what are the properties of these coils and how do they interact(at resonance)

the widened central hole weakens the magnetic field but what more is going on with these coils?

Matt Watts showed the resonant coil can be distanced without loss of amplitude.

https://vimeo.com/224646050

I've got two of these coils.
it consists of 2 pancake coils. the center one has a large hole. this is the coil that resonates. the coil wrapped around it on the outside is pulsed. I pulse it from the outside because they are side by side bifilar pancake coils made of wrapped up speaker wire.

I want to see how the resonant coils interact. at several distances and phase relations.

I took my toroid, with the bucking coil on it, and measured the impedance of both counterwound coils. they were 0,30mH and than I measured the primary of my double pancake coil, it was also 0,30mH
This is perfect for transferring bemf voltage spikes from the bucking coils into the bifilar coils.

On the toroid there is a primary wound (that could be made bifilar) that is being pulsed bu my igbt. This creates a magnetic field in the toroid, and the b emf is captured by the bucking coil. The bucking coil is grounded in its center tap.

Both ends of the bucking coil produce b emf impulse currents. of several hundred volts, equal polarity.
But it needs to be 50 khz to 100 khz because at higher frequencies it start ringin.

The b emf is fed into a bifilar coil, that makes a center bifilar coil resonate.
So the b emf, from a 12V pulsed toroid is used to get a bifilar coil into resonance.

The resonant bifilar coil, needed to tune down its resonant frequency.  This I did by adding a capacitor in parallel over the resonant coil.
This, became a problem. because the resonant Sine voltage rise(from the few hundred volts b emf) is potentialy several thousand volts in amplitude.

I didnt realise this at first. So I just used 2 caps in parallel to get 7nF. one of the caps had a 100V rating. The sine showed a rappid growing but in a strange way. then I noticed a strange smell, and I saw the 100V rated capacitor melt away :cool:

Woww what?
So I cooled down, and got 10nF 630V rating capacitors, and dialed down the resonant frequency to around 80khz. And it became hot.
 fast. So I stopped again. (not realising the resonant voltage rise had a potential of much more than the 630V).

To protect the capacitor, I added a 330 ohm 1/4W resistor. The sine now was lower (i remember something like 35V). But still the resistor became very hot!!!

To me this indicates some power. But at the same time, the power is dependant on how high the resonant voltage sine rise gets.
more volts is more power...

I also tried, a 8V lightbulb, across the capacitor, but it killed the resonance, the voltage kept low, and the power was gone, the lamp showed a little glow. I reproduced the glow with a dc power source and it resembeld 0,2W or something low like that. 

So this gave 2 problems, and at the same time a very exciting result.
I could use very low power conduming b emf, to create a very high voltage resonant sine wave.
But at the same time. I didnt have the right capacitor to tune the resonant coil down...
To fix this, maybe I could use several larger capacitance in series, so it will take the voltage better. best would be a 10nF high voltage capacitor... that i dont have right now.

In the past I found out it was a factor 70, so 70x 500V (bemf) would produce a 35000V sine wave potentialy...
WHAT?!?!?!? :-X
ok so thats why the caps were melting away...

Also: No magnetic reaction of the compass that i usually see at the resonant frequency of the bifilar coil.

just like this video from Nelson.

https://youtu.be/XGyz31yaCdw

I could also tune down the resonant frequency by adding 2 caps, from the bridge, that connects both windings, to the start of each winding. but i dont think it will matter. the voltage rise is just to high.

But Nelson seems to use a capacitor to store the B-emf, the capacitor will than have more joules at a lower voltage, resulting in a lower voltage resonant rise. So B-emf into a cap, and the cap discharges into the bifilar coil. hmm... I have to try this. and work it out

one big difference with the cd sleeve coil and the phi hole coil side by side, is that the phi coil is able to be tuned down much lower in frequency, and still have a large resonant voltage rise.

the cd sleeve coil didn't give a proper voltage rise when it was tuned to low.

but... i didnt test the cd sleeve coil with the b emf pulses. so that also could be the trick. because of the lack of magnetic field at the resonant frequency (normally observed with a compass).

the b emf seems to be pure dielectric energy.

while the b mmf (from a cap) is pure magnetic.

one question remains... why did the cap melt? was it the high resonant voltage? or was it from being part of the resonant circuit where a longitudunal current passed through the capacitor?

http://www.radio-electronics.com/info/data/capacitor/polypropylene-film-capacitor.php

I may have used the wrong type of capacitor, causing it to fail.
a malory 150 cap is made of metalised polyester. it works great in audio frequencies but at radio frequencies its different.

poly stryrene also would work but is much more sensitive for temperature.

orange drop capacitors... made of polypropylene. little more expensive... but does the job well.

i just reas the 630v rating was DC not ac
ac rating 250V... duh no wonder it melted

Jackpot :clap2:

I remebered I had a device I build years ago.
does someone recognise what it is?
a hint, its full of diodes too.

found it, its got 22x  0.22uF 400V ac 10%
wima mkp10 capacitors. pretty expensive as i remember.

2 of those in series can take 800V.

but... i need 10nF thats 0.01uF so I will put several in series.

Once you get pretty sure about the values you need, these guys have some beefy parts to chose from.  Pricey stuff though, so you'll want to be certain you get only what you need and no more (unless money is no limitation).

thanks Matt. indeed pricy very pricey.

those wima mkp10 are metalised polypropylene. should work...
lets see if they hold.

22x 0.22uf in series makes 10nf. perfect.
but they are 10%
and I have no idea how so many capacitors in series performs in a resonant circuit.
still if it works Ill know what the circuit really does.

well it works :)
prefect resonant sine around 600V pp
stable at 99khz.
I only used one side of the bucking coil, disconnected the other side and left it open.

If i lool at the B emf pulses... they are not impulses anymore. must be due to the parallel connected bifilar primary.
so this could be improved (diodes?)

the other thing is... I haven't got 2 resonant coils as before. only one, because i only have one 10nf cap (22x series 0.22uf)

I could share the cap and connect the coils in parallel like before but then the resonant frequency is higher probably around 200khz.

the b emf pulses would even be less quality.

its worth a try, but... it would be better to prove the b emf...

All together, this looks promising. as the power needed is low.

in the end, capturing the b emf into a capacitor and use the pulse from the back mmf of the cap would also be very interesting

so many capacitors in series.
hihi

hm made a little error. I thought by bypassing some of the 22 series capacitors the capacitance would go down...
but they are like parallel resistors... so the capacitance goes up, tuning the frequency even lower.

but then i noticed my white bypass wire is getting hot! indicating current.

the voltage rise went down with bigger capacitance and lower res frequency.

I want to find maximum resonant voltage rise, so I should start with a smaller capacitance than 10nF.

I added 6 0.1uf in series and the resonant frequency became 124.65 with less voltage rise than 22 caps at 99.15khz

but I have to account for the internal esr (resistance of the cap) and now I have 3 different kinds of caps so i can come to a good conclusion.

I haven't got enough caps for this.. not the right ones.
I dont know where the right frequency is.

the only guide I can make is using the 22x 0.22uF caps. and write the numbers down when i go from 0.01uf to 0.22uf

but... damn i have to make sure all else  constant... so i will remove the b emf toroid
and connect straight to igbt

if I connect a ac volt meter is says 400V but that reading is off
the scope shows 18(guess)x50= 900V pp
18 is a guess because i have only 8 blocks of height.

when i connect the white wire (that is feeling hot) the sparks can be seen and heard.  10nF 99khz seems to be the biggest voltage rise, but as the resistance of each capacitor also counts I can be sure.
I will order a set of capacitors to find the right capacitance.

but as the voltage rise is already so high, and the wire is getting hot, I might already be very close.

I tested with a compass.
when connected straight to igbt the resonant coil has a magnetic field that affects the compass.

but when I use the ring toroid there is NO measurable magnetic field seen by the compass.

the igbt gives a pretty clear pulse

the toroid shows a pulse with resonant sine

why is there no magnetic field shown by the compass with the toroid? and it does show up with the igbt?

even with the toroid, when i wind the white wire that gets hot 3 times around the compass the needle is unaffected.

I wish I had a second 10nF capacitor that could take the 900V pp to make another resonant coil.

how did nelson do this? he seems to use normal caps. indicating the voltage rise isnt  that big.
maybe now i have a big dielectric field, and weak magnetic field and should I reverse it.

this means not pulsing with voltage b emf
but pulsing with current b mmf from a capacitor discharge.

charge a capacitor with bmmf from a joulethief circuit, and discharge the cap into the coil. this should create more current.

I wonder if a joulethief can charge a capacitor fast enough with b emf, from a coil, to be able to discharge the capacitor again at the resonant frequency of the coil

to see if this magnetic strong field dielectric weak resonance if formed in the bifilar coil,
I will need to disconnect one side of the 10nF parallel cap. charge it while disconnected with bemf from a joulethief coil (with high current).
then connect the charged 10nF cap back to the bifilar coil and let it discharge.

this should create a ringing between the capacitor and (in phase) capacitance of the bifilar coil, and the magnetic field of the coil. (maybe add some inductance? with ferrite?)

the joulethief that charges the cap, can run on or be stopped. to save energy.
If it runs on, the bemf could be switched back to the battery power source to charge it. but this is more switching... I still only have one igbt. for later study...

7x .22uf inseries measures 31.7nF

with 31.7nF i get a resonance at 57.25kHz.
at 72% duty cycle.

if i calibrate the sine at 8 blocks
with to 10nf, res @101 khz

than with 31nF res@57.25khz
i only get 5.5 blocks.
so... maximum voltage rise is around 100khz with 10nF in parallel to tune.

max voltage at resonance also indicates max current.

so if I us 31nF for 3 resonant coils... current wont be max.

will proceed anyway.

still had 2x3 0.1uf caps. added in series
makes 3x 20nF
resonant voltage rise will be higher (better) and also higher in frequency.

going to try to get out of phase resonant signals from the 2 output coils...

resonant at 72 khz with 20nF
duty is 57% with higher values the current limiter if the igbt kicks in, indicating to big source current.

the added 0.1 uf caps get hot. these are not  like the wima mkt10 made for this. Ill remove them again and work at 30nf

ok back with 31nf (7x .22uf mkt10 in series)
resonant at 57.85 khz @62 duty.

now add the 2 coils on both sides, and ground them so they are opposite phase (if that works