PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2019-06-17, 02:02:48
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: [1]
Author Topic: Possible access to electron precession energy  (Read 660 times)
Group: Moderator
Hero Member
*****

Posts: 1140
Some years ago I suggested the possibility of applying a pulse with a fast leading edge to ferromagnetic material in order to gain energy from an anomalous change in the Larmor precession angle of the electrons responsible for the magnetization.  The paper is given below.  Unlike NMR and FMR it does not attempt to cohere the precessions to get resonance, if the effect is real it will apply to all non-cohered precessions that occur in bulk material.  The pulse rise time has to be small compared with the precession cycle time, and this poses a problem when dealing with permanent magnets because the Larmor frequencies are so high.

Recently it has occurred to me that it is possible to use a pair of NdFeB disc magnets in a "Hemholtz Coils" configuration to get a reasonably uniform field in a space between them, see FEMM picture below.  Now if we put a bonded NdFeB magnet in that space we can get the internal field of the bonded magnet near to zero without it becoming demagnetized.  The electron dipoles in the bonded magnet are still aligned, still precessing about the same axis, but their Larmor frequency is now much lower.  The needed pulse rise time is now much lower and more easily achieved, and this could allow a much easier experiment to be carried out.  Is anyone interested in in doing this?

Smudge
   
Group: Elite
Hero Member
******

Posts: 3512
It's turtles all the way down
Dear Smudge

I enjoyed reading your paper, though I admit I must give it a second look to attempt to extract the basics of an experiment.

If you could outline a simple experiment and define the setup, I might be willing to give it a shot, this so we can be sure it is understood correctly. I do have various large coils that can be used for the Helmholtz coils, or if magnets are preferred for this, I could dig some up.

I'm more concerned with the drive parameters and arrangement for the excitation / output coil.

Quote
The circuit must allow the flux to oscillate
rapidly and freely at this internal resonance, not forced to oscillate at electrical external
resonance. This spike collection circuit must be designed along UHF principles.

If this is the case, it would seem that the drive / output coil must have very low inductance, perhaps just a single turn around the magnet, and stray capacitance minimized, if I understand it correctly.

Best regards



---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Hero Member
*****

Posts: 843
Smudge,

Why a bonded NIB for the test PM?  I assume then that the outer bias PMs are sintered?

Regards,
Pm
   
Group: Elite
Hero Member
******

Posts: 3512
It's turtles all the way down
I found this but still not sure if I understand why bonded magnets must be used. Is it the micro-powder that is created? Maybe Smudge can elaborate.

http://www.permanentmagnet.com/bonded-NdFeB-sintered-neodymium-magnets-20150907.html

Quote
Bonded NdFeB Magnets VS Sintered Neodymium Magnets
Preparation Method of Bonded Neodymium Magnets

Preparation of bonded NdFeB magnets adopts MQ or HDDR to produce magnetic powder.
MQ
The melted ingots are poured on rotating water-cooling copper roller, which can make non-crystal or micro-crystal thin strips. Rough particles (about 200Pm), after proper heat treatment, will be changed into nano-state.
HDDR
Ingots are treated by hydrogen. Through hydrogenation, disproportion, desorption and crystallization, micro-powder comes into being.

Through slight breaking, MQ or HDDR powder can make into bonded neodymium powder. Bonded NdFeB magnets are produced by compression molding, injection molding and other method, using magnetic powder mixed with high-molecule compounds, such as epoxy. MQ are mainly used to manufacture various kinds of isotropic permanent magnets. Magnetic powder of HDDRA can be made into isotropic and anisotropic bonded neodymium magnets.

Advantage and Disadvantage of Preparation Method of Bonded Neodymium Magnets
Performance of bonded neodymium magnets is less excellent than that of sintered neodymium magnets , but they, with more precise sizes and simpler techniques, are more stable, easier to produce in batch.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Moderator
Hero Member
*****

Posts: 1140
Smudge,

Why a bonded NIB for the test PM?  I assume then that the outer bias PMs are sintered?

Regards,
Pm
If you are looking towards using one magnet to (almost) demagnetize another one then the one to be (almost) demagnetized must have a lower Br than the other.  This is because the field drops rapidly with distance  from the one that is doing the demagnetization.  Hence I chose bonded NdFeB against standard NdFeB.  A feature of NdFeB is that it can have reverse field applied that will almost negate its own field without the dipoles flipping, without the magnetization reversing (the reversal can occur in the third quarter of the BH characteristic). Other low Br magnets such as ceramic cannot be driven this way as their reversal occurs at higher B levels in the second quarter.  The objective of getting near zero field is to have dipoles that are still aligned but their Larmor precessions are at much lower frequency than the GHz values that apply near the Br field.

Smudge 
   
Hero Member
*****

Posts: 843
If you are looking towards using one magnet to (almost) demagnetize another one then the one to be (almost) demagnetized must have a lower Br than the other.  This is because the field drops rapidly with distance  from the one that is doing the demagnetization.  Hence I chose bonded NdFeB against standard NdFeB.  A feature of NdFeB is that it can have reverse field applied that will almost negate its own field without the dipoles flipping, without the magnetization reversing (the reversal can occur in the third quarter of the BH characteristic). Other low Br magnets such as ceramic cannot be driven this way as their reversal occurs at higher B levels in the second quarter.  The objective of getting near zero field is to have dipoles that are still aligned but their Larmor precessions are at much lower frequency than the GHz values that apply near the Br field.

Smudge

Smudge,

OK, here is a test using two 1/8"x 1" dia N52 grade NIBs for bias with a .25" x .5" dia N35 grade NIB for the test PM.  I 3D printed a bobbin and wound 30T of 26 ga with an inductance of 9.1uH on the test NIB.  I must say that it was a job to get the three NIBs together and clamped!  ???

Anyway, here is a scope shot and I'm not sure what to look for so any suggestions that you would like me to try?  The CH4(grn) trace is the current measured with a current probe and CH2(blu) is the current measured thru a 1 ohm CSR.  The CH1(yel) is the input pulse from a mosfet driver.

Pm

Edit: It just dawned on me that I probably don't have enough mmf for this test, correct?
   
Group: Moderator
Hero Member
*****

Posts: 1140
Hi PM,

Thanks for doing that test.  I would concentrate on that ringing at around 63MHz.  That 200nS pulse is too long and I would try reducing it to around 16nS when it might cohere with that ringing.  Then remove the outer biassing magnets and see whether the ringing is still there, or whether it has changed.  What I would be looking for is an anomalous negative voltage spike at the back end of the pulse.  Of course the ringing might have nothing to do with the magnetism and everything to do with induced eddy currents and coil to magnet capacitance.  If you could create the same geometry using non magnetic materials and see what happens, that would be a good thing to do.  Does altering the magnet spacings change things?  It would be nice if you could discover something that is definitely related to the magnetism, then we could concentrate on exploring that further.

Smudge
   
Hero Member
*****

Posts: 843
Smudge,

Here are some tests with shortened pulses that are at the limit of my Rigol SG in the pulse mode.  I originally had short clip leads between the driver and the PM assembly so I removed them and attached the coil leads directly to the driver to eliminate the extra lead inductance.  The resonant ringing frequency increased as a result as you can see.  I also eliminated the current probe to save confusion.

The first two scope pix are with slightly different pulse widths for comparison of the back side ringing.

Before taking the last scope pix, I removed the PM array from the clamp fixture and must to my surprise, the inner 1/2" N35 NIB had it's field reversed and was now in attraction mode!  I must say that I applied a much higher MMF to the coil than what was used in these tests, so maybe this created this situation?!  Anyway, the last scope pix shows the coil with the 1/2" NIB inserted but no other material attached.  The waveform difference is considerable!

Perhaps I have the PM strengths reversed as it would seem that the weaker Br or N35 should be the bias while the N52 should be the test PM!?

Regards,
Pm

Edit: Replaced scope pix EP5 with EP5C which has corrected current waveform.  Was not making correct contact on the previous pix.
« Last Edit: 2019-03-19, 20:13:12 by partzman »
   
Sr. Member
****

Posts: 488
... Now if we put a bonded NdFeB magnet in that space we can get the internal field of the bonded magnet near to zero without it becoming demagnetized.  The electron dipoles in the bonded magnet are still aligned...

Hi Smudge and all,

This is an interesting idea but the alignment should be checked, as we have no evidence that the magnetic domains will not flip.
Partzman's observation that the field is reversed in the bonded magnet is even an indication to the contrary. This may not be inconvenient if the reversal occurs only at a given threshold, in which case staying below the threshold solves the problem. But if the effect is progressive, the initial idea is no longer feasible with a NdFeB bonded magnet.

It is possible to verify this by the Barkhausen effect. With a coil of high inductance around the bonded NdFeB magnet connected to an audio amplifier, either you hear crackling when you approach the two sintered magnets and then you can suspect a progressive reconfiguration of the domains, or you don't hear anything except at a unique threshold and then you can pursue the idea.



---------------------------
"Chance favours only the prepared mind."  Louis Pasteur
   
Group: Moderator
Hero Member
*****

Posts: 1140
This is an interesting idea but the alignment should be checked, as we have no evidence that the magnetic domains will not flip.
If the magnet is taken to a point in the second quadrant of the BH loop before the start of magnetization reversal, then there are no flips taking place.  For some grades of NdFeB magnetization reversal takes place in the third quarter, especially at the lower temperatures.  So it should be possible to take such magnets down to near zero field as I showed in my image in the first post.
Quote
Partzman's observation that the field is reversed in the bonded magnet is even an indication to the contrary.

My guess is that PM just clamped his three magnets together without any spacers (perhaps PM would verify this).  Doing that is quite likely to invoke a reversal.  I have used NdFeB to reverse ceramic magnets, and even to get partial reversal at the mating surface so that the ceramic magnet became what some people wrongly call tripolar (i.e. N-S-N or S-N-S) buts was actually a linear quadrupole (i.e two dipoles in series).  The problem with this is the NdFeB magnet is weakened magnetically, and since that weakness is locally concentrated it leaves the NdFeB magnet in a state of stress.  Use it several times and eventually the magnet just shatters.
Quote
It is possible to verify this by the Barkhausen effect. With a coil of high inductance around the bonded NdFeB magnet connected to an audio amplifier, either you hear crackling when you approach the two sintered magnets and then you can suspect a progressive reconfiguration of the domains, or you don't hear anything except at a unique threshold and then you can pursue the idea.
I agree, I think that threshold can be near zero B and this is a good way to check that.
Smudge
   
Hero Member
*****

Posts: 843
My guess is that PM just clamped his three magnets together without any spacers (perhaps PM would verify this). 

Yes, I clamped the Nib trio without any spacers.

Pm
   
Group: Moderator
Hero Member
*****

Posts: 1140
Yes, I clamped the Nib trio without any spacers.

Pm
Perhaps you could repeat the experiment starting with large spacers so as not to get magnetization reversal (use the coil to check this doesn't happen during the assembly, any reversal would cause a large voltage spike).  I think the ringing you saw is not what we are looking for.  We want a fast rise of current in the coil and the current was rising quite slowly and linearly throughout your 200nS pulse.  Maybe try longer pulse duration or try series R to get shorter L/R time constant.  Gradually reduce spacer thickness to get the inner magnet closer to zero field.
Smudge
   
Hero Member
*****

Posts: 843
Perhaps you could repeat the experiment starting with large spacers so as not to get magnetization reversal (use the coil to check this doesn't happen during the assembly, any reversal would cause a large voltage spike).  I think the ringing you saw is not what we are looking for.  We want a fast rise of current in the coil and the current was rising quite slowly and linearly throughout your 200nS pulse.  Maybe try longer pulse duration or try series R to get shorter L/R time constant.  Gradually reduce spacer thickness to get the inner magnet closer to zero field.
Smudge

I haven't given up on this but I've had a medical issue which has slowed me just a little.  Anyway, after giving more thought to this and looking at the curves for the various NIB materials, I concluded that to perhaps achieve what you are looking for, three identical NIBs could be used with variable spacing.  I came to this conclusion after studying the formula to calculate the surface area flux density from residual flux density Br.  It seems that by having a variable gap capability, the correct bias point down near the bottom of the load line could be achieved.  This is probably all wrong as I have tried this and it didn't work as I had hoped.  I've attached a pix of the assembly where three 1/2" N35 NIBs are held in place with moveable plungers in each end which allows the vise to easily adjust the gaps once in place.

With all variations of pulse width, current, and series resistance, I can see no anomalous pulses other than what I think are simply from the circuit inductance.

Pm
   
Group: Moderator
Hero Member
*****

Posts: 1140
OK and thanks PM for doing this work.  Disappointing but not surprising.  Maybe conductive magnets is not the way to go because of eddy currents.
Smudge
   
Pages: [1]
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2019-06-17, 02:02:48