Ok! Let's try to summarize from all my previous posts elsewhere, about the specific setup I have been testing, which is shown in the recent YouTube content I've uploaded - "N Tesla Single-Wire (pt. II)" which was uploaded 6 days ago (March 21, 2024):

https://youtu.be/YYXr6vgef04

Also, here's the "Part I" video showing the same setup (posted on YouTube in November 2023):

https://youtu.be/Gpv_eNBdM74

This system has a 2-turn primary (unifilar) of 12awg (3.3 mm²) copper speaker wire with both channels shorted together, wound around the perimeter of a 42 turn secondary (bifilar) of 16awg (1.3 mm²) FLAT copper stranded wire. Both step-up Transmitter (TX) and step-down Receiver (RX) coils are encased in epoxy, fixed to an acrylic sheet.

Anyway, the step-up TX transformer feeds its HF HV output into a single 18gauge (0.8mm²) wire of ~4m, which feeds into the step-down RX coil: into the center of the 42-turn high-voltage 'primary' (grounded at the outside), and the 2-turn low-voltage 'secondary' is tapped for output.

This output goes into a bridge rectifier, and charges up my 27V 10F Maxwell graphene Ultracapacitor assembly. From the ultracap, I power the DC load - an automotive radiator fan, with the fan blades removed and an acrylic disc (or stack of 3 discs rather) in its place.

Running the load from the capacitors is what allows me to power 'inductive' loads, whereas any such load throws the system out of resonance if run straight from the rectifier.

Also, for simplicity I am referring to my impulse coil (aka self-induction coil, aka kicking coil) as L0 (as opposed to L1 like Ivo does, and like I'd previously done). This is just so that I can call my primary coil "L1" and my secondary coil "L2".  In the next post I'll talk more about L0.


 

ON THE SENDER/TRANSMITTER SIDE:
I'm using Ivo's oscillator circuit, which uses two series Silicon Carbide MOSFETS, as a high-side switch, controlled by a function generator's square wave - which opens and closes the MOSFETS. When 'open', the (+)  from my DC power supply is conducted through my 'L0' kicking/self-induction coil, to ground. When 'closed', the circuit is broken, which results in a Back-EMF inductive spike, i.e. a negative voltage impulse from my L0, which is conducted into my Primary Coil L1. Thus, the primary coil is excited by negative voltage impulses, at a rate determined by the frequency of the function generator.

The primary coil has a 'switchboard' of tuning capacitors, connected in series, giving 'series resonance' in the primary. The secondary coil has capacitors connected in parallel, for 'parallel resonance' in the secondary. The Primary Coil L1's series capacitance has a value of 103nF. The Secondary Coil's parallel capacitance has a value of 4nF.

*I'd also been using an "L3" Extra Coil in series after the secondary', but since i started working with this 'flat/interleaved' bifilar pancake secondary, I can't yet seem to find the right geometry/type of extra coil which boosts the voltage or gives any advantage - though with the previous varieties of 'stacked/vertical' bifilar secondary, i COULD get an extra coil to produce such benefits.

Regarding L0, I have used a variety of coils to produce the 'coil discharges' which excite the primary L1. Sometimes I use bifilar pancake coils, but sometimes I alternately use the shielding conductor of a coiled-up length of RG6 coaxial cable (25ft I believe), which is what I'm using in the videos, as pictured:

Robert Handwerker - LMD waves in a Tesla Coil revised

AMInnovations LMD Transference of Electric Power

Advanced Energetics for Aeronautical Applications vol II

In recent months, I have been working on an improved system for 'single wire power transmission' (resonant, high-frequency). The previous system (shown in previous posts, above) used 'bifilar pancake' tesla coil secondaries. With the 'bifilar pancake' system, I was able to achieve a transmission efficiency of (only) about 40%. I asked a gentleman 'Hakasays' (YouTube & EnergeticForum handle) about improving the efficiency, and he recommended I make an attempt using 'open frame helical' Tesla coils, such as those favored by Eric Dollard and Adrian Marsh (AMInnovations). So I have built up a system with such coils, which has been designed much more carefully than the previous 'bifilar pancake' coils. I am still using Master Ivo's Tesla Oscillator circuit (2021 version, the immediate successor of the 'Radiant Half Bridge'), and it works quite well even with this different system configuration.

*note: inspiration also came from a great video by Hakasays, called Build a Tesla Coil (the way Tesla built them).

My earlier 'pancake' secondary coils were basically constructed on the basis of: 'how much copper wire do I have available, what is the biggest coil I can manage, and what sort of 'freestyle' primary coil can I construct to excite the secondary?'. With this current system of helical coils, on the contrary, I have applied 'optimized' spacing between windings (rule of thumb being at least 160% of the wire diameter, is the center-to-center spacing between windings) using 3D printed 'winding spacer' pieces attached to the frame, I have constructed the primary coils to be of approximately equivalent copper mass to the secondary coils, and I have independently tuned the primary coil 'tank circuit' to resonate at the same frequency as the secondary coil's resonant frequency. These steps were taken to try to achieve the highest Quality Factor (Q) for the tesla coil (primary and secondary together). I have made a 'matched set' of 2 coils, for use as 'transmitter' and 'receiver' coils in a Single-Wire Power Transmission setup.

The 'frame material' with the lowest associated 'dielectric loss' turns out to be UHMW (an extra dense version of HDPE high density polyethylene), and so that is what I have build the frames with. Long story short, early rough draft coils I made using plywood had Q factors in the low 40's - but with these UHMW frames have Q factors of 170. This is an admirable value for Q, I gather.

So here's a picture of the final product, and next I will list some more specifics about my particular 'open frame helical coils', regarding dimensions and design variables etc. By the way, I will tend to default to metric units.



Wire Specs:
Enameled Copper Magnet Wire, size = 20AWG (0.52 mm²), diameter = 0.82 mm.



Frame Specs:
The frame's 8 vertical pieces were cut into pieces of 40.5 cm (16") height and 9 cm (3.5") width. The 2 circular pieces have diameter of 28 cm (11").

*note: I had enough UHMW (white) for everything except for 2 of the circular pieces, which are made of acrylic (with blue protective wrap). This did not apparently affect the Q factor.



Winding Spacers:
Each coil has a total of 100 turns of wire. To space the turns/windings, each of the frame's 8 vertical pieces has two 3D printed 'winding spacers' attachments, each with 50 'teeth' (since my 3D printer wasn't big enough to print all 100 'teeth' as single pieces). So each attachment of 50 teeth is 12.5cm (5"); thus each coil's 'height' is 25 cm (10"). Therefore, the spacing from center-to-center of each winding is 2.5 mm.

*note: This amounts to 305% spacing between windings, center-to-center, which is more spacing than strictly necessary - the rule of thumb being 160% - although printing pieces with closer spacing might be a challenge for my (cheap) 3D printer.



Overall Dimensions:
Everything assembled, the effective circumference of the coil is 87 cm (34"), so the effective diameter is 28 cm (11"). The coil height being 25.5 cm (10"), the height-to-diameter ratio is slightly less than 1:1 (i.e. 0.9).

Total wire length calculated as 285 ft (87 m). This 20awg magnet wire has 314 ft in a pound, so the total weight is calculated at 0.91 pounds (0.41 kg) including the enamel coating. For bare copper wire, specs for 20awg says 0.0046 kg/m, thus a total copper mass of 0.40 kg (0.88 lbs).

The resonant frequency of the secondary coil alone (separated from primary, and without topload) is right around 1,100 kc/sec. Next I will describe some the way I am measuring Q as well as finding/tuning the primary and secondary coil's resonant frequencies (using a VNA, Field Strength Meter, oscilloscope, and function generator.

Lately been learning how to use a vector network analyzer (DG8SAQ model), as well as a RF Field Strength Meter, for tuning the coils and for measuring the Q factor of the coils (individually and as a system).

There's a HUGE amount of theory involved which I am just beginning to grasp, so I do not explain things very well in this video, just demonstrate the way I have been using the new tools, as I learn them.

https://youtu.be/mn2K-K-9es8

In order to tune a Tesla Coil 'primary' so it resonates with secondary, an adjustable capacitance is more-or-less necessary. Master Ivo introduced me to a great way to achieve this, i.e. his several prototype circuits featuring Wima FKP1 high voltage PP film capacitors, on perf-board, and with toggle switches to connect/remove each cap from the big parallel lumped circuit (for series/parallel resonant coil tuning).

This enables a convenient 'variable capacitor'. However, it's a lot of point-to-point soldering to connect the switches to the capacitor leads - so i made my first ever PCB design for building just such a capacitor switchboard. It lets one switch 0.1nF increments, from 0nF to 11.0nF; it features Wima FKP1 2000VDC/700VAC film capacitors. FYI, the necessary C for my present Tesla Coil 'primary' tank circuit is 5-6nF. Anyway, the pcb 'gerber file' (zip) is linked below.

PCB Gerber File

I used 10x 'Wima FKP1U011004D' 1000pF caps [15mm pin spacing], 10x 'Wima 'FKP1U001004B' 100pF caps [15mm pin spacing], and 20x 'Adam Tech SW-T3-1A-A-A3-S1' SPST mini toggle switches [2.5mm pin spacing].



This will be useful with my present SSTC setup using master ivo's circuit, but I also built a new SSTC driver circuit following a design by 'LabCoatz' - a Class E 'Musical' Single-FET SSTC. I like the circuit since it has no feedback or automatic tuning; I like being able to manually control the frequency and it has a built-in adjustable-frequency oscillator. Anyway, it has a 'resonant capacitor' slot for tuning the circuit (it's placed across the MOSFET, not across the primary like might be expected, needs testing). So the variable capacitor PCB will be perfect to fine tune this circuit to achieve ZVS/ZCS and keep the FET from getting hot. This PCB is blue.

Also, I've been using Master Ivo's high-side MOSFET switch for a couple years now - all with the same circuit featuring 2 Wolfspeed C3M SiC MOSFETS in series, with 2 gate drivers and 2 BNC inputs etc. But lately Ivo had put out an updated streamlined version, with a single MOSFET on a smaller PCB but otherwise with all the same components as previously used. I built it and am nearly ready to fire it up for its first test drive. This PCB is black.