I want to replicate Nelson Rocha's blue center coil of his radiant box device.

I want to make 2 identical unifilar pancake coils, with a phi ratio hole, and distance them and fill the gap with epoxy (as a good dielectric)

what should the distance be? the radius of the  hole? or the diameter of the hole?

the epoxy resin can be poured in. a mal (casket?) should be made around the first coil and the epoxy poured in and hardened. then the small layer epoxy again poured on top and then the second coil is placed on top. the coil is then hardened out.
the epoxy should not fill the phi ratio hole.

baking oven paper can be used to build the form. and a silicon spray can isolate the epoxy from the paper.

the coils can then be connected in series into a side by side bifilar. the connection can be made with a capacitor.

i can use speaker wire, but i also have a roll af some rather thick enameled coper wire for this. but it is stiff, and will need a guide to roll up into a proper phi holed pancake coil.
I also have some plates of a induction stove.
they are ideal for this job. allready formed. flat multistranded... unifliar.

lets build this.  Im open to suggestions in this build.

with the induction plates i would only need to widen the center hole. the back plates could be kept on. there would be a little distance to the ouside coil. I dont like that.
maybe I could glue them on the baking paper.

Quote from evostars on July 2nd, 2017, 08:15 AM

what should the distance be? the radius of the  hole? or the diameter of the hole?

Radius of the hole at most, since you have two coils (with dielectric between).  Optimal spacing could actually be less than the hole radius, not certain yet, but I think one hole radius separation is a good starting point.


Winding the side-by-side coils is pretty easy, but I found the single layer bifilar coils to be much more difficult--managed to do one, used a lot of hot glue and had to fuss with it non-stop.  Now that I know the proper dimensions for the Phi hole, I'm going to take another stab at the PCB coils.  These should be real precise and fairly easy to do some thorough testing with.  I am however concerned their SRF may be too high to deal with using practical, available components.  Guess there is only one way to find out.

My goal is the same as you Evo--need to replicate Nelson's coil stack and figure out exactly what kind of inputs it needs and where its output is.  Then we can begin trying different input parameters and see if we can find any sign of overunity operation.  There's only a finite number of combinations using Nelson's Radiant Box example, eventually one of us will hit on the right combination.

the back plates can be easily removed. the windings should rotate the same direction so the plates should be removed any way.
and again fixated on someting else.

agreed Matt. we will find out by experimenting with it.

the resonantfrequency of the pcb could be lowerd by enlarging the plate area.
in other words, widen the printed wires so there is more surface area for the dielectric field. the distance between the wires would be detriment by how high the voltage is you want to use

Im going to remove the ferrite rods from the back  of these 2 coils, and then hook them up in series. maybe place some dielectric in between them. see if it learns me something

I have a hunch we can make this work without the ferrite, but I do suspect the ferrite increases the output power for no additional cost.  Just trying to think like Nelson, he probably recognized half the energy is leaking out the back side and found a way to put/reflect it back into the system.

I still think the ferrite was to balance the speaker wire coil windings. but we will see.

I just made a stack of 4 coils. the 2 in the middle are series connected and probed unifilar induction stove coils. (not a phi hole)these have a lot of distance due to the back plate.

the outside coils are 2 speakerwire phi coils.

in i used a capacitor to series connect the unifilar coils. but I could not lower the  frequency below 2mhz. so i will try again with a parallel capacitor.

series connected, one end to ground (outside rim) and parallel variable capacitor added.
560pF

and it works. 889khz primary resonant frequency. with a 300V peak to peak resonant rise in the center coil.

if a good dielectric is used between them, and the center hole made phi ratio.

and a bigger capacitor

also tried grounding at the bridge and measuring both ends. signals are nice out of phase. perfect to rectify and add up into a capacitor. only a slightly higher res frequency of 942khz with the same 560pf cap in parallel connected at both probes.

I could use two caps to fine tune both coils to the center ground. done that before.

Look'n good Evo.   :thumbsup:    Wish I could wind coils like that.

Quote from Matt Watts on July 2nd, 2017, 10:13 AM

Look'n good Evo.   :thumbsup:    Wish I could wind coils like that.

:D
just focus your mind and rip them out of a induction cooking plate

If i would choose the thick copper wire as a unifilar coil, I could do that with 2 perspex plates, separated with a wire thick phi ratio disk. as can be seen on several youtube channels. the advantage is would be to have more tight fitted windings.

but the multistranded coils of the induction stove seem to be widened. perfect for more plate area. but in the end more windings with the same plate area would be best.

the induction coils will have to do for now.
I can simply remove the backplates. fixate it on a thinner plate. remove the windings in the center until phi ratio is achieved.

once that is achieved, a dielectric could be placed in between. to test. epoxy would be the last phase because it is rather permanent.
a temporary dielectric could be acrylic plates.
as long as they have the phi holes in them

I do believe the excess wire of the center hole shpuld be cut off and reconected to an other diameter wire to less influence the resonance.

how to proceed... the induction stove coils need to be widened in the center hole to male it phi ratio. i need to unwind 3 windings in the center.

I also need to get them of the backplates. they are glued on there, and are easily loosened. before they are loosened the coils need to be fixated on something else. maybe a temporary tape would suffice.

I would love to use al the coil windings, and shape the coils to the same diameter of the other speaker wire coils, but... that would be to hard. the wires are multistranded and twisted... pretty hard to work with. And the diameter is already closely matched.

the excessive wire from the center should be cut. but still kept long enough to make the series connection.

so...
step 1 fixate with tape
step2 remove backplate
step3 remove inner windings to meet phi ratio.
step 4 cut excessive wire, and reattach connector.

Quote from evostars on July 5th, 2017, 10:50 AM

step 4 cut excessive wire, and reattach connector.

If I'm not mistaken, those windings are Litz wire.  So you will have to remove the insulation from each strand before you bind them back to a connector.  If you don't you'll have floating conductors in the bundle and no telling how bad that will throw things off.

Just go slow and careful.  Try a little piece and see what it takes to get the insulation removed and some solder to stick well.  In my experience it has always been:  super clean, heat and lots of flux.  At some point you'll see the solder swarm all the stands and make a solid connection, then you can add the connector.

yeah thats gonna be the worst part.
maybe im lucky and the heat will do the most work. I've got a 80W 450 degree digital solderstation. that should do the trick.
what i do withsingle stranda usually is burn, an sandpaper.

coil fixed with tape came of easily. took away enough windings.... now for the hard part

got 2 coils now...
multistranded. lets apply some hot solder and see what happens to the coating

too bad. 450 degrees celsius didnt do the trick.

next... fire...

Quote from evostars on July 6th, 2017, 02:55 PM

next... fire...

Yes fire.  I use a little torch until the copper glows.   That usually gets the finish off, then fine grit sandpaper and lots of flux and you should be good to go.  Probably best to heatsink the copper just a little ways up the strands so only the tips glow.

Those coils sure do look nice.  If you get them working, don't underestimate the power they may be able to transfer.  1000 watts plus is what they are designed for; you might get that in a burst.  Be ready.

burned the coating of the single wires off.
now... sandpaper... pfff piece by piece


yeah 1000W that would be a nice start ;)

would acetone desolve enamel over a night?

I think it will, but the coating will stay gooey.  You'll still have to get it off somehow.  Flux (and heat) is your friend with these fellas.  The cleaner the better.

Im sandpapering single wire after single wire.
better do it good or else i might regret it.
works pretty well... just feeling like a zen monk

one done one to go... break time