Ok, so this is only a sixteenth-baked, and maybe some1 already has talked about it that i didn't notice:

But it seems to me that if we can first somehow come up with a way to actually see magnetic fields in 3D in real time... Then it will be a heck of a lot easier to develop devices that run on magnetism as the motive force: Because then when any changes/adjustments are made, there will be instant visual feedback to the results; not mention the general increase of understanding how it works, or a greater ease of proving the device's concept to the scientific community.

Like some sort of hologram projector where suspended ionized particles can be seen only when they intersect with the flux lines. Or maybe a fluid or gas or smoke  that is non-conductive and does not affect the field, but will somehow be colored by it; and then immerse the device in the liquid or gas while operating and making adjustments. Or a steady stream of gas blown onto it; with something like suspended carbon-iron filings but that would "float" without falling immediately, and also not affect the circuit (like a 3D analog of the old 2D trick with magnet, paper, and filings).

Or some use of coherent light, in a gas bath around the device, so that the light reflection qualities of the gas are somehow affected by the flux when the laser hits it.

If we could actually see what was going on with the fields with certainty, i think the big breakthroughs would come a lot easier and faster (especially with something like the TPU).

I got this crazy idea while watching an ultra-sound monitor at an out-patient facility; thinking how cool it would be if sound waves could show magnetic lines of force like they do tissue.

But i have absolutely no clue how to really do this lol... or how to make any of the above concepts work. I do believe it's possible someone out there can devise a way to do this, probably by some totally different means... hopefully in a way that a home inventor can use it without too much cost or trouble.

Thx, that little bottle thingie is interesting.

The problem with puter sims is without good baseline data to create it with it won't be accurate, GiGo. I guess you would have to make measurements of the flux density every cubic mm of the space all around the device, then input it in to a data analysis proggy, then it should work OK but only for that particular config. Because you couldn't actually model it "cold", since we don't exactly know what to expect yet without the actual measurements being done for every single permutation of the adjustments (different component values, coil configs / core composition and wire lengths, excitation F's... etc.)  One heck of a data acquisition job Once you did a few you could take risks and predict trends with the model but they still could be wrong very easily; until you actually had hard measurement-sets for every possible permutation you couldn't really trust it.

But if you first built some sort of raster-like scanning device as a physical carrier for the flux density measuring transducer, something like an 3-axis x/y/z plotter (or a robotic arm) with output to an a-to-d , to take the hundreds of thousands of measurements quicker and more accurately  (and that yet didn't affect the circuit in any way), it would be possible enough to display the results in 3D color-graded graphics that could be spun around and looked at from any angle; not in real time but close enough (many test & measurement data acquisition analysis proggys will do this graphing, although to sync the x/y/z coords of the "plotter" with the measured data would take some tweaking). The advantage of all that would be that once you had many complete sets of data you could compare and do all kinds of calc's like FFT's to figure out the best "sweet spots" for best excitation f's and output voltages for different components /adjustment settings.

X/y plotters (pen "two-axis printer-plotters" used for drawing, but not the kind that use a roller for the "x") can be bought used fairly cheap, they are obsolete now since wide-frame ink jet printers are better & cheaper.. 2 of them cobbled together could make a "3-D plotter"; or again, use a fairly precise-resolution robotic servo arm (more expensive but a heck of a lot easier to program and make work). There are prolly small flux transducers with direct serial output and fast enough sample rate (...but needing to be of at least 12-bit res.) making the actual single data-point capture part easy; but these would likely be fairly expensive too. 

Even with a fast raster, it may take 60 minutes or more to sweep the area once at 1 mm res ( 1 cubic M = 1 billion samples lol... each with a unique x/y/z coordinate). And with aliasing; you would prolly need that resolution for it to be worthwhile... 1 mm would prolly give you a true resolution of 2 mm's (+/-1); depending on the accuracy of the plotting device.

This is something that the engineers of a well-funded corporation, university, or independent lab that was developing the technology would likely spend the money and time on, however; because the rewards are worth it. And once you had many and varied complete data sets, future predictive modeling could be done with much better reliability.... But not until then.

A bottle of baby oil and a brillo pad is looking better and better lol

Here's an idea...

The clear bottle idea, perhaps made into something more like a fish tank cube or rectangle (not too large). And having electro magnets which can be turned off and on at varies stages to "reset" the filling to one end or to a sealed top. Then have that magnet turned off and another turned on to attract them to it. And the test magnet put in the middle as these are being drawn past.

While this is not meant to be the "ONLY" method, it would perhaps bring about an "IN MOTION" reading so to speak. The whole thing would then be video recorded from at least 2 right angles, better to have top view, side view and end view. These videos can then show a side by side effect which would become a standard for viewing these things. Of course the "IN MOTION" mode, is only 1 mode, and without that there would be other still factors, still with the video recording all 3 angles. The end result would still be the need to "reset" the steel wool fibres (filings). Perhaps colouring them in copper plating  coating or something like that to stand out better or other similar colour changes might also help.

If they are selling things like this for "a couple of hundred bucks" then who knows?

Though personally I think money as a motivator is a muddier of ideas for the purity of their evolution.