hoverboard

How much copper will it take to pave all the roads in the USA?

But then it would be cool to have traffic jams made of of cars that can't steer.

Wheels are not a problem that needs to be fixed.

I have noticed the stripe pattern, so I assumed that the sticky fields were along the plane of the fridge door, not perpendicular, which would have a huge air gap.

I just tried flipping one over, and there is zero attractive force from the flip side. So the Halbach thing is real.

They were used as wiggler magnets at the FEL where I worked for a while.

There is a Halbach cylinder too.

That gives you a constant field inside. A colleague wants to build two of them, one outside the other. Then spinning one wrt the other gives a variable internal field. It would be kinda fun.. in magnetic materials weren't so dang expensive.

I didn't realize that Halbach was a contemporary.

George H.

"George Herold" wrote in message news: snipped-for-privacy@googlegroups.com...

On Monday, October 26, 2015 at 11:53:39 AM UTC-4, John Larkin wrote:

They were used as wiggler magnets at the FEL where I worked for a while.

There is a Halbach cylinder too.

That gives you a constant field inside. A colleague wants to build two of them, one outside the other. Then spinning one wrt the other gives a variable internal field. It would be kinda fun.. in magnetic materials weren't so dang expensive.

I didn't realize that Halbach was a contemporary.

George H. ========================================================

One of the big permanent magnet sellers (forget which one) sells nested Halbach spheres so you can vary the internal field from zero to about 2 T, I think. The problem with a sphere or cylinder is the shape and varying direction of magnetization of the pieces. The shapes are all the same, just custom for you particular dimensions, but they need a different set of magnetizing pole pieces for each different direction of magnetization so that gets expensive fast, especially if you want 8 or more segments for best homogeneous volume. I found an article online that used four flat rectangular sections to give a square cross section cavity, with better homogeneity and strength than just the usual two (so non-Halbach) pole pieces, but you still need two different directions of magnetization. I got curious and spent some time playing with FEMM for fun, and it looks like you can do a decent cylindrical Halbach array using just square sections. Picture bars say 1/2" by 1/2" by 3" long, magnetized with north pole perpendicular to one of the 1/2" by 3" faces. You can use 8 of these to form a cylinder, and get even better strength and homogeneity over a larger volume than the four pole version, and do it with standard catalog magnets. I started with the centers of the segments on a circle, but found that if you then tweak it slightly by moving the side poles a bit you get even better field shape. Anyway, you could do a pair of these, nested, to play with much cheaper than ordering two custom "real" cylindrical Halbach magnets. If you want bigger pieces you can build up say 1"x1" with four

1/2"x1/2" bars, and still wind up cheaper than having them custom made, although I bet just having them magnetized in this direction rather than the most common axial direction wouldn't add much cost given the simple shape. Using N52 material in FEMM it looked like fields up to maybe 0.5 T with a clear bore of 3" was possible, but that's from slightly fuzzy memory and I don't have any way of measuring field and homogeneity so I never built one to test. If your friend does, I's love to hear about it.

----- Regards, Carl Ijames

Den mandag den 26. oktober 2015 kl. 18.39.36 UTC+1 skrev Carl Ijames:

We got a 0.5T ~2" bore one a few years ago, I think it was at least $10k and weights roughly 70kg, (the actual bore might be bigger but the 2" was for the specified homogeneity)

I know ~1.7T is doable with ~1" bore, but even with high temp material 50'C is pushing it

-Lasse

Den mandag den 26. oktober 2015 kl. 18.39.36 UTC+1 skrev Carl Ijames:

We got a 0.5T ~2" bore one a few years ago, I think it was at least $10k and weights roughly 70kg, (the actual bore might be bigger but the 2" was for the specified homogeneity)

I know ~1.7T is doable with ~1" bore, but even with high temp material 50'C is pushing it

-Lasse ==========================================================================================

Yes, but that is with optimally shaped pieces. Using my suggestion for the

2" bore the performance is a bit less, but the cost would be about $1.5-2k for a 6" long cylinder. It all depends on what you need, and what you can live without, like any engineering design :-). I just thought this was a useful way to save a lot of $ for not much loss in performance. Oh, I guess I forgot to specify, I'm talking about a cylinder with the magnetic field perpendicular to the axis of the cylinder, not parallel to the axis.

----- Regards, Carl Ijames

Not just steer, also propel.

Yeah, that's the big flaw. Rollerskates and ice skates, and XC skis, can be directed and one can apply thrust by lateral forces on the edges/wheels. That lateral force is unavailable to the ground-interaction of hovercraft or maglev or canal barge, you need the fan/eddy-current-phased-drive/rudder-mule-and-towrope to propel your craft.

These 'hoverboard' gizmos are like undirected rafts or non-dirigible balloons. The principle, though, has good scaling properties for high-speed trains (large enough to have superconducting magnets, fast enough to hover on relatively thin aluminum). The conductive roadbed thickness only needs to be skin depth at the polarity reversal frequency.

Braking will always be a problem (if there's an emergency, the motor-reverse type of brake is not necessarily reliable); maybe build in some drogue parachutes?

Hi Carl,

If you remember who that was I would be interested.

The problem with a sphere or cylinder is the shape and varying

We sell an NMR permanent magnet that is two donuts of magnetic material with an iron yoke around the outside. But (of course) you can't vary the field. (well only a little with temperature.) Homogeneity is not bad. I wonder if a cylinder thing would do better? I got

I've basically said all I know about them, I think he was using an eight piece cylinder. And he said the outer one had to be a lot bigger, to get the two fields to cancel when the two cylinders are in opposition, (so to speak). I think this is an idea looking for an experiment. At the moment I don't think there is any plan to build one. A variable DC magnetic field from 0-1 Tesla for less than ~$5-10 k might be nice. (The $5-10 k is a made up number. Just a guess at what resistive magnet and power supply would cost.)

George H.

Lasse, is that a variable double Halbach cylinder type thing? Or just a single fixed B-field cylinder? And if a two cylinder can you tell me where you bought it? Custom?

George H.

just a fixed B-field cylinder

-Lasse

Downhill skis can steer too! Usually.

That lateral force is unavailable to the ground-interaction

I guess even a simple magnet will lift a train if it's going fast enough.

Fun: drop a good rare-earth magnet down a copper pipe.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

The serious NMR guys use a superconductive magnet that has several superconductive field shim coils. Then a lot, like 24 maybe, room temp shim coils, each with its own supply. Resonance linewidth is limited by field uniformity, and they get Qs above 1e9.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Den mandag den 26. oktober 2015 kl. 21.20.17 UTC+1 skrev John Larkin:

superconducting is really the only way to get really high fields, and the SNR increase by power of 3/2 with the field strength

but then you have the mess of needing liquid helium

-Lasse

"George Herold" wrote in message news: snipped-for-privacy@googlegroups.com... [big snip] We sell an NMR permanent magnet that is two donuts of magnetic material with an iron yoke around the outside. But (of course) you can't vary the field. (well only a little with temperature.) Homogeneity is not bad. I wonder if a cylinder thing would do better? =================================================================================

I thought all of the benchtop NMR's, including yours, were based on Halbach array magnets since my understanding is that you can get the same homogenous volume with less material using a Halbach array compared to a yoked dipole like you describe, and the price of a magnet is basically proportional to pounds of material. I'm no expert, I've just picked up some by osmosis from a previous boss who was a physicist and knew a good bit about them, and some reading, while all of my hand's on experience has been with commercial solenoidal supercons. I thought it was Dexter Magnetics that sold the spheres, but a search of their web site didn't turn anything up. Maybe they only made a few as an attention getter, and when they ran out they never made any more? :-) There's a guy named Richard Stelter at Dexter who has many patents on arrays similar to what I described, but mostly adding iron pole pieces to get higher fields in smaller volumes. I played some with similar arrangements but since I was mostly curious about scaling and large field volumes I stuck to simple geometries that didn't need custom magnets. It was just a home hobby research project, for fun and to learn how to use FEMM. Contact Dexter and see what they say about your current design and what they can do for you, maybe you can cut your costs or improve your specs. I know you aren't competing in the analytical instrumentation field like Bruker and Thermo, but getting a quote won't cost you anything :-).

----- Regards, Carl Ijames

I guess that's why we didn't invent airplanes.

--

Rick

Wow, Thanks Carl... The NMR is not my baby, so I only have a periphery view. I helped with the thermal design of the magnets. The magnets are made and mostly designed by a company in California, (I think.) There's basically zero chance of a new magnet design. Well for 5-10 years anyway. (There are x,y,z and z^2 gradient coils on the poles... old technology.)

But given that, I didn't know Bruker made a table top NMR with permanent magnets. I wonder what sort of field homogeneity they get? In theory all magnetic materials and coils are uniform and perfectly aligned. In practice it's not so good...

Hey, how about a HB cylinder with X rods of magnetic material. (X = 6 or 8 looks right.) (Magnetization across one diameter of the rod, for those following along at home.) You then need the mechanical "jig" to twist the rods to the right angle, tweak "some" positions the right way* and away you go....

George H.

• getting the mechanical design right.. which "positions" to tweak, is a huge part of the "art" in instruments. (IMHO) I've got a strong desire to try and count the degree's of freedom I need and then try and put in "just" that many knobs. (I'm thinking of more than HB cylinders, three legged tables, for instance, kinematic mounts in optics. I can't help but think of RV Jones,
  formatting link
  He was a master at mechanical stuff. Any other good books/ authors?

Wow, Thanks Carl... The NMR is not my baby, so I only have a periphery view. I helped with the thermal design of the magnets. The magnets are made and mostly designed by a company in California, (I think.) There's basically zero chance of a new magnet design. Well for 5-10 years anyway. (There are x,y,z and z^2 gradient coils on the poles... old technology.)

But given that, I didn't know Bruker made a table top NMR with permanent magnets. I wonder what sort of field homogeneity they get? In theory all magnetic materials and coils are uniform and perfectly aligned. In practice it's not so good...

Hey, how about a HB cylinder with X rods of magnetic material. (X = 6 or 8 looks right.) (Magnetization across one diameter of the rod, for those following along at home.) You then need the mechanical "jig" to twist the rods to the right angle, tweak "some" positions the right way* and away you go....

George H.

• getting the mechanical design right.. which "positions" to tweak, is a huge part of the "art" in instruments. (IMHO) I've got a strong desire to try and count the degree's of freedom I need and then try and put in "just" that many knobs. (I'm thinking of more than HB cylinders, three legged tables, for instance, kinematic mounts in optics. I can't help but think of RV Jones,
  formatting link
  He was a master at mechanical stuff. Any other good books/ authors? =========================================================================================

Here's an article from CEN that shows the big and little ends of NMR:

When I was first looking into using rods instead of segments of a cylinder to make a Halbach cylinder (in FEMM), I started with round rods and planned to mount them at the appropriate angles. Then I quickly realized that with up to 8 rods you only need orientations of 0, 90, 180, and 270 degrees anyway so square cross-sections work and are both cheaper and give you a stronger field. I think the best benchtop NMRs have minimum linewidths of less than

10 Hz, and I think that that means a homogeneity between 10 and 100 ppm. I don't know how many shim coils that takes but I would guess at least 5. I figured that with whatever your sales volume and the NRE to get the magnet designed and the probe to go with it, you probably couldn't change to a new one anytime soon, but my attitude is that it never hurts to look and keep up with the field :-).

Lasse, if you don't mind my asking, what is that high homogeneity magnet for? I was pretending to design magnets possibly suitable for moderately large (1-2") trapped ion cells for Fourier transform ion cyclotron resonance mass spectrometry, with a clear bore large enough for the cell and surrounding vacuum chamber - I keep wanting to build one in my garage :-).

----- Regards, Carl Ijames

OK thanks, I couldn't find any homogeneity numbers there. I did see some Bruker page that said 1ppm w/ shim coils. I did find this,

"Bruker says it is developing a line of Fourier-transform-based benchtop NMRs, but they are not yet ready for the market. A spokesman says that "we are interested in new markets and applications." "

We can "just do" Fourier transform spectroscopy with our system.. FID decay times in the 30ms to 100 mS range (IIRC) 20 MHz resonance.

The one thing about your square pieces is that you never get the magnetization to be perfectly aligned. (As Lasse said, I think, 1% is typical) A cylinder might be better, cause you could tweak the angle a bit.

George H.