2 Neo magnets with some Mumetal shielding

[gareth01422]

Hi guys

I have a simple question.

The attached picture shows 2 Neo magnets with some Mumetal sheilding between the magnets. Let's say the gap between the shield and the magnets is 2mm. So the overall distance between the magnets is 4mm.

will the magnets still repel from the shielding? If so will this force be the same as if the shielding wasnt there?

Gareth

 

[marcusl]

Yes, in this configuration they should still repel. You put your "shield" at a nodal plane (where the field is everywhere zero) so it has no effect whatsoever.

 

[gareth01422]

Ok I read something like that on a site, but didnt fully understand it. Is there material that would stop the interaction between the magnets and not replel the magnets.

Or is there a way of neutralizing the magnets say for 1 second? could this be done with an electrical current?

Gareth

 

[Neo_Anderson]

Mu-metal shields are among the most effective magnetic shields available. Magnetic shielding is always a function of the thickness of the mu-metal.

If a 1-cm circular disk magnet provides 200K Gauss of magnetizm, and if a 1"x1"x1/8" piece of mu-metal provides 400K of shielding, then the magnetic field is completely contained by the mu-metal. Proof: Pry open an old harddrive. Take out the neodymium magnet that's stuck to its mu-metal plate. Now get a 1cm circular neodymium magnet and place it on the mu-metal. Take a compass and place it at the other end of the mu-metal.

The compass needle would not deflect at all.

 

[gareth01422]

Mu-metal shields are among the most effective magnetic shields available. Magnetic shielding is always a function of the thickness of the mu-metal.

If a 1-cm circular disk magnet provides 200K Gauss of magnetizm, and if a 1"x1"x1/8" piece of mu-metal provides 400K of shielding, then the magnetic field is completely contained by the mu-metal. Proof: Pry open an old harddrive. Take out the neodymium magnet that's stuck to its mu-metal plate. Now get a 1cm circular neodymium magnet and place it on the mu-metal. Take a compass and place it at the other end of the mu-metal.

The compass needle would not deflect at all.

Great , that has clarified it for me.

So If I have it correct, The shielding has a K factor (400K in the example you said). So am i correct in saying that if i had a piece that was 1/4" thick this would be 800K of sheilding? and as in your example, if i places a magnet on one side, will the megnet attract or repel from the sheilding?

Gareth

 

[Neo_Anderson]

Great , that has clarified it for me.

So If I have it correct, The shielding has a K factor (400K in the example you said). So am i correct in saying that if i had a piece that was 1/4" thick this would be 800K of sheilding? and as in your example, if i places a magnet on one side, will the megnet attract or repel from the sheilding?

Gareth

The magnet will not be attracted to the mu-metal, because mu-metal is made with non-magnetic metals (nickel, copper, etc). This is why your neodymium ("neo") magnet slides right off the mu-metal.

Unlike your mu-metal, a metal/alloy such as steel or iron will attract a magnet.

If your mu-metal is thick enough, then it will provide a circuit for all stray magnetic lines of force. This means that if you have two neo magnets repelling each other (north-to-north or south-to-south orientation) and if you place the mu-metal in-between the magnets, then there will be no further magnetic repelling. Ditto for a north-south attraction between your neo magnets: place a mu-metal barrier of sufficient permeability between the two, and the magnets will no longer attract each other.

The "K-factor" is magnetic permeability. For mu-metal, the permeability is quite high, in the order of 20K or so. Metglas Magnetic Alloy goes as high as 1 mln, but is quite expensive. Look at the spreadsheet below for more metals, and their permeabilities:

http://en.wikipedia.org/wiki/Permeability_(electromagnetism )

 

[gareth01422]

The magnet will not be attracted to the mu-metal, because mu-metal is made with non-magnetic metals (nickel, copper, etc). This is why your neodymium ("neo") magnet slides right off the mu-metal.

Unlike your mu-metal, a metal/alloy such as steel or iron will attract a magnet.

If your mu-metal is thick enough, then it will provide a circuit for all stray magnetic lines of force. This means that if you have two neo magnets repelling each other (north-to-north or south-to-south orientation) and if you place the mu-metal in-between the magnets, then there will be no further magnetic repelling. Ditto for a north-south attraction between your neo magnets: place a mu-metal barrier of sufficient permeability between the two, and the magnets will no longer attract each other.

Thats great

Ok, I am going to carry out an experiment when I have the money. This is what I would like to achive so please bare with me.

I will have one fixed Neo magnet and one sliding back and forth like and engines piston, At top dead centre is when the magnets will be closest (the closer the better). As the magnet travels back to the fixed piston I would like to shield the gap. then once at top dead center the shield will flick out of the way, applying a repelling force pushing the magnet back down.

The magnets I would like to use are 3" DIA x 1" Thick N42 Neo magnets. The gauss at this distance between the magnets is 3258 http://www.kjmagnetics.com/calculator.repel.asp" or so the site says in the link. I don't trust these figure 100% but its a place to start.

Will a Mu metal sheild do the job I am after?

Gareth

 

[Neo_Anderson]

Thats great

Ok, I am going to carry out an experiment when I have the money. This is what I would like to achive so please bare with me.

I will have one fixed Neo magnet and one sliding back and forth like and engines piston, At top dead centre is when the magnets will be closest (the closer the better). As the magnet travels back to the fixed piston I would like to shield the gap. then once at top dead center the shield will flick out of the way, applying a repelling force pushing the magnet back down.

The magnets I would like to use are 3" DIA x 1" Thick N42 Neo magnets. The gauss at this distance between the magnets is 3258 http://www.kjmagnetics.com/calculator.repel.asp" or so the site says in the link. I don't trust these figure 100% but its a place to start.

Will a Mu metal sheild do the job I am after?

Gareth

Lol, I get it! The moveable magnet is like an internal combustion engine's piston, and the fixed magnet is like the stratefied charge of compressed gasoline vapors, and the mu-metal being flung out of the way is like an intake valve closing!
Lol.
Fellow forum members, why won't it work?
I know why; I just want to see if anyone else does, that's all...

 

[gareth01422]

Lol, I get it! The moveable magnet is like an internal combustion engine's piston, and the fixed magnet is like the stratefied charge of compressed gasoline vapors, and the mu-metal being flung out of the way is like an intake valve closing!
Lol.
Fellow forum members, why won't it work?
I know why; I just want to see if anyone else does, that's all...

PM the answer if you will, because I havnt a clue why? lol

 

[marcusl]

Wait wait wait! I'm afraid almost everything contained in Neo_Anderson's posts is incorrect. Here are some of the wrongest:
1) Of course mu-metal is magnetic. A magnet will stick to it quite nicely.
2) It reaches high permeability only under very specific conditions at low fields. How did N_A get 400k, by the way? He just made this up.
3) Mu-metal is not a good choice to shield strong fields because it saturates at low applied fields. Paradoxically, iron or low-carbon ("mild") steel is a better choice for intense fields.
4) Finding thick mu-metal is expensive. It is usually available as foils and thin sheets. By the way, it needs special annealing and heat treatment in a hydrogen atmosphere for full efficiency, so machining and forming is usually done prior to this processing. Even the act of cutting a finished sheet with a shears will partially degrade its shielding properties.
5) Placing a shield between the magnets as you have drawn will be ineffective. The magnetic field on the plane halfway between your magnets is already zero.

 

[gareth01422]

Wait wait wait! I'm afraid almost everything contained in Neo_Anderson's posts is incorrect. Here are some of the wrongest:
1) Of course mu-metal is magnetic. A magnet will stick to it quite nicely.
2) It reaches high permeability only under very specific conditions at low fields. How did N_A get 400k, by the way? He just made this up.
3) Mu-metal is not a good choice to shield strong fields because it saturates at low applied fields. Paradoxically, iron or low-carbon ("mild") steel is a better choice for intense fields.
4) Finding thick mu-metal is expensive. It is usually available as foils and thin sheets. By the way, it needs special annealing and heat treatment in a hydrogen atmosphere for full efficiency, so machining and forming is usually done prior to this processing. Even the act of cutting a finished sheet with a shears will partially degrade its shielding properties.
5) Placing a shield between the magnets as you have drawn will be ineffective. The magnetic field on the plane halfway between your magnets is already zero.

so what would you suggest?

Is there other materials that would do the job?

Gareth

 

[Bob S]

Yes, in this configuration they should still repel. You put your "shield" at a nodal plane (where the field is everywhere zero) so it has no effect whatsoever.

In the midplane between the two magnets, the axial field is zero. But the radial field in the shield from the two magnets adds, and is very large. Mumetal is a very good magnetic shielding in certain applications where very low fields (< 1 Gauss) are required, but mumetal saturates at about 0.2 to 0.4 Tesla, while soft iron saturates at about 1.4 Tesla, so soft iron is the better choice in this application. The magnetic field from each magnet pole enters normally (along the axis) into the soft iron (remember B_normal is continuous), travels outwards in the iron radially about 2”, and exits. This means that the total soft iron cross section should be about equal to about half the combined area A of the two magnets at the magnet radius: So
A = 2*pi*R*t where t is the thickness of the soft iron and R is the magnet radius. So we get for t (for R=1.5"):

t = A/(2*pi*R) = pi*R²/(2*pi*R) = R/2 = 0.75”

This will keep the peak radial field everywhere in the iron below 1 Tesla. When the soft iron shield is pulled out or pushed in, eddy currents in the soft iron will resist the force.

Bob S

 

[Bob S]

Two items:
1) Magnetic forces from PM magnets are conservative forces, meaning that a closed-loop path integral of ∫F·dξ over a path (or series of actions) where F is a force and dξ is a displacement, will not lead to any excess useful energy. The loop integral is zero. Perpetual motion as a source of energy is simply not true. More likely is loss of useful energy due to eddy current heating.

2) The stored magnetic energy in the system is the volume integral

W = [1/2]∫B·H dV = [1/2μμ₀]∫B² dV_volume

where μ is the relative permeability which = 1 everywhere except inside magnetic materials, such as the soft iron shield. When the soft iron shield is inserted, the total stored magnetic energy in the system decreases due to the B field inside the soft iron ( μ ≈2500).

The change in the stored magnetic energy W is directly related to a force:

dW/dx = F_x

So as the soft iron shield is inserted, it is pulled in by this force. If only one magnet is present, the force is halved. If both magnets are present when the soft iron is removed, the force (work required) to pull it out is doubled.

If the magnetic shield were a Type I superconductor that excluded all magnetic fields (Meissner effect), work will be required to push the shield in (because the stored energy W increases), and useful work could be recovered when the shield is removed.

Bob S

 

[gareth01422]

so the idea just won't work at all?

The shield will have to be too thick, the sheild wil have to be a superconductor, and the work needed to move the shield will be wasted.

Gareth

 

[Neo_Anderson]

Wait wait wait! I'm afraid almost everything contained in Neo_Anderson's posts is incorrect. Here are some of the wrongest:
1) Of course mu-metal is magnetic. A magnet will stick to it quite nicely.
2) It reaches high permeability only under very specific conditions at low fields. How did N_A get 400k, by the way? He just made this up.
3) Mu-metal is not a good choice to shield strong fields because it saturates at low applied fields. Paradoxically, iron or low-carbon ("mild") steel is a better choice for intense fields.
4) Finding thick mu-metal is expensive. It is usually available as foils and thin sheets. By the way, it needs special annealing and heat treatment in a hydrogen atmosphere for full efficiency, so machining and forming is usually done prior to this processing. Even the act of cutting a finished sheet with a shears will partially degrade its shielding properties.
5) Placing a shield between the magnets as you have drawn will be ineffective. The magnetic field on the plane halfway between your magnets is already zero.

I must apologize to marcusl, to gareth01422, and to the forum.
After reading this last night, I took a neo magnet/mu-metal sub-assembly I pulled from a harddrive and removed the magnet. And found that mu-metal is indeed a magnetic material, contrary to what I stated in my earlier posting. "Who in their right mind would consider nickel a magnetic material," I thought, "In light of the fact that a 5-cent nickel does not adhere at all to a magnet; even ones as powerful as neodymium magnets!"

I was incorrect, and did not approach the issue in a scientifically methodological manner. Trying to obtain a magnetic property from a nickel is hardly the Scientific Method! (although I did consider the high copper content in a nickel...)

Marcusl must admit his own shortcoming; that being that mu-metal is quite magnetically permeable even at freq's as high as 100KHz. And, unusually resistant to eddy currents and other factors that govern thermal activity in the alloy. Aside from mu-metal's magnetic culpricicity (it's a word), it functions as a near-ideal magnetic shield!

 

[Neo_Anderson]

But gareth's engine can still work, but only temporarily, if he uses mu-metal (or even metglas 2714A) to shield one pole from its like pole.

To put things into perspective, this is the engine gareth proposes: a magnetically-driven device that depends on the opposition of magnetic forces for mechanical motion.
In the customary internal combustion engine ("ICE"), there are the following components:

• Intake valve and its drivetrain (camshaft).
• Piston, which is necessarily connected to a connecting rod; the other end of which is connected to a driveshaft.
• A stratified charge of gasoline vapors, which drive the piston downwards.

Gareth's engine is similar, with the following analogues:

• A fixed, permanent magnet takes the place of the stratefied charge in the ICE.
• A high-permeable magnetic shield takes the place of the intake valve.
• A moveable magnet, necessarily attached to a con-rod of some type takes the place of the piston/con-rod/driveshaft in the ICE. Gareth's design must include a driveshaft-like component.

I insist it cannot be done on the basis that such a device would constitute a perpetual motion device, which is forbidden under Newton's mechanics (conservation laws). It cannot be done, because the rate at which the magnetic shield is "flung away" from the moveable magnet at TDC must be faster than the rate at which the moveable magnet is driven away (lit. "downwards") from the fixed member.

The energy taken to move the magnetic shield away from the moveable magnet at such a high speed would be greater than any energy gained from the opposition of magnetic forces when the moveable magnet is at TDC.
We cannot consider applied materials as a limitation, either, since magnetism is accomplished on a molecular, crystalline scale, with the heavier magnetic crystalline structure the better magnetic crystalline structure.

Can anyone else point out a different limitation to the design from a technical standpoint?

 

[gareth01422]

Hi guys

Thanks for the answers to my questions. I understand now that using magnetic shielding isn't such a good idea. So I've been having a re think on the idea. In stead of having the stationary magnet at the top, What about an electromagnet to perform the same task but only on the down stroke, which could work off a micro switch to turn the power on and off. I have never made or used electromagnets so I'm not too sure how they work (only from what I've read on the net).

Gareth

 

[marcusl]

Yes, this is a better approach. Use an electromagnet and substitute a steel rod for the magnet. When you energize the coil it will pull the rod into its center, also pulling along whatever load you want to connect to it. A coil spring can push it back when you de-energize the coil.

This is called a solenoid and is used, e.g., to turn the water on and off in dishwashers and washing machines.

 

[gareth01422]

Yes, this is a better approach. Use an electromagnet and substitute a steel rod for the magnet. When you energize the coil it will pull the rod into its center, also pulling along whatever load you want to connect to it. A coil spring can push it back when you de-energize the coil.

This is called a solenoid and is used, e.g., to turn the water on and off in dishwashers and washing machines.

Ok - I don't think I will need a spring as this idea could have a few cyclinders, But would the coil of the solenoid shut down and energize fast enough? All I want to do is replace the stationary magnet with the solenoid design then I don't have to worry about shielding. And would it possible to make a solenoid about the same dimensions as the magnets (3"DIA x 1" or 2") that will produce enough energy as the magnet it is replacing?

Gareth

 

[marcusl]

What are you trying to do?

 

[gareth01422]

What are you trying to do?

PM sent