STEORN DEMO LIVE & STREAM in Dublin, December 15th, 10 AM

[PaulLowrance]

Paul, I have a number of those textbooks and a lot of industry literature. The thoughts I posted are based on that literature. I'm afraid I don't have the math background to do more than simple algebraic calculations, but I can follow the dynamic relationships pretty well.

If you would take some time to consider some of the the properties I mentioned, you may see that they too can influence the overall behavior. High permeability is one of the desired characteristics. But by no means is it the only one.

For the sake of further discussion, let's talk about remanence for a moment. A high remanent magnetization will provide a magnetic bias that reduces the DC input energy costs to saturate the core.

Using longitudinally annealed Metglas 2714AS as an example, we see it has a Bsat of 0.57T and a Br of 0.45T, for a Br/Bsat ratio of about 79%. That's pretty high compared to most materials.
(page 3 of: http://metglas.com/downloads/magamp.pdf )

Now compare that with Finemet FT-1H, which has Bsat of 1.35T and Br of about 1.21, for a Br/Bsat ratio of 90%.
(page 8 of: http://www.hilltech.com/pdf/hl-fm10-cFinemetIntro.pdf )

The Finemet will retain 90% of its magnetization when the coil is deenergized where the Metglas will only retain 79%. Which material will require more input energy to push into saturation?

Oc, I'll make this brief because you keep missing the point. So I don't think I can help you much right now. Briefly -->

First of all you don't base it on percentage. You need to know how much does the field increase, the permeability, and the core characteristics.

Second, your question is pointless in terms of the Orbo design because *yet for the umpteenth time* LOL, what does work in the Orbo design is when the field is well within the saturation curve. I keep telling you guys over and over and over and over that the work done in decreasing the magnetic attraction between the magnet and core is well within the saturation curve. Furthermore, the amount of energy require to get the Metglas core to the point where it's within the saturation curve is so ridiculously small that it's irrelevant. For common cores, yes, it's somewhat relevant.

The Finemet MAGAMP core is like placing ~ 2.5 Metglas MAGAMP cores in the same space. *If* it takes 2.5 times as much current in the Finemet core to force the dipoles out of being aligned with the magnet as compared to the Metglas, then that's 6.3 times as much power-- P=I^2*R. So you would get 2.5 times the force, which equals 2.5  times the work, but it takes you 6.3 times the power, thus making it 2.5 times less efficient. IMO, there is a high probability that it will take more than 2.5 times as much current in the finemet core, thus require more than 6.3 times the power. The only saving grace for your finemet cores is if by chance somehow it takes less than 2.5 times the current to force the dipoles away from the magnet, but I wouldn't bet anything on that.

[PaulLowrance]

Real quick, so if you say, "Well, if the Finemet core is like 2.5 times Metglas cores in the same space, then just move it farther away."  Of course you can do that, but moving it farther away you end up with the same force.  If moving the core farther away has other side effects that are good, then the Finemet is a better choice. We'll have to see.

[0c]

A into simple concepts:
the remanence is the ability to retain a magnetic polarity.

this can help the motor torque only when the next magnet have inverted magnetic polarity, otherwise you reduce the DC input energy but you reduce also the attraction force.

Immediately after deenergizing, the core will retain a remanent magnetization, oriented in a circular polarity that is not externally detectable. This "biased" core will have very near the maximum permeability and attraction to an approaching magnet (permeability is higher near the knee of the BH curve than it is at zero). As the magnet approaches, the domains in the core will be rotated to align with the field of the magnet and the core will essentially become a diametrically magnetized magnet, oriented for maximum attraction to the approaching magnet.

Rotating the prealigned domains of the core due to remanence consumes less energy than aligning randomly oriented domains (zero remanence). This the reason why permeability is higher at magnetization levels above zero.
(This is a reference for ferrite materials, but the principle is the same, see figure 1:
http://www.tscinternational.com/tech4.pdf )

So contrary to what you said, attraction to the approaching magnet will be greater when it is remanently biased than it would be if it had no remanence.

in any case you have energy loss in the core (heating).

There will always be heat, but the heat generated is proportional to the amount of work done. Since rotating prealigned domains (remanent magnetization) requires less work than aligning randomly oriented domains (no magnetization), less energy will be lost to heat for a remanently magnetized core.

the best solution is to have an "elastic" property when the magnet approach the spin becomes aligned with the magnetic field, then with the coil current you flip the spin alignment along the toroid core and saturate the core (it appear like air to the magnet).

The challenge is to have the new spin alignment with gain of energy that recover the work of alignment done by the magnets.

That's almost what happens with a remanently magnetized core, except the prealigned domains are only required to "rotate" 90 degrees instead of "flip" 180 degrees.

metglas annealed along the core have the possibility to do this job?.

Possibly, but I'm not sure. I think other materials with higher Br/Bsat ratios should be investigated.

[gravityblock]

The Finemet will retain 90% of its magnetization when the coil is deenergized where the Metglas will only retain 79%. Which material will require more input energy to push into saturation?

If the metglas retains 79% of its magnetization when the coil is de-energized and 79% of the energy stays inside the core, then PL can't possibly recover up to 95% of the energy from the pulse.  The most that could be recovered from the pulse is 21% in the perfect case. 

That is a 174% gain just by energizing and de-energizing the metglas core if you can re-capture 95% of the pulse according to PL, which is close to PL's 170% efficiency claim for his tiny Orbo replication.  We have OU just by energizing and de-energizing the metglas core.  I don't think so, lol.

PL, your COP calculations don't add up. 

GB

[0c]

The Finemet MAGAMP core is like placing ~ 2.5 Metglas MAGAMP cores in the same space. *If* it takes 2.5 times as much current in the Finemet core to force the dipoles out of being aligned with the magnet as compared to the Metglas, then that's 6.3 times as much power-- P=I^2*R. So you would get 2.5 times the force, which equals 2.5  times the work, but it takes you 6.3 times the power, thus making it 2.5 times less efficient. IMO, there is a high probability that it will take more than 2.5 times as much current in the finemet core, thus require more than 6.3 times the power. The only saving grace for your finemet cores is if by chance somehow it takes less than 2.5 times the current to force the dipoles away from the magnet, but I wouldn't bet anything on that.

Ahhh, but it's still an unknown. How much power is required to push an equivalently sized Finemet core into saturation? Can equivalent or better results be achieved with smaller cores? How much additional torque can be gained at the required distance? Do you have the numbers? I don't.

Paul, I'm not saying Finemet is the answer. I'm merely suggesting that there are other magnetic properties to consider besides permeability and other materials which exhibit those properies. I still think it should be investigated.