the coil will distort due to forces between the current in
one part of the coil interacting with current in another part, trying to
make the coil completely circular -these forces can be expressed in
terms of the current in one part and the flux due to the current in the
other part (i.e. Lorentz force- please look this up!). However- the coil
itself doesn't move. Similar things can happen in the end turns of
generator windings- in both cases bracing is needed. There need not be a
ferromagnetic core involved. However, in any case the sum of all the
forces and or torques is zero so no motion occurs.
A case where forces can exist is between parallel currents- for which
Lorentz applies nicely. Note that the current and the field due to the
second current produces a force perpendicular to both.

Put it this way- there is no way where you can get free energy from any
configurations of permanent magnets. It's like carrying water up a hill
and letting it fall down through a turbine- even at an impossible 100%
efficiency- energy in= energy out.

[Jos Bergervoet]

On 4/24/2014 12:57 AM, Don Kelly wrote:
> On 23/04/2014 2:34 AM, jaymo...@hotmail.com wrote:
>> Benj wrote...
>>> So then the real question would be if you make a toroidal coil like that
>>> carrying a current, what would be the force generated if you apply an
>>> external field?

...

>> If so, then I would say that a
>> quantifiable force will be generated on a toroidal coil that carries a
>> current if an
>> external field is applied.

...

> Now you are looking at something different- apply an external field in
> the region of the torus or the toroidal solenoid. There may be forces
> but there is no net force acting on the toroidal magnet or the toroidal
> solenoid as there is cancellation of forces.

This is not true for a time-dependent external field.
An incoming wave, for instance, will create a force
on the anapole moment.

Also, the stationary magnetized torus may have no
external field (although it has an external vector
potential) but if you shake it, or rotate it, it
will create outgoing EM waves. It has to be
accelerated motion of course, but that is the same
for a simple electric charge: movement at constant
velocity creates no radiation, acceleration does.

...

> Put it this way- there is no way where you can get free energy from any
> configurations of permanent magnets.

I think we all agree on this. Even if you move
them, the radiated energy comes from the work
you do by moving them.

--
Jos

[jaymo...@hotmail.com]

Don wrote...

> apply an external field in the region of the torus or the toroidal solenoid. There may be forces but there is no net force acting on the toroidal magnet or the toroidal solenoid as there is cancellation of forces.

Thanks for all the info on the subject.

If I were to hold a compass up to the side of a toroidal solenoid I assume
that the needle would point in either a cw or ccw direction around the
toroid. Following the direction of the field as dictated by the current direction
In the coil windings. In the same way as a compass needle would point if it were
placed beside a straight solenoid. Now this is because the needle is much smaller
and weaker than the solenoid. So the needle moves not the solenoid. But
if I place a very strong magnet beside the toroidal solenoid, hold it firm and
allow the toroid free movement. I would have thought the toroid would have
to move. Contrary to what you say above which is near any external field there
is no net force.?

[Wond]

On Wed, 23 Apr 2014 17:55:48 -0400, benj wrote:


>>> Magnetic fields do not act on each other. Forces are from magnetic
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>>

>

>> WADR, please clarify!
>
> What is your question? Are you thinking that when you have two magnets
> and they repel it is the magnetic fields that are repelling each other?

In a word, yes.


--
/home/wond/Documents/sig.txt

[Jos Bergervoet]

On 4/24/2014 2:32 PM, jaymo...@hotmail.com wrote:
..

> If I were to hold a compass up to the side of a toroidal solenoid I assume
> that the needle would point in either a cw or ccw direction around the
> toroid.

Not at all. If it is a "perfect" toroid without
leaking of the field it has no external field at
all and the needle would not react to it.

The toroid only has an external vector potential,
and compass needles do not detect this!

NB: As I wrote the toroid can create an external
field if you twist or shake it, but that will give
only very weak external fields unless you shake it
with MHz frequencies, and those would be too fast
for the needle to follow.

--
Jos

[jaymo...@hotmail.com]

Jos wrote..

> toroid.

>Not at all. If it is a "perfect" toroid without leaking of the field it has no external field at all and the needle would not react to it.

If I can go back to the toroid magnet rather than the solenoid. I didnt get
a definitive response from Don or Benj on this. Would an external magnetic
field have any effect on the toroid magnet? Like for instance a force
pushing it at right angles ( fleming) or any other motion due to
the two magnets being at close proximity.
Because although its clear there is no external field around the toroid
magnet. I would have thought that inside the toroid magnet the individual
atoms or domains would be aligned in specific directions and those on
one side of torus, closer to the external field *would* interact with external
magnet and/or its field.
So for instance on the side of torus thats nearest the external field, lets
say those atoms fields are pointing in one cw or ccw direction, either up or down,
that opposes or attracts the vertical external field. Would there be a subsequent
attraction/repulsion?
( from side view...external field is vertical and torus is standing on
Its rim like a wheel and you can see through its central hole , and slightly
to right or left of main vertical external north south field)

[benj]

Magnetic field do not interact with each other. In fact, it's an EM
principle known as "superposition". That means if you have two separate
magnetic fields you can treat them as entirely independent as if the
other did not exist. But you can of course, also add them up and treat
the sum as one field.

The basic interaction with magnetic fields is with the field and
charges. If you have a charged object moving in a magnetic field it
experiences the Lorentz force. Now if the charges are moving in a tube
like what we call a current, then those forces are applied to the tube.
This is how motors etc. work.

The catch here is that if one looks at the "magnetic field lines" which
are a graphic way of showing magnetic field density, one can have the
idea that forces are created by these lines such as when two magnet
poles come close and the lines "bunch up" or are "shoved apart". It
seems obvious that the forces are coming from the lines.

But they aren't. The fields aren't even shoving each other around. As
above it's just a sum of two fields: one from one magnet and one from
the other giving a total field picture. They do not "shove" on each
other or interact.

The forces between the magnets comes as always from Lorentz forces. All
magnetic fields as far as is known come from electric currents. And in
the case of permanent magnets it's tiny internal circulating atomic
currents. Thus all the forces you feel playing with magnets comes from
the forces of magnetic fields on those tiny circulating currents which
like a motor are transferred to the material of the magnet itself and
you feel the push or pull.

OK?

[benj]

On 04/24/2014 03:26 PM, jaymo...@hotmail.com wrote:
> Jos wrote..
> > toroid.
>
>> Not at all. If it is a "perfect" toroid without leaking of the field it has no external field at all and the needle would not react to it.
>
> If I can go back to the toroid magnet rather than the solenoid. I didnt get
> a definitive response from Don or Benj on this. Would an external magnetic
> field have any effect on the toroid magnet? Like for instance a force
> pushing it at right angles ( fleming) or any other motion due to
> the two magnets being at close proximity.

Generally speaking just as the geometry of the torus cancels all
external magnetic fields, the same thing happens to the forces generated
by an external field. But it's not all quite so perfect. As Jos has
mentioned the torus represents an "Anapole moment". This is found by an
expansion of the fields and finding the moment term. A moment means that
the torus responds to an external field with a torque (not linear push
or pull). However the Anapole moment is a second order effect and thus
very small compared to say a dipole moment which is small compared to
normal field forces (and zero for the torus).

So I guess the answer is yes, the toroid experiences a tiny action, but
it's a torque and not some linear force.

[Don Kelly]

On 24/04/2014 5:32 AM, jaymo...@hotmail.com wrote:
> Don wrote...
>> apply an external field in the region of the torus or the toroidal solenoid. There may be forces but there is no net force acting on the toroidal magnet or the toroidal solenoid as there is cancellation of forces.
>
> Thanks for all the info on the subject.
>
> If I were to hold a compass up to the side of a toroidal solenoid I assume
> that the needle would point in either a cw or ccw direction around the
> toroid.

Try it- The field of the magnetized toroid is internal to the toroid
barring imperfections in the toroid material, or, in the case of a
solenoidal coil-physical limitations of the coil. There is material on
the net for both. Do a search. Even for a rectangular transformer core-
there is essentially no externally component of this field except at
inner corners where fringing occurs. A weak field can be detected near
the core but this is due to the fact that unlike a permanent magnet, the
current is at a finite distance from the core. What may happen is that
the compass needle is near a ferromagnetic material so the compass will
point toward the core- in the same way that it will be deflected by a
nearby piece of iron. Why? The local earth's field will be distorted by
such pieces of magnetic material. In fact the distortion will be higher
with a steel or cast iron ring than with a ferrite as the permeability
of the steel is very much higher than that of the ferrite (say 5000 muo
vs 6 muo).
There is a form of mathematical modeling -finite element modeling which
is used for analysis of fields- electric, magnetic, thermal, mechanical
stress, etc- all of which have similarities-this approach is often used
where theoretical calculations are too complex- It fits the observed
facts. I have done this for a magnetized ferrite core. I needn't have
done so as a theoretical analysis takes less than a minute, for such a
core. External field is 0 -the outside world doesn't know the core is
magnetized. Look up Femm which is a free (but awkward program to do
this analysis).

Following the direction of the field as dictated by the current direction
> In the coil windings. In the same way as a compass needle would point if it were
> placed beside a straight solenoid.

The compass needle is in an external field as field lines are always
closed -so the field of a solenoid is closed externally to the solenoid-
hence affecting a compass needle (or iron filings) The field lines in a
magnetized torus(or ideal toroidal solenoid)- are closed withing the
torus-not externally.

Now this is because the needle is much smaller
> and weaker than the solenoid. So the needle moves not the solenoid. But
> if I place a very strong magnet beside the toroidal solenoid, hold it firm and
> allow the toroid free movement. I would have thought the toroid would have
> to move. Contrary to what you say above which is near any external field there
> is no net force.?
>

What I suggest is that you do searches regarding basic principles and
some specific material with regard to toroidal magnets or torus's
excited by current carrying coils There are good references even on
Wikipedia or from some physics or engineering sources.

[Don Kelly]

Benj is looking at it from a more fundamental approach. A magnetic field
can be expressed as due to moving charges- the problem is that it is
hard to measure the charges and their velocity- but it is easy to
measure the field that is produced.
So, in that way, one can think of the interaction between a field and a
current or two known currents(involving the field of one with respect to
the other) or between two fields as one would see with two magnets-
repulsion or attraction. Is this basic? No. Is it useful? Yes.

[jaymo...@hotmail.com]

Benj wrote..

>Generally speaking just as the geometry of the torus cancels all

> external magnetic fields, the same thing happens to the forces generated...

Interesting info thanks.
Incidentally my semantics may have been misread but I definitely never
thought that fields interact with fields on their own. You need the magnet
to get the field.
I would have thought that the geometry of the torus wouldnt cancel
out all the forces. Take for instance this illustration below. The "O" is the
torus magnet seen from the side, looking through the hole. Standing
up on its rim like a wheel and seen from the side. The N-S is the vertical field
of the external magnet and its slightly to the left of the torus.
N
O
S

The way I see it is if the torus magnet has a cw or ccw field then the field
either points up or down on the right hand side. Now because the magnet
is only a collection of atoms, then it must be safe to assume that an atom
at the left hand side of the torus must actually have its NS field pointing
either up or down. Depending on the field direction. And the reverse for
an atom located at the right hand side of the torus. Now logic dictates to
me that the atom on the left being closer to the external magnet and
its field, must have a greater force exerted on it then the atom on the right.
So the conclusion for me is that there is a net force either pushing down
or up on the toroids left side. Small, but net.That must be the torque you mention

[Wond]

OK. Thanks for that. (a graphic way of showing magnetic field density
= slaps forehead)(internal circulating atomic currents= update mental
model immediately) This is gonna be fun.



--
/home/wond/Documents/sig.txt

[Don Kelly]

Actually. given such a situation- there will be no up/down force. Assume
the torus moves up a miniscule distance (dy) due to a magnetic force
(F). This implies that work is done (f*dy) However such an motion
results in no change in the external and internal fields and no energy
change or transfer- that is no work is done. There is no change in field
energy with position.
There are a few other flaws to consider but that's enough.
If you have two bar magnets in proximity, a movement of one with respect
to the other can result in a change in magnetic field distribution.

[jaymo...@hotmail.com]

Don wrote...

> Actually. given such a situation- there will be no up/down force. Assume the torus moves up a miniscule distance (dy) due to a magnetic force (F). This implies that work is done (f*dy) However such an motion results in no change in the external and internal fields and no energy change or transfer- that is no work is done. There is no change in field energy with position

I dont quite understand this.. Are you saying that even if the toroid magnet
moves in response to the external magnet,..no force was used?
What moves the toroid if not an applied force from somewhere.
Do you maybe mean that when I place the toroid into position
I add the extra force, its stored in the two fields and when I release
the toroid that stored force is used to move the toroid. Leaving the
two magnets fields the same as before I put the toroid into position?

[Don Kelly]

1) I made a premise that there is a force due to interaction between the
external field and the torus field. No external energy source is being
considered as how the torus got to its original position in the external
field doesn't matter. I have as I understand your scenario, considered a
vertical external field and a vertical motion (implying a vertical force).

2)Then I noted that such motion requires that energy from the magnetic
field is converted to mechanical energy.

3)However, the field after the movement due to "magnetic" forces would
be the same as before (as would be the case for the scenario you gave).
and but for the conversion of field stored energy to mechanical energy
the field stored energy must decrease. This isn't happening.
4)since there is no energy change there is no mechanical energy produced
for a finite change in position- there can be no force.
The premise (1) is false.

[Jos Bergervoet]

On 4/27/2014 1:57 AM, Don Kelly wrote:
> On 25/04/2014 5:50 PM, jaymo...@hotmail.com wrote:
>> Don wrote...

...

> 4)since there is no energy change there is no mechanical energy produced
> for a finite change in position- there can be no force.

Exactly. And likewise there *can* be a force if the
external field is time-dependent, i.e. coming from a
source of EM radiation, which radiates energy.

In this case (the toroid anapole moment) you could
use an external field created by a 2nd toriod below
the first one (like two donuts above each other).
Then an AC current in the 2nd toroid is needed, or
else it would not have an external field! And the
1st one of course keeps its DC current and therefore
still has no external field by itself, as before.

With this setup, you will find a net vertical force
on the 1st torus, oscillating with the frequency
used. And you can extract energy out of its periodic
movement if you want to. Of course you could also
hold it rigidly fixed, or let it move up and down
freely, in which cases there still is no net energy
transfer (in the latter case only temporary energy
storage in the motion).

This setup is not the only possibility, you can
replace the 2nd toroid by a simple vertical electric
dipole point source at the same location (vertically
below the 1st toroid) which might be easier to
analyze mathematically.

--
Jos

[Jos Bergervoet]

On 4/24/2014 9:26 PM, jaymo...@hotmail.com wrote:
> Jos wrote..
> > toroid.
>
>> Not at all. If it is a "perfect" toroid without leaking of the field it has no external field at all and the needle would not react to it.
>
> If I can go back to the toroid magnet rather than
> the solenoid. I didnt get a definitive response from
> Don or Benj on this. Would an external magnetic
> field have any effect on the toroid magnet?

I didn't follow whether DOn or benj addressed this,
but there is a complication with the toroid magnet
(as opposed to the solenoid) in that it consists
of magnetic material. Even if it isn't magnetized
(i.e. no flux inside the core) any magnet will still
attract it, just like it can pick up iron nails
whether they are first magnetized or not.

So if you want to see whether the presence of this
internal toroidal flux makes any difference, it is
really better to analyze the (ideal) toriod coil
with DC current.

--
Jos

[jaymo...@hotmail.com]

Jos and Don wrote about toroids.


Thanks for those replies.I am considering trying it out myself using my
copper wire,and blow some cash buying a strong 12 volt car type battery
and try to magnetize a small steel washer.
The only problem is that unmagnetized, the washer is strongly
attracted to my alnico magnet. Will my attempts to magnetize
the washer be in vain? In that the induced toroidal field in the washer
wont be strong enough to give me the desired effect .
Maybe this will only work with a ferrite toroid core. Or,.. only a larger
steel ring than the washer?

[Don Kelly]

I had, some time ago, suggested a test with a torus in an external
field. Hang it on a pendulum and note the deflection. Now magnetize the
torus and repeat to see if there is any difference. What now interests
me is the effect of the external magnet on the distribution of the flux
in the torus--magnetized and unmagnetized.

[Don Kelly]

Hang the washer on a string in such a way that it is attracted to a
magnet as some distance to the side. put a paper with a scale behind it
so you can measure the deflection. Same set up after magnetizing the
washer. I wouldn't use a car battery but rather something like a large
wall wart with DC output and lots of turns of wire. A car battery has
too high a short circuit current to fool with (and is more
expensive-spend some on a decent multimeter).

[Jos Bergervoet]

On 4/28/2014 12:16 AM, Don Kelly wrote:
> On 27/04/2014 3:45 AM, Jos Bergervoet wrote:

...

>> So if you want to see whether the presence of this
>> internal toroidal flux makes any difference, it is
>> really better to analyze the (ideal) toriod coil
>> with DC current.
>>
> I had, some time ago, suggested a test with a torus in an external
> field. Hang it on a pendulum and note the deflection. Now magnetize the
> torus and repeat to see if there is any difference. What now interests
> me is the effect of the external magnet on the distribution of the flux
> in the torus--magnetized and unmagnetized.

But they just add up! The flux of the external
field will be added to the flux in the torus from
the DC current in its winding, I mean.

So which question do you want to be answered with
this experiment?

--
Jos

[jaymo...@hotmail.com]

Don wrote...

> I had, some time ago, suggested a test with a torus in an external field. Hang it on a pendulum and note the deflection. Now magnetize the torus and repeat to see if there is any difference. What now interests me is the effect of the external magnet on the distribution of the flux in the torus--magnetized and unmagnetized.

Yes this is what I was trying to do but failed, laughably. If thats a word.
I had the steel washer and copper wire, magnet and thread. But only 9v
Battery. Because the steelwasher is so small and because I assumed
from reading up on toroid solenoids that its hard to avoid a break
in the internal field where the (current carrying wire)winding comes on
to the torus core. I decided on the method used in memory cores
where one puts a current from a battery through a wire through
the hole in the middle of the torus. Of course is doesnt work as
I find on further reading that a 12 volt battery is needed. Im still
debating on whether ,in my limited budget, to go out and buy the battery
to conduct an experiment where Ive already been told the outcome by
the experts. I suppose, like all travellers, I just want to see for myself.
But I also noted how strong the washer is attracted to the alnico bar
magnet. and realised how difficult it will to conduct any experiment
under such crude conditions with so little experimental experience.

[jaymo...@hotmail.com]

Don Kelly wrote...

>Hang the washer on a string in such a way that it is attracted to a magnet as some distance to the side. put a paper with a scale behind it so you can measure the deflection. Same set up after magnetizing the washer. I wouldn't use a car battery but rather something like a large wall wart with DC output and lots of turns of wire. A car battery has too high a short circuit current to fool with (and is more expensive-spend some on a decent multimeter).

I went to the local auto supply store yesterday and was just about to
buy a large 12 volt acid ,car type battery, take it home, and hook up 6 inch
piece of copper wire to the two terminals. ( threaded through the hole
of my steel washer) But the cost bothered me and I thought Id wait
a bit. Good thing I did. I didnt realize that this would have essentially
been shorting it and expoloding it in my face. I only found this out after
I got home and looked up car batteries etc on google. So thats it for
any ideas I have about experiments with electricity. Mains or batteries.

Im stuck with just asking questions here. There was one more I thought
I would ask. Its a variation on my toroid inquiries.
If I atttached two bar magnets together in an H format. So that the vertical
bars of the H were the two bar magnets and the crossbar of the H was a
rigid armature holding the two magnets in position. Then at the center
of the cross bar of the H, the structure is attached to a stiff but flexible wire
shown in the illustration below as a dot, representing the wire coming
out of the screen. This wire is also held firm. The purpose being to
hold the H magnet structure in place in the external field. But giving
it a little flexibility so we can see if there is any net force on the H
magnet structure from the external N S field. Either up, down,across
or even rotational.
One magnet has N up. The other has north facing down. This H structure
is then placed just to the right of a strong external magnetic field.
N
N S
| |
---*---
| |
S N
S

Would this setup, like the torus setup yield no motion.
Or would the left magnet on the H armature experience
a stronger downward force, then the right magnet . Being
closer to the external field, and result in some sort of motion
Of the whole Two bar armature..like maybe a ccw rotation.?

[benj]

On 04/29/2014 08:06 AM, jaymo...@hotmail.com wrote:
> Don Kelly wrote...
>> Hang the washer on a string in such a way that it is attracted to a
>> magnet as some distance to the side. put a paper with a scale
>> behind it so you can measure the deflection. Same set up after
>> magnetizing the washer. I wouldn't use a car battery but rather
>> something like a large wall wart with DC output and lots of turns
>> of wire. A car battery has too high a short circuit current to fool
>> with (and is more expensive-spend some on a decent multimeter).
>
> I went to the local auto supply store yesterday and was just about
> to buy a large 12 volt acid ,car type battery, take it home, and hook
> up 6 inch piece of copper wire to the two terminals. ( threaded
> through the hole of my steel washer) But the cost bothered me and I
> thought Id wait a bit. Good thing I did. I didnt realize that this

> would have essentially been shorting it and exploding it in my face.

> I only found this out after I got home and looked up car batteries
> etc on google. So thats it for any ideas I have about experiments
> with electricity. Mains or batteries.
>
> Im stuck with just asking questions here. There was one more I
> thought I would ask. Its a variation on my toroid inquiries.

You are so uninterested in science that you let a little car battery
explosion in your face deter your scientific curiosity? Shame on you!

Huge explosions are what makes science FUN! There are things with a LOT
more power than car batteries! :-)

The answer of course to a science problem is (wait for it) MORE SCIENCE!

In this case what you need to study is called Ohm's law. It says that
I = V/R. I is the current through the core. V is the voltage of the
battery. and R is the RESISTANCE you've inserted in series with the
battery to keep it from blowing up! By picking the resistance you can
set the current you put through your cores. The battery won't give big
sparks then.

There are lots of things you can use for a load, but the usual is some
kind of nichrome battery load or heater. Harbor Freight sells things
like this cheap. Best is to see if you can locate some nichrome heater
wire especially the kind made in wide flat ribbons which will have a low
Ohms and take a lot of heat without burning up (virtually all your
energy from the battery goes into the load).

There are many today who think that science advances by mathematical
speculation. But the TRUE TEST of science is always observing what
REALITY says about it. Decent experimental physics requires a LOT of
skill and talent typically not possessed by theoreticians and pretty
much portrayed as irrelevant lab assistant technician work by everyone
else (post Einstein). Designing a workable measurement without
"loopholes" as you see can require a lot of understanding often in areas
well removed from the particular problem you are trying to study!

Good Luck!

[Jos Bergervoet]

On 4/29/2014 2:06 PM, jaymo...@hotmail.com wrote:
...

> One magnet has N up. The other has north facing down. This H structure
> is then placed just to the right of a strong external magnetic field.
> N
> N S
> | |
> ---*---
> | |
> S N
> S
>
> Would this setup, like the torus setup yield no motion.

You now made a magnetic quadrupole! This one
will feel forces from an external field, even
if that external field is static. It also does
generate its own external field (without
the need to first twist or shake it).

In both these respects it differs from the
earlier discussed toroid, which is an *anapole*
and feels no force from an external static
field and has no external static field of its
own. But still, mysteriously, it's not the same
as an object without any internal field, as
soon as time-varying effects are involved!

(And of course quantum mechanically this
anapole is detectable by the Aharonov-Bohm
effect, but classically you can also do it
just using a time-dependent field.)

--
Jos

[Don Kelly]

I just did a finite element plot- with a torus (material Alnico5) and a
bar magnet to one side. With the torus unmagnetized this program showed
the expected sort of flux distribution and also allowed me to find the
found the weighted stress tensor. With the torus magnetized, the flux
distribution was different and the stress tensor was about 25% of that
in the magnetized case.
with the unmagnetized torus, the B inside the torus on the far side is
3/4 that on the near side-not quite balanced ( mu is about 1.5mu0)
With the torus magnetized, the variation in the internal field is quite
small- and it appears almost as if the external field is forced out.

I used femm 4.2 (free) and a crude approximation of the torus as 4
quadrants magnetized in the average direction of each quadrant. Use of
more sections would be better but, as is, it appears that there is
enough information to suggest that the magnetization of the torus
reduces the force due to the external magnet.

[Don Kelly]

You don't need to worry about a break in the field you won't get it-
use the winding to magnetize the washer and then remove the coil-leaving
a magnetized washer. 12V isn't needed -but you need ampere turns - many
turns at a low current is as good as 1 turn at a high current. It may
well be that with a steel washer- the field that you get is easily
destroyed but on the other hand it takes more amp-turns to magnetize a
ferrite core.

All that I can suggest to you is that you find some good references on
magnetism and electricity as well as force and energy relationships.
Wiki is generally reasonable and there are some university sites but Do
not go to Keelynet!!!!

[benj]

On 04/29/2014 08:47 PM, Don Kelly wrote:

> All that I can suggest to you is that you find some good references on
> magnetism and electricity as well as force and energy relationships.
> Wiki is generally reasonable and there are some university sites but Do
> not go to Keelynet!!!!

Hey, what's wrong with a little "free energy"? Better than paying!
Anyway, a nice N-machine can give all the current you'll ever need.

[Jos Bergervoet]

On 4/30/2014 2:32 AM, Don Kelly wrote:
> On 27/04/2014 11:04 PM, Jos Bergervoet wrote:
>> On 4/28/2014 12:16 AM, Don Kelly wrote:
>>> On 27/04/2014 3:45 AM, Jos Bergervoet wrote:
>> ...
>>>> So if you want to see whether the presence of this
>>>> internal toroidal flux makes any difference, it is
>>>> really better to analyze the (ideal) toriod coil
>>>> with DC current.
>>>>
>>> I had, some time ago, suggested a test with a torus in an external
>>> field. Hang it on a pendulum and note the deflection. Now magnetize the
>>> torus and repeat to see if there is any difference. What now interests
>>> me is the effect of the external magnet on the distribution of the flux
>>> in the torus--magnetized and unmagnetized.
>>
>> But they just add up! The flux of the external
>> field will be added to the flux in the torus from
>> the DC current in its winding, I mean.
>>
>> So which question do you want to be answered with
>> this experiment?
>>
> I just did a finite element plot- with a torus (material Alnico5) and a

I see. Not an (air-core) toroid coil but a torus of
magnetic material. That's different.

> bar magnet to one side. With the torus unmagnetized this program showed
> the expected sort of flux distribution and also allowed me to find the
> found the weighted stress tensor. With the torus magnetized, the flux
> distribution was different and the stress tensor was about 25% of that
> in the magnetized case.
> with the unmagnetized torus, the B inside the torus on the far side is
> 3/4 that on the near side-not quite balanced ( mu is about 1.5mu0)
> With the torus magnetized, the variation in the internal field is quite
> small- and it appears almost as if the external field is forced out.

Probably this happens for most shapes with closed
internal flux: when magnetized (close to saturation)
there is less room to increase the flux further by
external field. But decreasing it in one part is
still possible I would guess. So (at most) half the
effect could be gone?

> I used femm 4.2 (free)

Thanks. I made a note!

> and a crude approximation of the torus as 4
> quadrants magnetized in the average direction of each quadrant. Use of
> more sections would be better but, as is, it appears that there is
> enough information to suggest that the magnetization of the torus
> reduces the force due to the external magnet.

And if the torus is replaced by a simple bar shape,
then I expect the force to be reduced as well if your
external field is aligned with the internal bar flux
and in the same direction, but not reduced if it is
opposite to the internal flux. But the bar cannot
be magnetized very far (no closed magnetic circuit)
so it might not be visible.

--
Jos

[Don Kelly]

One consideration that I had not considered is that in the model, the
near side of the torus had a field which was oriented essentially in the
same direction as the external field. I repeated by reversing the
polarity of the external field and found results close to the previous
situation- reduction of force. As before, the external field has little
effect on the filed within the torus -it is less than what would be
expected from a vector summation of the individual fields and the
external field is distorted.

Replacing the torus by a bar would yield the same result as for 2 bar
magnets (In this model- I used a bar magnet near the torus for the
external field-as this gave a non-uniform field in the region of the torus.

Certainly I could have used a coil in the model with or without a
magnetic core. I have some reason to believe that with a high
permeability core there would be less effect.

[Don Kelly]

Begone-minion of Satan- all that free energy helps his heating bill.

[hth...@gmail.com]

On Wednesday, May 22, 2013 at 11:53:38 AM UTC-4, jaymo...@hotmail.com wrote:
> Ive tried making a torus shaped (circular or donut shaped)
> magnetic field by arranging a circle of bar magnets.
> Unfortunately all I get is a series of north south poles
> In a circle.
> My question is.. are torus shaped permanent magnets
> possible to make and buy. Ive noticed horseshoe shaped
> Fields are available so thats half way there.
> Thanks.

I've seen them used on the backs of some stereo speakers.

[benj]

Of course there are torus shaped magnets. But how do you want the
magnetic field to go? Speaker magnets are magentized front to back
through the thickness of the toroid. You can magnetize them like bar
magnets in a circle but then the field just goes round and round inside
and doesn't come out. Not useful or much unless you saw a slot in the
toroid. And if you do that it really is just sort of an odd shaped
horseshoe magnet.

Search for magnets online. Google is your friend.

[Jos Bergervoet]

On 5/10/2016 3:37 AM, benj wrote:
> On 05/09/2016 09:29 PM, hth...@gmail.com wrote:
>> On Wednesday, May 22, 2013 at 11:53:38 AM UTC-4, jaymo...@hotmail.com
>> wrote:
>>> Ive tried making a torus shaped (circular or donut shaped)
>>> magnetic field by arranging a circle of bar magnets.
>>> Unfortunately all I get is a series of north south poles
>>> In a circle.
>>> My question is.. are torus shaped permanent magnets
>>> possible to make and buy. Ive noticed horseshoe shaped
>>> Fields are available so thats half way there.
>>> Thanks.
>>
>> I've seen them used on the backs of some stereo speakers.
>
> Of course there are torus shaped magnets. But how do you want the
> magnetic field to go? Speaker magnets are magentized front to back
> through the thickness of the toroid. You can magnetize them like bar
> magnets in a circle but then the field just goes round and round inside
> and doesn't come out. Not useful or much unless you saw a slot in the
> toroid.

Without the slot it still gives a strong vector potential
outside! That would be visible by things like the
<http://en.wikipedia.org/wiki/Aharonov–Bohm_effect>
but perhaps that is indeed not extremely useful in the
sense of practical applications.

Also, without the slot, you would have this interesting case:
1) The magnet would radiate waves if you rotate it.
2) The magnet has zero external field if kept stationary.

That looks like something that might in some cases be useful..

--
Jos

[benj]

On 05/10/2016 03:12 AM, Jos Bergervoet wrote:
> On 5/10/2016 3:37 AM, benj wrote:
>> On 05/09/2016 09:29 PM, hth...@gmail.com wrote:
>>> On Wednesday, May 22, 2013 at 11:53:38 AM UTC-4, jaymo...@hotmail.com
>>> wrote:
>>>> Ive tried making a torus shaped (circular or donut shaped)
>>>> magnetic field by arranging a circle of bar magnets.
>>>> Unfortunately all I get is a series of north south poles
>>>> In a circle.
>>>> My question is.. are torus shaped permanent magnets
>>>> possible to make and buy. Ive noticed horseshoe shaped
>>>> Fields are available so thats half way there.
>>>> Thanks.
>>>
>>> I've seen them used on the backs of some stereo speakers.
>>
>> Of course there are torus shaped magnets. But how do you want the
>> magnetic field to go? Speaker magnets are magentized front to back
>> through the thickness of the toroid. You can magnetize them like bar
>> magnets in a circle but then the field just goes round and round inside

>> and doesn't come out. Not useful for much unless you saw a slot in the

>> toroid.
>
> Without the slot it still gives a strong vector potential
> outside! That would be visible by things like the
> <http://en.wikipedia.org/wiki/Aharonov–Bohm_effect>
> but perhaps that is indeed not extremely useful in the
> sense of practical applications.
>
> Also, without the slot, you would have this interesting case:
> 1) The magnet would radiate waves if you rotate it.
> 2) The magnet has zero external field if kept stationary.
>
> That looks like something that might in some cases be useful..

Interesting! Please explain why you think that rotating such a magnet
would produce waves. This would mean that if I took a long
current-carrying solenoid and bent it into a toroid and then rotated it,
it would radiate, right?

[Jos Bergervoet]

Rotating would of course periodically rotate the external
vector potential A. And if A changes then there is an E-field
from dA/dt which is then also periodical, so inevitably there
is also a non-zero B-field from rot B = dE/dt. (Of course you
can also get them via Liénard-Wiechert or Jefimenko's Eqs.)

> This would mean that if I took a long
> current-carrying solenoid and bent it into a toroid and then rotated it,
> it would radiate, right?

Yes if you rotate it that would be the same. But if you use
AC current, you can keep its position fixed! (Although the
former gives circular polarization, it is essentially the
same as two 90 degree crossed versions of the latter.)

--
Jos

[benj]

AH! Yes. I figured it out after I wrote this. I was rotating about axis
through the hole in the toroid which does not change the field which is
the natural way you'd think of a "wheel" rotating.

However, to get radiation you need to rotate the toroid about an axis
sideways through it. Since the A field is not spherical about the toroid
it creates dA/dt just as you say.

>> This would mean that if I took a long
>> current-carrying solenoid and bent it into a toroid and then rotated it,
>> it would radiate, right?
>
> Yes if you rotate it that would be the same. But if you use
> AC current, you can keep its position fixed! (Although the
> former gives circular polarization, it is essentially the
> same as two 90 degree crossed versions of the latter.)

Obviously AC creates dA/dt. So that wasn't the problem.

As you say it is an interesting phenomena since here you have a toroidal
piece of material that has no external magnetic or electric field when
sitting there, but when you spin it, it creates EM fields in the
formerly field-free space around it. Cute.

And I'm pretty sure from my knowledge of TV and Hollywood that if you
rotate the toroid fast enough, a "stargate" is formed right in the
center hole!

[Jos Bergervoet]

On 5/10/2016 9:06 PM, benj wrote:
..
..

> As you say it is an interesting phenomena since here you have a toroidal
> piece of material that has no external magnetic or electric field when
> sitting there, but when you spin it, it creates EM fields in the
> formerly field-free space around it. Cute.

Also, it shows that you don't need a spooky Aharonov-Bohm
effect to detect a pure A-field, you just need to rotate
the source, which brings it back to the domain of classical
physics.

> And I'm pretty sure from my knowledge of TV and Hollywood that if you
> rotate the toroid fast enough, a "stargate" is formed right in the
> center hole!

Well, perhaps, but that may be a bit outside the domain of
classical physics..

--
Jos