Here is some more important info. I have just re-established contact with Graham Gunderson. He worked with Brian Ahern on the Manelas stuff for a while, and took apart one of Arthur's earlier devices. Here is his response. I did spend a month living with Brian Ahern, when he had several of Manelas's devices in his home. We also reviewed Arthur's labs, still pretty much intact, at Arthur's widow's home. The one machine that appears to have delivered convincing results had been disassembled, and was mostly an empty frame. Still, it yielded some important clues.
WhenI met Arthur he had already had his stroke. He was bedridden and in a care facility. It seemed he could understand me, but he was unable to speak and could not really move. I said I was there to help him and his family get the recognition they deserve for the discovery. And that I would need his help in locating his notes -- which he had claimed to have hidden somewhere in the walls of the family home. Arthur appeared to understand, and looked at me with an expression that read in his eyes like pain or regret. I could not make out why.
I will get through the bad news and move to the good.
For me the worst of it is the early demonstration machine of his I took apart. I saw a photo of Arthur in a house full of visitors, and this machine. It was lighting a small array of LEDs. The visitors were there to see it and meet Arthur. This machine was still intact when I ran across it. It was a transistorized oscillator connected to what appeared to be 6x4x1 billet wrapped with triaxial coils, in the fashion I suppose you've seen before, and probably in other photos of Manelas's work.
The machine did not operate, and when prodded to do so with restoring its loose wiring and adding a DC input I measured efficiency at about 30%. At the time I had concluded it was an unstable and poorly built relaxation oscillator. I say "poorly built" since the single transistor there was a power transistor with a low current gain, requiring substantial base current to trigger it. These currents are provided through resistors, incurring substantial circuit losses. There were other issues which looked on the surface like poor design, and some chance those "bad" things are what makes the machine actually work. I kept an open mind, at least.
The field pattern on the billet was very strange, in that both of its 6x4 surfaces were polarized North, while the region between those surfaces (the peripheral, 6x4" rectangular surface about 1" high) was the opposing South pole. Magnetizing a block with that pattern would require a giant H field and Arthur's equipment wasn't of that magnitude. With approval from Brian and others I made the decision to unwrap the billet and have a look, since the machine did not have any OU. It was said that the conditioning process would only hold for some time, and after the machine is shut down, it dims away. Floyd Sweet said the same thing.
Once the "billet" was unwrapped, it was actually 2 billets each 1/4" thick, in a sandwich explaining the pole pattern I'd observed; they were oriented repulsively and the space between them was filled with thin Nokia Li-ion batteries removed from old cell phones. These batteries connected to leads which exited the billet structure and powered the device. Of course we talk about Ockham's razor championing the simplest explanation, and in this case I think that points to deception. It is one thing to have batteries visible, quite another to hide them.
After that I saw Arthur's work with a rather more skeptical eye.
The second item of "bad news" relevant here, beyond the tragedy of Arthur's stroke, was the early car he had self-running -- quite convincingly -- and then the second car. A Texas investor had bought Arthur a nicer car, a Chevy Volt, for Arthur to outfit with his invention. Arthur could not deliver, and I heard the relationship between the men had soured to the point the investor had tried repossessing the Chevy Volt. I saw this car in Arthur's garage (among many, including antiques) and it had only minor modifications to its wiring, apparently to insert a power source, but nothing more.
Finally, Brian had 2 large power sources built by Arthur in his home. These used the arrangement of 5 spool shaped coils, each fitted with a relaxation oscillator and transistor circuit on top, connected to a billet and some batteries. I operated both devices across as wide an operating range their silicon would allow. The highest efficiency I measured was in the tower at just over 80%. After a while I concluded I'd covered the operating area and its variables and there wasn't any OU, and it's about that time I returned home.
Arthur used a strange, custom made wire that is bifilar and resembles twisted magnet wire. In Arthur's wire, one conductor helices about the other conductor which is nearly straight. Both conductors are connected in conventional parallel, as the conductors within Litz wire would be when the wire is soldered at its ends. In Arthur's arrangement, fast wavefronts will see different propagation factors in each conductor, since one's a helice and one is straight. Arthur had this wire especially made at New England wire, which makes Litz and other custom formulations.
There was a lot of talk at one point about the inventions using "bifilar" coils, but the ones I saw aren't like the radical ones used in Sweet's VTA (which inductively cancel, but with a distinct "fingerprint" in their interference pattern). Arthur's were a bit more like Litz wire, offering parallel paths and more-or-less conventional induction. Although fast wavefronts would have propagation dispersion as mentioned above, the transistors and diodes he used have slow switching times on the order of 1 uSec. However, the rather dated silicon diodes in his rigs had large junction areas and pronounced reverse recovery current. Many diodes of that kind do create rapid impulses when suddenly reverse biased, as they continue to "conduct" until the PN junction is exhausted of residual charge carriers. The exhaustion ends abruptly, and the equivalent "turn off time" can be sub-nanosecond and electrically "loud" like a slamming door. Early IGBTs behave the same way, and emit a peculiar form of EMI. I didn't observe this in Arthur's machines, but the silicon he used can do it.
OK, now, the "good" news!
Arthur left hand written notes documenting a machine that did not resemble any of the machines I saw,except the skeleton of the device he had installed in his early self-charging car.
That device had temperature probes and a data logger installed, which documented internal temperatures 3-5'C below ambient in the machine volume, about the size of a shoe box; and coldest at the billet. The output of the machine was monitored and also charted. The absorption of ambient thermal energy correlated to output power, both of which varied over the course of a day -- the more output power, the more heat it appeared to absorb. There were no other regions showing elevated temperature, so the data appears to suggest thermal absorption rather than merely heat being pumped between locations.
The logging showed occasional sudden glitches in the output and the machine would sometimes nearly fail, for a few minutes, but then revive. A young scientist friend of Brian's connected these events with sunspot or solar-flare activity, and found they match, with a delay.
I am encouraged by the distinguishing features of Arthur's "Car" device, although I only saw it in a partial state and read detailed notes that matched its construction. It is not like the others I saw.
1) The notes suggest a single, solid billet, wound with triaxial coils.
2) The billet is magnetized with a gradient, as though a circular coil about the size of a coffee mug is centered in the 6x4" pole face and pulsed, creating a circular region of field reversal in the center of the magnet. One picture showed Arthur levitating a steel needle above this region, and the needle is trapped in a clear plastic tube whose axis is normal to the magnet surface. That constraint is required by Earnshaw's theorem, which this effect does not override, but the needle can and does levitate inside the tube; I've done it too.
3) The power transistor used in the Car invention is a TV horizontal flyback type, not the slow, and low-voltage 2N3055 type used in Arthur's other inventions. Instead, the flyback types have even lower current gain (10-20), high cutoff voltage 1.2 - 1.7kV (VCEo) and fast, abrupt turn off over some 500nSec or less.
4) The 3 coils around the billet are triaxial. Coil "A" wraps around the periphery, forming a 6x4" picture frame shape. Coil "B" wraps around the long 6" axis of the magnet. Coil "C" wraps around the 4" axis of the magnet. The connection of these coils in Arthur's running circuit matches the role of the same coils in Floyd Sweet's conditioning process.
This is what blew me away. I'll share it in some detail.
Coil A: In Sweet's process, this frame coil is excited with low level sine wave AC. The magnetic axis is parallel to the anisotropy axis, "Easy" axis, and crystal alignment axis of the magnet. In Arthur's process, this coil is connected to the base of the flyback transistor, and triggers it. As the transistor has low current gain, a reasonable amount of current must flow in this coil to trigger the transistor. So, although it appears to be a "sense" coil, the current flow is non trivial and its H field along the easy axis is also non trivial. In both Sweet and Manelas machines, this "A" coil gates the field activity on the other axes, and is also weaker than the field activity on the other axes, and is directed along the dimension that the magnet's sensitivity, nonlinearity, and susceptibility is the greatest.
Coil B: In Sweet's process, this coil is pulsed with DC current pulses that are timed to fire at the peak of the AC wave in the A coil. (It isn't recorded whether this is a voltage peak or a current peak; only that there is synchronization). Each pulse has the same polarity of H field, and the pulses are repeated many times. In Arthur's "Car" device, this coil is connected to the Collector of the flyback transistor, the current and EMF in it varies abruptly. A take-off diode conducts the EMF spikes occurring at the transistor turn-off into a battery bank, like a flyback converter. There is no capacitive bypass, meaning that the battery chemistry and internal ions get jolted and kicked; and the impedance of the battery chemistry is reflected into the B coil and the magnet's changing field structure when the take-off diode conducts.
Coil C: In Sweet's conditioning process, this coil is pulsed with alternating DC current pulses, driven by the same source powering B. Whereas the polarity of the pulses (and resulting H field) in B donot reverse, the polarity of current and mmf in the C coil does reverse on each successive pulse. In the early days Sweet accomplished this by flipping the magnet by hand, for each next pulse, and re-installing it in the surrounding coils. In Arthur's machine, the DC input power to the device from its primary battery runs through this C winding in series, providing a fairly static H field along this axis. If I remember right, there was some capacitive bypassing in this circuit, yielding steadier currents (and mmfs) than seen along "B".
In summary, each system has a "tickling" influence along sensitive axis A, and the area and inductance of the A coil are large, and the mmfs here control the timing of the rest of the process.
Each system has different and complementary roles on the B and C coils, which are both transverse to the magnet's sensitive axis, and whose area is thinner and the inductance per turn lower. One of the pair I would call disruptive (the alternating, repeated current pulses in Sweet's C coil and the hard-hitting pulses in Arthur's B coil); while the other in the pair has influence I could call supportive (as the non-reversing, repeated current pulses in Sweet's B coil and the DC-like current flow in Arthur's C coil). Manelas and Sweet used opposite roles for B and C, but the coils' structure is similar and the magnet material sees a hard-axis mmf from both. If the magnet were 4x4" or square, I don't think the claims would change. The 6x4" geometry is only a de facto standard left from early magnetic separators using large inexpensive barium ferrite magnets.
I hope this helps. Much of the above I've never written down, but I share it with you out of my debt of gratitude to you. I certainly encourage sharing these details with anyone you wish. Note that when Graham wrote that he was not aware of the five toroids on top of the billet. That information only came to me a few weeks ago. Smudge
|
Author
Topic: Manelas Device (Read 6036 times)
