Magnetic Field "Blocker"

[calbiterol]

I was actually hit by a rather interesting idea last night. Going with the scitoys example here, so if things are confusing, just ask. Using electromagnets in a halbach array, and using a magnet in place of the last steel ball bearing (before the projectile, that is), and some compex timing circuitry, you could make a fully-auto gauss rifle. I don't have any idea what kind of rate of fire it would have, but it would probably just get better as the speed increased.

 

Heres the setup:

 

The dark green are the halbach electromagnets, and the yellow arrows indicate the original augmented direction. The grey circles are spherical steel ball bearings, the squiggly's denote an ommitted portion (in other words, part of the barrel/firing chamber is missing there), the red circle is a spherical magnet, and the blue thing is a projectile. Note that this thing has no scale whatsoever. So the initial force (hitting the first electromagnet on the left) transfers it's kinetic energy to the first steel ball bearing, which (because it's a halbach array) has very little electromagnet attraction to the cancelled side (there's still a small field there, though, so the ball should stay attached to the electromagnet). This transfer of energy knocks the ball bearing to the next halbach array, and so on, until it gets to the "spherical magnet." This is actually not a plain old magnet, it's a halbach array with a steel hemisphere at each end, held by both glue and magnetic attraction. The augmented field direction is the same as the original direction of the other halbach arrays, if that makes any sense - but it's a fixed magnet array. Anyway, this gets hit forward just like everything beforehand, but then it hits the last halbach array. There, it hits (and flips) a switch (the bright green box). This switch flips the polarity of all the halbach arrays. Magnetic attraction pulls the fixed-magnet array back towards the electromagnet array it originally came from, reversing the cycle. This same process is used before the first electromagnet array on the left to reverse the process again, bringing us back to where we started.

 

If that makes sense, then... you should understand how it becomes self-resetting. If it doesn't make sense, just ask.

[d22k]

it makes perfect sense, it would be fascinating to see it in action its just it would be a bitch to build =D

 

i don't see the need for any complex electronic though... i imagine building the IC board would be a piece of cake, its the 30 tiny electro magnets that would be difficult...

 

where do you live btw?

[calbiterol]

Northern Illinois, Chicago area. Not so much a bitch to build, you just have to find the right materials. A lathe would be nice, too.

 

I was just guessing at the complex electronics. After having thought about it (read: made the schematic) I realized I didn't. But I forgot I'd said that complex electronics would be needed. The complex part would be the loading mechanism. It always is - in all of my dealings with designing unbelievably insane paintball guns, the loading mechanism is the issue, especially since paintballs can be crushed. But that's a really, really long story, and one crazy tangent.

[d22k]

heh heh, a gauss paintball marker would rock! assuming the ball wouldn't rupture....

 

you've given me a lot of food for thought

 

some small diameter nails ~5mm would be perfect cut into cross sections.... any electric motor would provide all the wire neccesary.

 

Ill have a go at building a small halbach array using them, and let u know how it goes

[calbiterol]

Have fun, and succeed where I always fail - building the things I think up.

 

[indian]

I stumbled on this searching for interesting "Halbach" applications (to play with)

Unfortunately I do not "get" this gun idea.

How can it be true that the "next" ball keeps on being faster than the previous?

[labview1958]

Let's say I have a hollow sphere covered with a superconducting material immersed in liquid nitrogen. If a weak magnet is brought near it would the magnetic field penetrate inside the hollow sphere? If not I have created a magnetic shield!

[MetaFrizzics]

OMG. The Electrical Engineer inside me forces me to speak out after reading the beginning of this thread:

 

Mu-metal does indeed 'shield' items or 'block' electromagnetic waves.

However, to think that you will 'block' the lines of force and hence the 'pull' of a magnet is a complete misunderstanding of what Mu-metal shielding actually does, how it does it, and what it can be used for.

 

Mu-metal cages and covers are used to 'block' (by absorption of field energy of) MOVING magnetic fields (from A.C. currents!), which induce electrical currents in sensitive circuits. That is, a Mu-metal box or 'Faraday' cage captures the energy from radio waves and other changing electromagnetic fields, such as those radiated by (60 Hz) transformers or chokes in power supplies.

 

Mu-metal cannot block a permanent or D.C. magnet. All that happens is the Mu-metal becomes magnetized and hence becomes part of the magnet!

 

The theory behind Mu-metal 'blocking' is not 'true' blocking like in the sense of electrostatic 'shielding', where forces are screened out. In the case of Mu-metal, significant portions of the energy in the electromagnetic field are diverted and converted into heat, or re-radiated in a different direction /orientation than before, improving the shielding of sensitive parts, which usually have more and less sensitive orientations.

 

Mu-metal really is best understood as an 'antenna' that steals electromagnetic radiation by absorption or diverts it by electrical currents. Something like a large metal framed building would reduce or reflect your T.V. signal causing 'ghost' images and poor quality reception.

 

At best a Mu-metal cap or plate might act as a 'magnet keeper' by providing a more direct and low resistance route between the poles of your magnet. But any piece of iron or nickel would do the same thing.

[DQW]

Mu-metal cannot block a permanent or D.C. magnet. All that happens is the Mu-metal becomes magnetized and hence becomes part of the magnet!

I was going to say "hogwash" !!! But I'll be polite and so, I'll merely ask you to back up your claim with a reference to this in a peer-reviewed journal or something that may be considered a text-book.

[swansont]

Mu-metal cannot block a permanent or D.C. magnet. All that happens is the Mu-metal becomes magnetized and hence becomes part of the magnet!

 

The theory behind Mu-metal 'blocking' is not 'true' blocking like in the sense of electrostatic 'shielding'' date=' where forces are screened out. In the case of Mu-metal, significant portions of the energy in the electromagnetic field are diverted and converted into heat, or re-radiated in a different direction /orientation than before, improving the shielding of sensitive parts, which usually have more and less sensitive orientations.

[/quote']

 

 

Sorry, but...no.

 

Mu metal has a high permeability so the field lines "like" to be there, and thus not in the shielded area.

 

It's true that you can't use this material to shield one magnet from another, but you can shield a region of space from magnetic fields, and that region can be either inside or outside the shield, depending on how you set it up.

[MetaFrizzics]

Solving jointly all of the equations with regard to the integration constants, we finally obtain:

 

[math]S = \frac{H^I}{H_O} = \frac{4\mu_s r^2_e}{r^2_e(1 + \mu_s)^2 - r^2_i (1 - \mu_s)^2} = \frac{4}{\mu_s} \frac{1}{1-\frac{r^2_i}{r^2_e}} [/math]

 

This equation indicates that magnetostatic shielding effectiveness depends upon magnetic permeability of the shield and its radius and thickness. Even if the shield's magnetic permeability [math]\mu_s > 1[/math] and the shield thickness is small, [math] ( r_e = r_i ) [/math], the shield effectiveness is not very good. Only very thick shields with large permeability provide protection from static and low-frequency magnetic fields.

 

Shielding effectiveness [math]A_m = 20 log |1/s| in dB [/math].

 

While the Equation above is simple, the magnetostatic problem is not! The following complicating factors make the design and evaluation of magnetostatic shielding a very tricky business:

 

(1) High magnetic permeability materials are non-linear: with change of field intensity the permeability varies, reaching saturation quickly at key induction values. Shielding effectiveness is also affected by some parameters that are specific to magnetic materials ( i.e. mu-metals), including hysteresis, magnetostriction, and core loss. All these factors must be accounted for not only in shield design but also in performance measurements.

 

(2) Material properties vary broadly with respect to manufacture, handling conditions, and especially temperature, flexing, and impact shock. As a rule the higher the permeability of the material, the larger the instability in shielding effectiveness, with experimental variances observed as large as TENS of decibels.

 

(3) You have probably noticed that the Equation above doesn't include frequency, (rightfully so for magnetostatics) So strictly speaking it is only applicable at d.c. or low frequencies where equivalent penetration depth [math] \delta = (2/(\omega\mu\sigma))^{1/2} [/math] is large. But because of high magnetic permeability, the skin effect becomes significant at VERY low frequencies, leading to a reduction in the effective thickness of the shield. This in turn leads to a deterioration of the magnetostatic shielding effectiveness. The only good news is that with frequency rise other mechanisms can enter the picture.

 

To give a practical grip on the picture, Here are some measurements done with Helmholtz coils:

 

A 1/16th in. iron pipe provided around 20-25 dB shielding effectiveness at 50 - 5KHz. By comparison, an RG59 cable shielded by relatively thin but high permeability double-layer conetic braids yielded only 14-20dB and 16-23dB. As a rule it isn't simple to obtain large cable shielding effectiveness in magnetostatic fields period.

(For refs see Sellers J et al, "Flexible Braids for Improved Magnetic Shielding of Cables" 1978 EMC symposium Alta. & Rikitake, T. "Magnetic and Electromagnetic Shielding" 1987 226 p etc.)

[MetaFrizzics]

"Tranformers are notorious avenues for introducing noise on powerlines. ... The magnetic field cannot be confined and can be a source of troublesome hum. Since the field is highly directional, the problem can sometimes be minimized by orienting the transformer in the direction that produces the least interference, or it can be located further away. Magnetic shielding is an expensive last resort. More exotic constructions use high Ni alloy (mu-metal) as a casing but the benefit obtained from high theoretical permeability is partially lost in fabrication and assembly techniques. A greater attenuation (not elimination) of the stray field is achieved if a double-case construction is used: an inner case of magnetic nickel steel surrounded by a 2nd case of mild steel. " Handbook of Transformer Design and Applications W. Flanagan McGrawHill 2nd ed. 1993 (sec 3.7)

 

Here again my points are re-affirmed.

[MetaFrizzics]

Spontaneous Magnetostriction

 

When a material becomes ferromagnetic at the Curie point, spontaneous magnetization appears within the domains and with it an associated spontaneous strain e or magnetostriction [math]\lambda_O,[/math] along a particular direction. The amplitudes of such are independant of crystallographic direction. Within each domain the strain varies with angle from the direction of spontaneous magnetization, according to this: [math] e(\theta) = e (\cos \theta )^2 [/math]

 

The average deformation throughout due to magnetostriction can be gotten from integration (assuming randomly oriented domains):

 

[math] \lambda_O = \int^{\pi/2}_{-\pi/2} e( \cos\theta)^2\sin\thetad\theta = e/3 [/math]

 

This is caused by the ordering of the magnetic moments at onset of ferromagnetism. Because the strain is in all directions, the sample will change dimensions but stay the same shape. For details on the quantum mechanical aspects of this, see:

Introduction to Magnetism and Magnetic Materials D. Jiles 1991 Chapman & Hall. Jiles is from the Ames Lab at the U.S. Department of Energy)

 

As a practical matter, the best technology for both d.c. and low frequency magnetic shielding is still relatively simple steel alloy casings. (refs. Magnetic Circuits and Transformers ,1943 (nothing has changed) Dept EE, M.I.T.)

 

For anyone who wants to properly understand the electromagnetic field at a deeper level than the almost idiotic 'Maxwell' equations of Heaviside from the last century, A difficult to locate but valuable book is:

Antennas in Matter RWP King, GS Smith, with M Owens & Tai Tsun Wu 1981 - M.I.T. press This 850 page tome deservedly commands the title 'Advanced'.

[MetaFrizzics]

Mu-metal cannot block a permanent or D.C. magnet. All that happens is the Mu-metal becomes magnetized and hence becomes part of the magnet!

I was going to say "hogwash" !!! But I'll be polite and so, I'll merely ask you to back up your claim.

I may have exaggerated a little, but anyone who wants to play with mu-metal cages and magnets can confirm the basic observation. (As a hi-fi audio designer I have done so for many decades, and speak firsthand.) Only incredibly thick pieces of mu-metal could hope to be effective, and its a law of diminishing returns, due to skin-effects, hysteresis, magnetostriction and reduction of useful area near the magnet.

It's true that you can't use this material to shield one magnet from another, but you can shield a region of space from magnetic fields

Well, that's another one for me. In what sense can you shield 'nothing'? (a region of space)? Obviously, to talk about effective shielding you have to put something in there, and show it was shielded. Yet if we put a magnet in there, the house of cards collapses.

 

If I just admit I exaggerated a little, will you let me go now? I'm just a bit quirky in my old age.

[swansont]

Here again my points are re-affirmed.

 

Ummm,

 

Transformers are AC, not DC, so the points raised by DQW and myself are not addressed by this, and the source says that shielding works but is expensive, which would seem to contradict your point.

[swansont]

Well' date=' that's another one for me. In what sense can you shield 'nothing'? (a region of space)? Obviously, to talk about effective shielding you have to put something in there, and show it was shielded. Yet if we put a magnet in there, the house of cards collapses.[/quote']

 

In the example that's in my lab, you shield atoms in a vacuum system from having changing Zeemann shifts from the slow fluctuations in the external magnetic field of the earth.

 

The complete set of shields used in the frst apparatus in the lab had a shielding factor of about 30,000. The new set I designed should be greater than 100,000.

 

I may have exaggerated a little' date=' but anyone who wants to play with mu-metal cages and magnets can confirm the basic observation. (As a hi-fi audio designer I have done so for many decades, and speak firsthand.) Only incredibly thick pieces of mu-metal could hope to be effective, and its a law of diminishing returns, due to skin-effects, hysteresis, magnetostriction and reduction of useful area near the magnet.

[/quote']

 

 

As an atomic clock designer, I have to say that my experience contradicts yours.

[swansont]

This equation indicates that magnetostatic shielding effectiveness depends upon magnetic permeability of the shield and its radius and thickness. Even if the shield's magnetic permeability [math]\mu_s > 1[/math] and the shield thickness is small' date=' [math'] ( r_e = r_i ) [/math], the shield effectiveness is not very good. Only very thick shields with large permeability provide protection from static and low-frequency magnetic fields.

 

 

Depends on what is meant by "very thick." I have a 4-shield set. Three of them use 1/16" mumetal, and the fourth (innermost) is 1/8" thick. Currently the innermost shield is the only one in place, and is giving a shielding factor of several hundred, which is good enough for preliminary results.

 

Previous shields were 1/16" and gave the resuts I listed earlier - SF > 30,000

[DQW]

Solving jointly all of the equations with regard to the integration constants, we finally obtain:

What equations ? What integration constants (having solved enough and more Laplacians, I can guess) ??? Holy cow ! If you are going to throw equations at us to "prove" your "point" you ought to do better than using fancy words which mean absolutely NOTHING. I hope you realize this reeks of scientific disingenuity.

 

[math]S = \frac{H^I}{H_O} = \frac{4\mu_s r^2_e}{r^2_e(1 + \mu_s)^2 - r^2_i (1 - \mu_s)^2} = \frac{4}{\mu_s} \frac{1}{1-\frac{r^2_i}{r^2_e}} [/math]

I'll accept this equation without proof, for now. But next time you come up with something like this, as "proof" I'm going to insist that you show the derivation or else provide a link to the place you cut-and-pasted from. How are all those symbols supposed to make sense to someone who has no idea what they mean ?

 

Nevertheless, let's start from here :

 

 

[math]S = \frac{4\mu_s r^2_e}{r^2_e(1 + \mu_s)^2 - r^2_i (1 - \mu_s)^2} = \frac{4}{\mu_s} \frac{1}{1-\frac{r^2_i}{r^2_e}} [/math]

 

I make the substitutions, [imath]r_e \equiv r~;~~r_i = r-t~,~~t<<r [/imath] where r is the outer radius and t is the wall thickness. Then,

 

[math]{1-\frac{r^2_i}{r^2_e} = {1-\frac{(r-t)^2}{r^2} = \frac {r^2 - (r-t)^2}{r^2} = \frac {2rt - t^2}{r^2} \approx \frac{2t}{r} [/math]

 

Mumetal® can have a relative permeability of 230,000 or higher. For now, I'll use a conservative number [imath]\mu _s = 2 \cdot 10^5 [/imath]. Now if the radius of your spherical shell (or whatever the heck it is you wrote that equation for) is 10cm (about the size of a bowling ball) and has a wall thickness of 10mm, this gives you S = 1/10000 !!! Even if you had giant sphere of radius 2m (a 13 ft dia enclosure, bigger than my kitchen) with a wall thickness of only 4mm (less than a quarter inch, hardly enough for just structural strength) you have an impressive shielding ratio of S = 1/200 !

 

In other words, when you say...

Even if the shield's magnetic permeability [math]\mu_s > 1[/math] and the shield thickness is small, [math] ( r_e = r_i ) [/math], the shield effectiveness is not very good. Only very thick shields with large permeability provide protection from static and low-frequency magnetic fields.

...you are TALKING THROUGH YOUR HAT.

 

And what on earth does "Even if the shield's magnetic permeability [math]\mu_s > 1[/math]" mean ? I'll tell you what : it is misleading the casual reader into believing that you are making a rare exception by permitting the shield material to have a relative permeability greater than 1. For every metal that you can name with [imath]\mu _s < 1 [/imath], you know I can name tens of metals with [imath]\mu _s >> 1 [/imath]. You must know that diamagnetic susceptibilities are several orders of magnitude smaller than ferromagnetic susceptibilities ! Yet you make a statement like this ? You know that most shielding materials have [imath]\mu _s = 100,000 ~to~1,000,000 [/imath] or more, and you have the gall to make it look like having [imath]\mu _s > 1 [/imath] is a rarity among shielding metals ?

 

And then you go on to obfuscate the matter even more (Hoping what ? That nobody will check the details ?), rather than accept that you were wrong.

 

Mumetal is a very good DC magnetic shield, and there's NO DENYING THAT.

 

http://www.mushield.com/

http://www.goodfellow.com/csp/active/STATIC/E/Mumetal.HTML

http://farside.ph.utexas.edu/teaching/jk1/lectures/node52.html

 

"Magnetic Shielding for a Spaceborne Adiabatic Demagnetization Refrigerator (ADR)", Brent A. Warner, Peter J. Shirron, Stephen H. Castles, Aristides T. Serlemitsos, Adv. Cryo. Engg., 37, 907 (1992)

 

" An analysis of magnetic shielding against DC power lines based on homogenization", Waki, H; Igarashi, H; Honma, T, Int J for Comp. Math. in Electrical and Electronic Eng., 24, 2, 566-580 (2005)

[MetaFrizzics]

Ummm' date='

 

Transformers are AC, not DC, so the points raised by DQW and myself are not addressed by this, and the source says that shielding works but is expensive, which would seem to contradict your point.[/quote']

Actually, it seems the other way around to me:

d.c. (stationary magnetic) fields don't do any real work, since there is no motion of the 'blocker' relative to the magnet. The magnetic flux leaks and extends into nearby space waiting for some unsuspecting (moving) object made of iron like another metal bar that it can pull on or magnetize.

If you could have the mu-metal rotate around the magnet, there might be hope of demagnetizing it and weakening the field leakage.

 

I find the whole model of 'faraday' lines repulsive (no pun intended) since everyone takes them too literally, and assumes they actually exist.

 

 

I see your Link (3) also agrees with what I said, when you read the fine print at the bottom (even though their units are different than mine):

 

hus, if [math]\mu\simeq 10^5 \mu_0[/math] for Mumetal, then we can reduce the magnetic field strength inside the shell by almost a factor of 1000 using a shell whose thickness is only 1/100th of its radius. Clearly, a little Mumetal goes a long way! Note, however, that as the external field strength, [math]B_0[/math], is increased, the Mumetal shell eventually saturates, and [math]\mu/\mu_0[/math] gradually falls to unity. Thus, extremely strong magnetic fields (typically, [math]B_0\stackrel {_{\normalsize >}}{_{\normalsize\sim}}1[/math] tesla) are hardly shielded at all by Mumetal, or similar magnetic materials.

That is, just like I said, mu-metal quickly fails to shield anything when you are trying to stop the close proximity field of an ordinary magnet.

 

Your first Link (1) was cute by the way. The effort (and cost!) that goes into shielding a computer monitor inside an NMR medical scanning room is a great example of why it is probably more sensible now to replace a CRT with a liquid crystal display! (another obselete product...) But can only imagine the engineering efforts that went into that shielding project, that caused them to turn to mu-metal as the best 'kludge'. And here the magnetic fields are well below safety standards for human beings for long-term exposure. NOT the close-proximity fields of permanent magnets! The same goes for the other examples there: Shielding the earth's magnetic field? Come on: It's so weak you have to use a magnetic compass to detect it! All the examples given where mu-metal shielding is effective happen to be with fields so weak you need special equipment just to measure them accurately!

 

The other examples on the website page are equally complex and incredibly expensive and difficult shielding projects, and while 'mu-metal' is inevitably involved, I'd suspect a far more important factor would be the creative engineering skills of the scientists overseeing the projects! (as I pointed out in my previous posts describing the problem of shielding.)

 

You know that most shielding materials have \\mu _s = 100,000 ~to~1,000,000 or more, and you have the gall to make it look like having \\mu _s > 1 is a rarity among shielding metals ?

Whoa. Let's take it back a few notches: The point wasn't to speak of permeability per se, but to show that a second factor, thickness of the shield was far more significant relatively speaking for shield effectiveness.

 

This is clearly indicated right in the bit you quoted: "...and the shield thickness is small...only very thick shields...", but is even more clearly indicated in the equations themselves.

 

Mumetal® can have a relative permeability of 230,000 or higher. For now, I'll use a conservative number \\mu _s = 2 \\cdot 10^5. Now if the radius of your spherical shell (or whatever the heck it is you wrote that equation for) is 10cm (about the size of a bowling ball) and has a wall thickness of 10mm, this gives you S = 1/10000 !!! Even if you had giant sphere of radius 2m (a 13 ft dia enclosure, bigger than my kitchen) with a wall thickness of only 4mm (less than a quarter inch, hardly enough for just structural strength) you have an impressive shielding ratio of S = 1/200 !

This is good work: But look at what you are doing as well, over and over again. You are talking about distances ranging from a bowling ball radius to a 20 foot square room. In these cases, supposing the placement of a strong permanent magnet the size of a golf-ball (no lightweight item), you are dealing with shielding field-strengths [math] \frac{1}{10^{10}} [/math] of the strength at the source.

 

Of course if you are either shielding incredibly weak fields (like the earth's) or working at gargantuan distances from the source (10 feet from magnet) mu-metal is your man!

 

Now look at the original post: He wants to completely shield a permanent magnet less than a mm from its core on the axis of the field lines, from end to end.

 

In a word, "impossible" (at least with a mu-metal shield).

[swansont]

(you seem to have combined your responses to both me and DQW. No matter)

 

"Clearly' date=' a little Mumetal goes a long way! Note, however, that as the external field strength, B[sub']0[/sub], is increased, the Mumetal shell eventually saturates, and [math]\mu/\mu_0[/math] gradually falls to unity. Thus, extremely strong magnetic fields (typically, B₀ >1 tesla) are hardly shielded at all by Mumetal, or similar magnetic materials."

 

That is, just like I said, mu-metal quickly fails to shield anything when you are trying to stop the close proximity field of an ordinary magnet.

 

1 tesla isn't an ordinary magnet, so one shouldn't pretend that this is typical. The cite is pretty clear - mumetal fails at large fields because it saturates. In those cases you use a material with a smaller permeability.

 

...

And here the magnetic fields are well below safety standards for human beings for long-term exposure. NOT the close-proximity fields of permanent magnets! The same goes for the other examples there: Shielding the earth's magnetic field? Come on: It's so weak you have to use a magnetic compass to detect it! All the examples given where mu-metal shielding is effective happen to be with fields so weak you need special equipment just to measure them accurately!

 

Nevertheless' date=' there are many research and commercial applications where it is necessary. Not because of human safety standards but because the device won't work with random stray fields.

 

 

Now look at the original post: He wants to completely shield a permanent magnet less than a mm from its core on the axis of the field lines, from end to end.

 

In a word, "impossible" (at least with a mu-metal shield).

 

 

I don't see a magnet strength nor a distance scale on the drawings.

 

Take a look at this device, used in optics labs. In the "on" position, it sticks to the optics table, and quite well. In the "off" position, it lifts off with almost no effort. How does it work? You rotate the maget so there is really good flux return in the "off" position - i.e. it is shielded by a material with a decent-sized permeability. In the "on" postion the field lines up with gaps in the shielding, and the field can leak out easily, and so can be attracted to the table. I've seen similarly-designed devices for use in machine shops.

 

Face it - magnetic shielding works fine for a wide range of applications. Not all. But hardly the "never works" scenario you were describing.

[DQW]

1. Typical bar magnets have a field strengths of 0.05 to 0.2T. NdFeB and SmCo5 type supermagnets are the only magnets you will find that have field strengths of order 1T (only at the magnet poles - the field decays outside the magnet, and the field along the sides of the magnet is much smaller than at the poles).

 

2. Mumetal saturates at about 0.8T. There are other alloys that saturate much higher, which have permeabilities in the many thousands.

 

3. Unless you have a VERY powerful magnet (> 3 tesla pole strength!!!), mumetal will achieve better than 90% (and in most cases, I guess better than 99%) shielding for the geometry described in the OP - at a spot that's half a magnet length away, along the equatorial direction. This is my opinion.

 

4. We have come a long way, from saying that DC magnetic shielding is a myth to making it look like it was meant all along, that you were talking about DC fields much in excess of 1T.

 

5. This statement : "In these cases, supposing the placement of a strong permanent magnet the size of a golf-ball (no lightweight item), you are dealing with shielding field-strengths [imath]\frac{1}{10^{10}}[/imath] of the strength at the source" shows that you do not understand the equation that you began an earlier post with. In that equation, Bo is not the magnetic field at the source. It is the field at the surface of the shield. The result quite clearly shows that the shielding efficiency is independent of the distance or size of the magnet (as long as Bo < B(sat)).

[Bluenoise]

Now most cause rifles use a coil of wire and send a projectile through it. But they seriously lack efficiency and power. I wonder if there is a way in wich you could create the same effect from a Halbach arrays by winding a wire. Seems you could get a much more efficient gauss rifle out of that. Though you'd have to reverse the polarities of the magnets to get the field to poin the projectile in the right direction.

[MetaFrizzics]

Why don't we just meet half-way, and say the following:

 

(1) Strong magnets (> 1 T) are difficult to shield at close range.

 

(2) High permeability means quick saturation (@ < 1 T), which is a trade-off.

 

(3) Low strength fields are easy to shield.

 

(4) High permeability Mu-metal is ideal for low strength fields (< 1 T).

 

---------------------------------------------

At this point, it seems to me we are all being equally disengenious at the least:

 

In the example that's in my lab, you shield atoms in a vacuum system from having changing Zeemann shifts from the slow fluctuations in the external magnetic field of the earth.

Again, ridiculously low strength fields.

The complete set of shields used in the frst apparatus in the lab had a shielding factor of about 30,000. The new set I designed should be greater than 100,000.

Yeah, until you lean a 2T magnet against it.

 

As an atomic clock designer, I have to say that my experience contradicts yours.

I don't see how that can be, since we use the same equations and electromagnetic theory.

 

1 tesla isn't an ordinary magnet, so one shouldn't pretend that this is typical.

Every magnet I have on my shelf here, taken from ordinary modern industrial motors is at least 1 T, and many are almost 2 T. They don't make good motors with bad magnets anymore. Sure these magnets are 'strong'. But they are also typical surplus nowadays. Who's going to buy a crappy magnet when there's a bin full of killer magnets over here for 50 cents each?

 

I don't see a magnet strength nor a distance scale on the drawings.

Its hard to get more disengenious than that. Is the OP going to build a magnetic flytrap? Surely he is hoping to build something significant and useful.

 

Take a look at this device, used in optics labs.

I have to admit those are cool. They didn't have them back when I was auditing the first holography courses at M.I.T. in /70-72, and they were virtually inventing everything on the fly. I used sand and heavy blocks in one nice experiment with an early laser.

Face it - magnetic shielding works fine for a wide range of applications.

...with a *narrow* range of T.

 

Not al(ways). But hardly the "never works" scenario you were describing.

True enough. I admit being facetious at least once. Will you?

NdFeB and SmCo5 type supermagnets are the only magnets you will find that have field strengths of order 1T

...and find them you will by the thousands. In calling these now common motor magnets 'supermagnets' you can only be a referencing to the 60's, when there were no drugs, only airplane glue and Beatle-boots.

Mumetal saturates at about 0.8T.

Thank you. Useful and accurate.

 

mumetal will achieve better than 90% (and in most cases, I guess better than 99%) shielding for the geometry described in the OP - at a spot that's half a magnet length away, along the equatorial direction. This is my opinion.

And your opinion is pretty safe, with that qualifier. Why even bother to shield a magnet normal to it's poles half a length away? Because it's easy?

 

making it look like it was meant all along, that you were talking about DC fields much in excess of 1T.

As far as I know we were. Do you think the OP was hoping to shield a toy bar-magnet in his grade 7 class?

Perhaps this is how we ended up on opposite sides in the first place. If it is a misunderstanding based upon coming at the problem from opposite ends, I can totally understand that. You guys are coming from the standard applications such as shielding sensitive equipment from very low field fluctuations, which is actually a practical use of mu-metal. I was coming from the other end of the diagram wondering how anyone is going to shield a hefty magnet. Let's meet in the middle, I say again.

 

But this???:

shows that you do not understand the equation that you began an earlier post with.

Ouch, bum sore now. Please. I was doing us all a favour by not posting a 10 page derivation. I think I know what I meant by my own diagrams and equations. I'm happy to post the derivation if you give me two days to set up all the LateX by hand. I have no auto-program, and I've been reverse-engineering everybody else's LateX posts up until now.

 

Thanks for the scolding. But since I can pull the equation out of the sky, I can also derive it, although perhaps not as rigorously as a mathematics specialist and author with 5 T.A.s doing most of the slog-work.

[MetaFrizzics]

Okay here was my starting point:

You have a wire with a mu-metal jacket (he he!)

There are basically three areas of interest as far as I can see,

Region C is of course the key area to be shielded.

 

I will slip on some LateX shortly.

 

OMG he's wearing a dental barrier!

[swansont]

It's going to be more effective if the tube is oriented in the direction of the field.