Author Options:

Split Tesla coils anyone?! Answered

I am collecting data in the form of old pics, patents, newpaper and magazine articles as well as just doing things.
Tesla coils are fun but is anyone experimenting with the original "multiplying transformer" that Tesla created?
And before you say, sure I have a Tesla coil......
I mean:
Large diameter primary coil with one part of the secondary in very close proximity while the other half of the secondary is located at the center ;)



25 days ago

I am still waiting for the last few parts to start building my coil winder :(
Today I got lucky on the scrap yard by getting a 5kg roll or copper wire for just 20 bucks.
Although copper is quite soft, this wire is almost a mm thick and made me rethink my winder approach.
Even at coil lenghts of only about 75cm and diameters of max 100mm (for this design anyway), the motor needs to have some torque.
When checking the Nema17's on my printer in a direct drive configuration the torque is really low and certainly not enough to wind 1mm wire onto a 100mm or bigger tube.
Then there is the problem of eliminating bends and kinks along the way without changing the wire tension too much.
For thicker wire a simple clutch break that applies even tension will suffice but for thinner wire this can cause breaks.
Although I don't plan on using hair thin wires I think it might pay off to allow for a motorised and sensor controlled spool unwinding.
After checking various designs here and on other websites it seems that so far all winders are either highly speciliased or dead simple.
The best I found so far was a "4D winder" to make your own carbon fibre pressure vessels in bottle form.
If I invest a considerable amount of time and money to build a machine to a job for me than at least I like this macine to be as versatile as possible.
In the most simple form you might just want a straight winder.
Some however would like to make their own transformer or even a radio coil in basket form.
For me this means the heart of it would be the controller.
Re-inventing the wheel is something NASA and a well know tire company already do but simply snatching some code blocks from CNC or printer solutions is not much easier.
From my early printer experiments I noticed that things like accelleration control can badly affect positioning.
Same for fancy stuff like trying to perferom step counts in sync with CPU timings.
For a long and tightly wound coil you simply can't afford any missed or miscalculated steps for the thing that moves the wire along.
An error of just 2% can mean disaster on a long coil.
And if you are like me and need more than standard then things like variable pitch and winding based on the actual length of the wire can be important.
For your switch mode power supply project you might need to use Litz wire and stick to a certain spacing...
Well, being able to quickly create some fibreglass pipes might come in handy one day too...
The hardware part is quite easy but the coding will be a nightmare.
Especially later things like reversing the direction with a set spacing between the turns.
How exactly do you stop a wire from slipping if there is nothing to hold it when you change direction? ;)
Starting to consider taking some pics along the way so if it works good I can make an Instructable from it.


5 weeks ago

Ok, some more "insights" I made during my coil design tests.

Tuning is a painful process and I now start to revert back to using a HV transformer and Leyden jard together with door knob capacitors.
Just easier to mess with water levels instead of trying to match a coil to the capacitor.
And well, the relations and facts observed during the short time my SSTC driver survived the torture were quite nice.
Although I could not match the coil diameters to the larger size I would have required the rest was matched close enough to confirm the magnifying effects Tesla claimed.
You could see such a tranformer as receiver that has multiple tuned inputs.
What would otherwise be wasted energy is used in a syncronised way instead.
The question reamains what to do with such a system in our modern world?
Even if efficient enough a system for wireless energy transfer over more than about a meter and at usable power levels would not be tolerated.
The generation and distribution of electrical energy isas old as it can get, so those behind would not like to see that most of their transmission lines could be declared a thing of the past.
And unless fully shielded the RF interference would render modern communications and electronics useless.
So either way a widespread use would be impossible.

But perfecting such systems without incvolving lots of modern electronics could open doors for more insight in other useful fields of application.
Like when using fully closed ferrite cores for HF applications, but instead of just using them to shield and increase magnetic flux their design could be optimised to make use of the RF waste it shields from.
Another interesting thing for experiments would be flexible coils.
For a fixed frequency the wire lengths could be perfectly matched but like these coiled air hoses we could stretch it out to change the inductivity and match it to the circuit.
Something we have done on a small scale back in the days or analog radios to tune them down to police bands or up for some airline and airport stuff.
Imagine a self tuning coil that is mechanically tuned...
Last but certainly not least is the understand of how multiple harmonics from different sources create high energy pulses or "humming".
Instead of switch mode power supplies using a single frequency approach it would be possible to use multiple frequencies instead.
All at much lower than actually required power ratings (if we just take their sum) as the main energy comes from the pulses caused by the slight frequency shifts.
Modern capacitor technology makes it all possible as in most cases we need a DC output anyway.
Similar story for motor control applications.
The main input frequencies could be filtered out more or less and the driver optimised to create harmonic pulses at the right frequency for the rotor position and timing.
And funny enough some similar sounding patents already exist for some of these things.
Although mainly by "brute force" methods.

I guess next step is to go a few sizes bigger and to test how much potential there really is for useful amounts of energy transfered over more than just a few meters.


6 weeks ago

Wow you really digged in deep into this matter! My respect! I just did a bit on tesla coils but regarding the resonance issue i realized a setup you might watch for a self resonating tesla device. But beware of getting too much expectations. It's not a huge spark and streamer thrower, just a small humble device.



Reply 6 weeks ago

For me it is not abut big sparks as they are just wasted energy.
We already use wireless charging for ages but I would like to go back to more reach and more efficieny.
Well, the later more in terms of what power can be taken from the system at a distance, not so much to be able to get 100% of the energy out I put in to run it ;)


Reply 6 weeks ago

Due to my experiments there was only a near field where you could get a reasonable amount of energy. My setup was oscillating around 1.2Hz and i could illuminate a CFL up to a distance of about 30cm pretty good while the apparatus was drawing 1A@12V. Then i tested how far i could pick up the 1.2MHz with an AM pocket radio. The signal vanished around 150m distance from the coil. So i couldn't see any long distance effects.


7 weeks ago

So you're asking about concentric secondary tubes?


Reply 7 weeks ago

Hey! I thought you wouldn't be here anymore :)
But no, let me try to explain:
A standard Tesla coil has one primary with just a few turns and one secondary with 700 - 1500 turns.
What I would like to do is to split the secondary coil into two seperate coils.
Resulting in a shorter secondary than the normal type plus another secondary with about the same diamter as the primary coil.
The big secondary is grounded on one side and the other coneects to the second secondary.
If in doubt check the drawing from patent number 119782 belonging to Nikola tesla.
Only difference would be that my second or inner secondary would be more like a barrel and not long as in the patent.


Reply 6 weeks ago

I've known several people with T coils. Don't remember details. I'll try to get in touch with them.


6 weeks ago

Ok, I wasted my remaing spools of magnet wire and experimented.
Turns out tuning can be a bitch if you mess with otherwise simple SSTC designs.
But it also showed that a split system has a great potential - in both meanings ;)
For my tests I just assumed that the drive part can be neglected for sake of saving a lot of wire.
So a simple Slayer type driver was used with a variac to adjust the input voltage.
Only short runs possible before serious overheating happened but enough to establish some baselines from a tiny model.

I might go overboard one day and make an Instructable out of it, but for now I need to work on the upscaling.
What happened so far:
I created a split secondary coil with a loose base on Tesla's patent for the magnifying transformer - No: 1119732.
But I based the dimensions on his notes that he scattered around through previous patents and articles in scientific papers at the time.
To put it into perspective:
Original setup running at 60V AC input: weak and about 8cm long arcs, a CFL would start to glow at about 15cm of distance.
This was using the standard setup as you find for these 15V mini tesla coils.
Using the split secondary coil and again 60V input the power requirement was about 20% lower, the arcs reached up to 14cm but the CFL already started to show a faint glow from over 3m away.

At this stage I assume a properly tuned full scale model of about 60cm in height would be able to produce arcs of over 1m in length - from a standard 240V outlet...
With a stacked set of neon transformers instead of a simple SSTC the max arc length would be mostly limited by the secondary coil and topload design.
Maxing out a 16A, 240V outlet with a 1.6m high setup should produce arcs of over 3m at a quite high power level.
Right now I am torn between going SSTC or investing into some doorknobs and transformers...

Unlike standard Tesla coils I realised that the design intentions behind Tesla's magnifying transformer are indeed based on his work on mechanical resonance.
He invented AC current as a means to eliminate the mass for resonance experiments.
The motor and other things were more or less byproducts along the way to finance his work and put his great ideas into patents that now form the base of our modern life.
With mass gone and only electrically or magnetically related losses to deal with he realised that AC can be utilised for so much more.
When we hear terms like "free energy", "cosmic energies" or just seemingly creating energy from thin air we already think BOGUS, FRAUD, FUNNY BUT USELESS....
But we also love to use the human language in a quite confusing form...
100 years ago a normal solar cell would have been magic, energy from the sun, free energy....
Same for a decent wind turbine in your backyard...
And despite using the same generator concept for water and steam, wind was long considered to be a myth only good enough to pump water from a well...
In todays times we use superconductors which have no resistance if cooled down enough.
Without them a normal CT scan would be impossible....

But what if we could not just reduce electrical losses with normal wires at normal temperatures but also create ways to keep these losses to a level that could be neglected?
Back in the day we all only had iron core transformers available to get down from mains voltage to something in the range below 50V.
Later we got ring transformer to reduce the losses, especially in audio equippment.
Now we basically only use switch mode power supplies.
A fraction of the size and cost of a similar iron core transformer and with efficiency rates above 80% as standard.

Going one step further and including sonic levitation...
We all love to experiment and something really satisfying happens once you place a glass plate above an ultrasonic transducer ad adjust the distance just right.
Suddenly you can make small objects levitate in the nodes of the standing and reflected sound waves.
As Tesla already pointed out: Resonance is the key!
Only if the distance of the glass plate equals a multiple of the wavelenght of the sound we can create these nodes of standing waves.
You can see it like a zones where the pulling and pushing forces are equalised.
If the object goes down a bit it will be pushed up, if goes to high it will be pushed down - until is keeps steady within the node.

Now allow your mind to make the step and to replace sound by electrical currents and electromagnetic as well as electrostatic fields.
We already know and use several ways to reach a close to resonant system when we think of switch mode power supplies alone.
The transformer that does the actual work is designed to work at a set frequency where the losses for the rated power levels are as low as possible.
The key is abandone what we think we learned in some job that involves electricity.
As a Tesla transformer does not use any core it is also a powerful RF generator.
Same for the electrical and magnetic fields.
We are dealing with some quite insane currents once you factor in frequency and how much current decent capacitors can provide....
But for our common Tesla coils the only goal is to produce arcs as long and as powerful as possible - by brute force.
I try to explain my findings so far with the research I did over the past few years:

In a "real" Tesla transformer the windings and design are as important as how the thing is powered to begin with.
Anyone with both an electrical and amateur radio background will have it easy to follow...
Both the primary and secondary system act like an electrical transformer.
But unlike a normal transformer the secondary also feeds back into the primary - a factor that is usually totally neglected as most coil setups are based on close proximity.
So we have a first amplification we are all familar with: Low to high winding ratio and the output must be higher by the factor (in an ideal world).
We also learned that adding more than around 1400 turns on a secondary won't be of any benefit, through that somehow limiting the possible output by electrical transformer action alone.
The topload, togehter with the primary forms a resonant circuit.
And in an ideal world the primary circuit would be closely matched to the resonant frequency of the secondary circuit.
As we want long arcs however the tuning is usually done so that the best resonance is reached with the longest ac length to feed to most possible power into it.
Now here comes the tricky part....
The coils and the coils with the capacitors (topload and mmc bank) also form a RF transmitting and receiving circuit.
Anyone with a decent radio background will know how coils are tuned for specific frequencies...
Now for the "hard core" part, skip if you think you got enough or if you are confused already:

From my personal understanding of Tesla bread crumbs his patent 1119732 is as good and bad as some Leonardos inventions.
Obvious flaws are only obvious if you know how the mechanism is really supposed to work - check his "mechanical tank" ;)
The coils must not only be matched to each other for their inductivity but also in terms of physical dimensions.
The later must match the tuned frequency as close as possible.
This goes as far as matching the wire length to the wavelength and angular velocity of the wave.
And like for example a 500 and 1000Hz sound combine to create a clear tone while just one Hz off would create a pulsing sound, a fully tuned Tesla coil system will do this:
The max levels of electrical waves will be within the coil with a zero crossing is at either end.
The electromagnetic field of the primary will enclose the secondary fully, same for the RF field.
The coils are designed to physically match the desired frequency so that again the zero crossing of the wave would be at the ends of the coil.
The last key to the puzzle is the elevated topload.
Changing the height changes the frequency.
So the final height is a way to adjust to secondary system slightly while the tapping point on the primary gives some adjustment there.
Problem is to keep both systems as closely matched in all factors as possible.
The connection to the topload serves two destinct purposes:
a) It allows for a far higher voltage without risking to get a strike on the primary coil - the elctrical and electromagnetical fields spread the arc away.
b) The connecting rod serves as a conductor to combine the different wavelengths and also to act as an antenna.
As a result the waves are all slightly out of sync when reaching the topload.

Unlike our glass plate for the sound the topload will store the energies that combine right at the top of the long rod.
I try to explain for those willing to follow some basic radio math ;) :
To make it easier I will explude the interactions of primary and secondary in terms of feeding energy back and forth.
If we only take the secondary system then the coil(s) form a full wave configuration in terms of antenna length and dimensions if in resonance with primary.
I might give the real values in a future Instructable but lets just say:
The primary forms a quarter wave antenna and the secondary system a three quarter antenna.
At the topload we combine the amplitudes from the two waves.
At the same harmonic frequency the electrostatic and electromagnetic fields combine at the topload too or better, slightly under it.
Last but not least the electrical wave through the primary coil will combine at another harmonic frequency in the same spot.
If you ever tried a piezo siren with two resonant frequencies (duo-tone) you know what I mean when I say that each harmonic frequency adds to the full "volume".

Why does it matter?
Well, it does not if we trust modern science.
It is just fun to try to get even longer arcs - or is it?
Tesla was a genius, no doubt about it, so he must have been up to something with his latest designs.
Size matters for Tesla coils, both physically and in terms on input power.
If I take dimensions for granted the same way I take for granted that Tesla already knew about the ionosphere before we had a name for it then his goal was clear.
Using the ionosphere as a reflector the same way we used the glass plate for our sound waves.
And with a slight out of focus "break out point" the reflection would not be total but at a slight angle.
Adjust this angle so that the "wave" will travel all around the world by reflection to combin back at the source and you can transmit a lot of energy with next to losses at all.
The input power just needs to be high enough so that a receiving antenna can drain enough power from it, like an oversized crystal radio if you like.
But even without this possible application there is more on a much smaller scale.

Going the full mile without shoes...
Imagine a fully tuned but small sized "magnifying transformer" in the 10kW range.
Portable enough to fit on a small truck with no problem at all.
Now imagine something big that might require power in a lot of places like a big outdoor exhibition or event.
You would only need one big generator but no wires going everywhere and causing all sorts of hazards.
Only receivers wherever power is needed.
And to fully utilize the resonance of such a small system it could be enclosed with a metal "dome" tuned to a harmonic frequency of the system.
Not only would all outgoing RF interference be eliminated but it would also be added to the overall efficiency of the system.
If my lame math is correct than the efficiency should be close to 98%.
Going from previous experiments and old documents I assume that with 10kW input at least 2Kw should be feasable at a distance of up to 200m for output.
More with multiple receivers.
At 500m distance this should be still more than enough for lights and some speakers.

Problems to face and solve:
Every little modification of one component affects the entire system.
If you need to adjust the inductivity of the secondary you migh as will build a new primary and topload instead.
Even if you stick to the established harmonic frequencies, upscaling from a small proof of concept model is a true headache.
Bigger coils mean you need to find the right match between what you can find as a core (like PVC pipe) and the length in relation to wire diameter and spacing.
Not all combos that look good on paper can be realised as it would simply mean you quite often would need touse several kilos of copper wire - with a good chance it won't work anytime soon...
A custom design from wooden spacers and winders might be an alternative but it also makes thing even more complex and time consuming.
So far the biggest problem is physical dimensions.
Unlike a normal Tesla coil with a loose magnetic coupling we now speak in terms of magnetic field strenght in relation to a genrated RF field.
So the losses to consider and the options to compensate are gowing exponentially.
I guess this was one reason why Tesla had to big right from the start.
It is far easier to adjust massive size coils by slight variations of the Leyden jar capacitors and height of the topload.
A little change here won't affect the properties of the coil to act as an antenna as the "frequency range" of a huge coil is a bit bigger than that of a tiny coil.
Example: If the overall wire length of a small primary coil is about 1.2m than a slight change will grastically effect where the nodes of the standing waves appear.
Like sandingoff a mm from your Wifi antenna ;)
But if the wire lenght is over 30m than this would only cause a fraction of the mistune compared to the short wire.

Just to complete the nonsense and for a good bedtime story....
We now for quite some time now that the earth has a magnetic field but also an electrostatic field.
The frequencies for them can looked up calculated I guess.
Now, if the electrostatic field fluctuates or changes like AC in certain areas than an induced outside field of the same frequency would either amplify or reduce these fluctuations.
Again, like sound waves with a phase shift of 0 to 180 degrees.
And changes in the electrostatic field should also relate to changes in the magnetic field.
After all we know how a simple electromotor works...
From vulcanic eruption and the resulting formations we know that he megnetic field has reversed many times throughout the know history of the planet we call home.
Scientists now start to put together the pieces of evidence with the data of thes pole reversals.
So far it seems that around the time of each reversal a more or less severe extiction level event happened in parts of the globe or worldwide.
Including the last ice age.
Now if Tesla only saw wireless communication as a byproduct of his wireless power system, then what about an entirely different end game?
If you check the date and time of Teslas big power up to J.P Morgan you might find that something entirely different happened at the same time in a different part of the world.
I leave it out to you to figure it out but give you hint: It was not a meterite that exploded in the atmosphere ;)
At this stage he must not only realise that his funding was cut off completely but also that the fields he tried to utilise to "amplify" his power output fluctuated.
From maps he already knew the polse wander a bit as really old maps compared to never ones show the corrections for true north over time.
If he however realised that these fluctuations are part of an ongoing cycle resulting a total reversal of the magnetic field?
"Steering" against it by adding energy where needed to stabilise the fields could have literally meant stabilising the entire planet.
Earthquakes and severe weather would be reduced, weather more predictable...
A direct "beam" of energy into a weak section of the field would sort it all out and tracking the waek spots would be the only hurdle.
On a larger scale it could even be used to "overload" the electrostatic field in the event of a severe solar eruption.
Or you could create your own northern lights in any part of the world...
But of course this is only based on assumptions and the fact that Tesla notes indicate he planned on publishing much more patents "once the time was right".
He already knew as a kid that he will build a massive power plant in America, he could create entire designs in his head and only once satisfied with the results he would sometimes put them on paper.
Where we use a computer simulation he used his phographic memory and his mind.
No calculator either....
If you know your electrical stuff from head to toe than try to fully create and test a 1kW switch mode power supply with adjustable output in your head.
Ok done? Now build it.
Nice work, now confirm it works and then maybe decide to create circuit diagrams and parts list....
We just can't do what Tesla and some othe great people with a photographic memory can do :(
But he also could do most complex math in his head like some of these wiz kids on Youtube and from TV shows.
How far did he understand the fields of our planet and how they affect us, or how far was he willing to go to prove a point?
If you by now found what happened during his biggest demonstration in another part of the world you might agree that there is more to know than what history provided us.
Edison was finally pushed from his throne once the US admitted he was only a ood salesman but ed the ideas of other people.
When will we get the full and unedited truth about Nikola Tesla?


7 weeks ago

There wouldn't be an easy to use simulator for electromagnetic coils available?
Something as simple as entering coil diameter, wire spacing, number of turns/layers so you get a visualisation of the magnetic field?