Tesla's Bifilar Pancake Coils

[7imix]

A capacitor is needed to resonate with it at the right frequency to achieve the desired effect.

Tesla talks about this in the patent. Does anyone have any idea how to calculate this? I searched google without much luck. Don Smith talks about measuring the capacitance of the secondary with an LCR meter, subtracting the capacitance of the primary, and using that.

However, in the patent, I believe Tesla is referring to adjusting the turn spacing to give the coil the required capacitance without requiring an external capacitor.

Thanks for any info!

[gyula]

Hmmm, actually no, I was going to wind it with no additional spacing between wires. Didn't even think of using fishing line...thanks for the idea!

I was originally going to order 42" diameter cercal glass which would have left me extra room anyway...I still have time to up that to a 48" diameter, but will probably need to jump from 1/2" thick to 3/4" thick for stability.

Forgive my ignorance...I come from the world of small digital circuits so these are new concepts at this scale. Is the spacing used to act as additional insulation between turns resulting in lower capacitance? Or is it the physical spacing/separation that lowers the capacitance?

I welcome (and appreciate) any and all ideas at this point!! Once this sucker is made, going back and making changes is a non-starter. It will have to wait for Rev II. ...at some point.

Hi,

The capacitance between the parallel guided wires is created by the facing wires surface areas, the dielectric constant of the insulation of the wires and the physical distance between the wires. How these 3 factors influence the capacitance turns out from the formula used for the parallel plate capacitor, see here for instance:
Parallel Plate Capacitor Capacitance Calculator

It comes that the higher the surface area of the parallel wires and the smaller the distance between them, the higher the capacitance you can achieve. You would consider using wire with rectangular cross section so that much higher surfaces could face each other instead of the usual circular cross section (circular wire is the worst case in this respect).
The above link includes dielectric constants of some insulating material and remember the thickness of this should be the minimum possible.

There are some audio cables of thin diameter like microfon cables that have one (or two) inner insulated wires in the center and wrapped into an outer braid that normally is the ground in audio systems, I think if you have a cheap source for such cables (NOT speaker wire) you could test it because its outer braid is also insulated so that you can make a pancake with a single such cable and have ready the two wires guided parallel and insulated. Coaxial cable like RG-174 also has a small diameter but maybe more expensive than an audio microfon cable. Such cables are characterised by a distributed capacitance between their wires by the length in picoFarad/foot or picoFarad/meter.

I have not tested but I think that the resonance could be calculated from the normal Thomson resonant LC formula where C is the distributed capacitance, measured between the parallel wires by a simple capacitance meter and L is the resultant coil inductance, measured between the wire start A and wire end B, as designated in the Tesla patent. Or, knowing only the distributed capacitance the resonant frequency can be found by measurement like in this video: YouTube - hydro4f3a 017 Pancake resonance
This guy, hydro4f3a has several videos on pancake coils worth watching.
Here is another useful video on testing pancake resonances:
YouTube - Various Coils (Pancake Coil Resonance) 002

This answers 7imix questions on the a capacitance too.

rgds, Gyula

[7imix]

I have not tested but I think that the resonance could be calculated from the normal Thomson resonant LC formula where C is the distributed capacitance, measured between the parallel wires by a simple capacitance meter and L is the resultant coil inductance, measured between the wire start A and wire end B, as designated in the Tesla patent. Or, knowing only the distributed capacitance the resonant frequency can be found by measurement like in this video: YouTube - hydro4f3a 017 Pancake resonance
This guy, hydro4f3a has several videos on pancake coils worth watching.
Here is another useful video on testing pancake resonances:
YouTube - Various Coils (Pancake Coil Resonance) 002

This answers 7imix questions on the a capacitance too.

rgds, Gyula

The key word "Thomson" is what I needed to get google to give me the information. Thank you.

For example,

Resonant circuits

[gyula]

The key word "Thomson" is what I needed to get google to give me the information. Thank you.

For example,

Resonant circuits

Yes, that is ok. Here is an online calculator.

Resonant Frequency Calculator

[Mike2kt]

Gyula - thank you for all the great info!

There are some audio cables of thin diameter like microfon cables that have one (or two) inner insulated wires in the center and wrapped into an outer braid that normally is the ground in audio systems, I think if you have a cheap source for such cables (NOT speaker wire) you could test it because its outer braid is also insulated so that you can make a pancake with a single such cable and have ready the two wires guided parallel and insulated.

Microphone cable you say? And as if being a packrat is such a bad thing...



pair of 22 AWG, shielded, about 500' left on the roll.

[gyula]

Well, The shield would be one of the "wires" and inner center (with red PVC insulation) wire the other one.
Maybe you have some specification on it like capacitance/foot?

[Mike2kt]

Spect sheet:

http://www.westpenn-wpw.com/pdfs/cm_PDF/291.pdf

[gyula]

Thanks for the link.

Well, the shield is Alu Polyester foil and I wonder how can you connect any copper wire to the Alu foil when you wish to make contacts at the cable ends, one thing is sure it is not solderable?
It has about 99pF/foot capacitance between the Alu foil and the inner conductors, If you connect the red and black inner conductors together, their surface area increases and this 99pF/foot will be nearly double.
(Connection: at one cable end you connect red and black to each other and at the other cable end you do the same, so you will have a paralleled red and black conductor as a single thicker insulated wire.)

[Mike2kt]

Well, the shield is Alu Polyester foil and I wonder how can you connect any copper wire to the Alu foil when you wish to make contacts at the cable ends, one thing is sure it is not solderable?

The foil is coupled with a (22-24 AWG equivalent) stranded tinned copper drain. The drain gets soldered..just as it would if I was connecting a XLR microphone jack to the end.

[Mike2kt]

Well two Netflix movies later and a bunch of staples...60 turn coil made from the mic wire. I figured why not - had the wire and a piece of plywood and staples are cheep. Please no wise-cracks on the construction...it's just a test concept. From the coils' center edge to its outer edge is 235mm. center is 120mm...so from the outer edges it's ~590mm.

Just got done with it, so haven't taken any measurements yet. I have a Fluke 189 multimeter and current clamp on hand...so I can take some basic measurements. What would you all like to see (please be specific).




----
60 turns of the cable...so 120 effective turns if it's setup as a bifilar coil. If my understanding is correct that is. lol

[7imix]

Well two Netflix movies later and a bunch of staples...60 turn coil made from the mic wire. I figured why not - had the wire and a piece of plywood and staples are cheep. Please no wise-cracks on the construction...it's just a test concept. From the coils' center edge to its outer edge is 235mm. center is 120mm...so from the outer edges it's ~590mm.

Just got done with it, so haven't taken any measurements yet. I have a Fluke 189 multimeter and current clamp on hand...so I can take some basic measurements. What would you all like to see (please be specific).




----
60 turns of the cable...so 120 effective turns if it's setup as a bifilar coil. If my understanding is correct that is. lol

Awesome. Love the construction technique. Going to build one like this myself.

[Mike2kt]

Awesome. Love the construction technique. Going to build one like this myself.

I really need to take a higher resolution photo of it...the spacing and tension came out fantastic! That picture doesn't do it justice. Hardest part is learning how to staple two cables down under one staple without piercing the outer insulation. Used a 3/4" (aka 23/32") 2'x2' plywood with T20 3/8" staples. The T20/T25s are for cable fastening...a T50 staple/gun will just rip into the cable.

I only used plywood because the cable had insulation...I still plan to use glass on my 28 AWG magnet wire coil. I think I scared the last glass place...time to find a new one, sigh.

[gyula]

Hi Mike,

Nice job! You may wish to measure the self capacitance between the shield end and the joined inner wires for I think it is beneficial to connect the two inner wires in parallel and treat it as a single thick inner wire as I wrote earlier so that their surface area nearly doubles, this means higher self capacitance to get and also reduces copper (ohmic) loss.

Also you may wish to measure the DC resistance between the shield ends and between the paralleled inner wires end. You have a very nice DMM, Fluke189!

If you could borrow an inductance meter too that would be great to check the self inductance of the shield wire in itself and the inner wires in itself, (these two should nearly be the same inductance value) then the total inductance when they are connected in series as shown in Fig. 2 of the patent.

If you cannot obtain an L meter, then, unfortunately an oscilloscope and a variable frequency signal or pulse generator would be needed to find the pancake coil self resonance I am afraid. Unfortunately I do not know yet how the number of turns may change the self inductance (I mean here either the shield wire or the inner wire self inductance, not the resultant inductance of the two when connected in series. The series connection will probably increase the individual inductances by nearly 4 times, as normally bifilarly connected parallel guided coils do. So to learn about the self inductance, I think the formulas given in the links earlier for pancake coils should give a good approximation.

Will ponder later on how to find self resonance by other means. Maybe the videos I linked to wrt pancake coils may give some more hints. But they use oscilloscope also...
The signal or pulse generator is always a useful 'gadget' in these tinkerings anyway, it is worth building such, perhaps a LMC555 or TLC555 timer IC based pulse generator is the easiest to build.

rgds, Gyula

[Mike2kt]

self capacitance between the shield end and the joined inner wires

0.04uF

Also you may wish to measure the DC resistance between the shield ends and between the paralleled inner wires end.

Shield 4.7 ohm
Inner: 1.9 ohm

[Mike2kt]

OK...here is a question...

Say you have one larger (say 38" dia) bifilar pancake coil with a total wire length of 2n.

Now, what if you have TWO medium sized (say 22" dia) bifilar pancake coils each with a wire length of 1n.


If you wire the two 1n coils in parallel and place them side by side, NOT stacked on one another, giving a wire length of 2n:

A) The performance will be about the same as the single coil
B) The performance will be much lower with multiple coils even if the total wire length is the same


I ask because it's a lot easier to get my hands on smaller sizes of glass.