Sciencemadness Discussion Board

Tesla Coil Inquiry

TheAlchemistPirate - 29-8-2015 at 10:02

Haven't posted here in a while...
Recently I have been learning about electronics and radio technology, and other than my whims of building a long range quadcopter that piggybacks the cell tower network, I have been planning to build a tesla coil. I wasn't sure if this forum was the type that would have much experience with these things (no offense), but I might as well see.

http://makezine.com/projects/make-35/six-pack-tesla-coil/
This is the tutorial I'm planning on following, and something doesn't make sense to me in this schematic from it-
https://i2.wp.com/cdn.makezine.com/uploads/2013/05/coil-sche...
The charging circuit info at the top left says it operates at 20-30 milliamps, but the primary circuit says it peaks at 100-200 amps! I figured that the capacitors discharging would raise the current to the few amps temporarily, but that number seems insane. It makes more sense in the top right secondary circuit info.

Another question, the recommended power source is a 10 kv Neon sign transformer, using a terry filter which I assume keeps it from exploding. It implies that the NST only supplies 20-30 milliamps. I was wondering if I could substitute something like this.
https://www.rmcybernetics.com/shop/cyber-circuits/pulse-modu...
It claims to be able to supply 9 amps continuously, along with several hundred volts. The most intriguing feature to me is the frequency adjuster, which can range from 0.01 hertz to 1.5 MHz! I would pair this with an ignition coil(presumably at 60 hertz) to achieve 10 kv. Could this replace an NST?

Lastly, radios operate by transmitting rapidly pulsing electromagnetic waves, with different ways of encoding the data into the frequencies. When a radio first receives a radio signal, it is a tiny amount of current, which can only be directly listened to with a crystal radio. This tiny amount of current is then amplified with transistors. While it is a very small amount, radio transmitters are providing wireless energy! My question is this- If tesla coils are among the most efficient transformers/transmitters, why aren't hobbyists tuning a large coil and a small coil to the same frequencies, then efficiently and wirelessly transmitting power miles away?! Tesla coils were one used in radio systems, why wouldn't that still be possible?

WGTR - 29-8-2015 at 14:37

Quote: Originally posted by TheAlchemistPirate  

Recently I have been learning about electronics and radio technology, and other than my whims of building a long range quadcopter that piggybacks the cell tower network...


If you ever do this, I'd like to see the details.

Quote:
...something doesn't make sense to me in this schematic from it- https://i2.wp.com/cdn.makezine.com/uploads/2013/05/coil-sche...
The charging circuit info at the top left says it operates at 20-30 milliamps, but the primary circuit says it peaks at 100-200 amps! I figured that the capacitors discharging would raise the current to the few amps temporarily, but that number seems insane. It makes more sense in the top right secondary circuit info.


This can be shown using complex algebra.

In any parallel resonant circuit that uses high "Q" inductors and capacitors, the circulating currents can be extremely high, even when the source driving it only supplies low currents.

The capacitive and inductive currents can be represented by imaginary numbers, and it just so happens that a tuned circuit at resonance has capacitive and inductive reactances that are both equal and opposite. The reactive currents cancel out, even though they may be extremely high inside the tuned circuit. In a parallel tuned circuit, the only energy that needs to be supplied is equal to the resistive losses (which are hopefully minimal).

In this particular circuit, the primary circuit is series tuned, and is being driven by a 60Hz signal. The impedance of the "six pack capacitor" at this frequency is extremely high (≈R - 500,000j), hence the currents are very low (≈R + 0.020j) during charging. When the spark gap fires, the tuned circuit oscillates for a short period of time at its natural frequency, which is much higher, around 400-450kHz. At this point the peak currents within the tuned circuit can rise very high for a brief period of time, apparently 100-200A from the schematic. This approximately checks out, since the impedance of the capacitor is ≈ R - 70j. The power supply would not see anything near this level of current coming out of it, provided that it is protected with a filter.

The "terry filter" shown in the schematic has wires going to nowhere. I'd suggest looking for a better schematic.

This is a very glossed-over explanation, but hopefully it explains things a little.


aga - 29-8-2015 at 14:49

Quote: Originally posted by WGTR  
If you ever do this, I'd like to see the details.

Probably means using a SIM card using 3G to get the data connection.

4G would not be much use outside a city.

[Edited on 30-8-2015 by aga]

TheAlchemistPirate - 29-8-2015 at 19:00

...
I had written a massive wall of text detailing how my cell tower-comandeering quadcopter would work, only to click "BB Code" to get some freaking italics and delete the whole thing.
Perhaps sometime in the future I will acquire a great wealth of motivation and rewrite the whole novel, on the offchance someone may actually understand it. My head hurts now and I haven't slept in days...

Pretend you have just memorized 57 electronics data sheets, and now have a rather severe aneurism from it. Maybe channel your memories of being in an electrical engineering class, after you had stayed up all night putting a university patrol car atop the massive dome building on the campus. I don't know.

Back to the tesla coil- I think I know what you mean now. So for a very brief moment an extremely efficient connection is formed in the tank circuit and hundreds of amps can flow freely through the spark gap into the primary, only for the power to immediately plummet to near 0. If you could measure the spark gap amperage with an ocilloscope wouldn't it be very jagged lines that are like extremely thin recatngles? I had more things to say, but I seem to have lost all of my higher-level mental abilities after that incident.
EDIT: I have no idea what's going on with these columns, maybe they're trying to hold up my wall of text at the top. I'm too tired to care.

[Edited on 30-8-2015 by TheAlchemistPirate]

IrC - 29-8-2015 at 20:44

Quote: Originally posted by WGTR  
The "terry filter" shown in the schematic has wires going to nowhere. I'd suggest looking for a better schematic.


If you carefully read the article you will see a link to the corrected schematic.

To the OP: Why bother with this design, throwing darts at a wall of numbers would end up giving better design values. I suggest you search the web and find a better secondary design. The diameter is too small and there are far too many turns of too fine a wire. The secondary self capacitance would load it so hard there will be nothing near the resonant rise one could achieve with a heavier gauge wire, taller coil form, and increased diameter. Not to mention the top load is too large with too great a self capacitance for the power level of this coil meaning far longer sparks could result if one were to compromise here by not trying to store so much energy in such a heavily damped low power design.

I did not see where they mentioned the Cl2 that will be bubbling out of the capacitors. Not to mention the tendency of that type glass to crack if one were ever to get a decent circulating current in the primary. Just find the correct value and order a proper capacitor with at least a 30 KV rating. The 1K resistors are going to severely limit charging current, by far too much. An experienced coil builder could write a book on all the reasons that design sucks. There are many sites with far better information and plans. One fairly decent site is linked in the article if you read it carefully. The corrected schematic will show you the proper filter circuit and it does have some features worth considering although I would make several changes.

IrC - 29-8-2015 at 20:46

How did my reply end up long thin and joined to aga's post?

aga - 30-8-2015 at 00:12

Sorry. That was me missing a bracket in my last post.

TheAlchemistPirate - 30-8-2015 at 08:10

After reading through the instructions again, I saw the answer to one of my questions. When talking about the power supply, it states "solid state voltage converters are not suitable for this application". So I would have to use a transformer instead of a dc power supply like the one I mentioned above? What is the difference between them if they both provide 25 mA at 10000v and 60hz?

I'll take your word for it and find another tutorial. I just want to begin with a simple(though not necessarily small) design. Also, I cannot for the life of me find GF protection-free NSTs anywhere. About the "six pack" capacitor bank(which are essentially Leyden jars), I don't think that the idea of using them is dumb as a whole. I also wondered about the potential of the jars to electrolyse the salt into sodium and chlorine, but figured it wouldn't happen if there wasn't much leakage. Would sodium bicarbonate work to lower the resistance of water? I really don't want to deal with trying to line the inside of a container with foil, or spending lots of $$$$ on fancy capacitors. I doubted the suitability of as an insulator too, after reading a very technical page about it liking to absorb current between the plates.

I'm rather awful at understanding these mathematical/algebraic equations people seem to pull out of thin air(not implying they're false), so I was wanting to start with something simple and possibly substitute parts with javaTC. While I haven't bought anything yet, I would also prefer the design to be use the variable power supply, if possible. How about this tutorial? http://www.instructables.com/id/How-to-build-a-Tesla-Coil/?A...

TheAlchemistPirate - 30-8-2015 at 08:11

After reading through the instructions again, I saw the answer to one of my questions. When talking about the power supply, it states "solid state voltage converters are not suitable for this application". So I would have to use a transformer instead of a dc power supply like the one I mentioned above? What is the difference between them if they both provide 25 mA at 10000v and 60hz?

I'll take your word for it and find another tutorial. I just want to begin with a simple(though not necessarily small) design. Also, I cannot for the life of me find GF protection-free NSTs anywhere. About the "six pack" capacitor bank(which are essentially Leyden jars), I don't think that the idea of using them is dumb as a whole. I also wondered about the potential of the jars to electrolyse the salt into sodium and chlorine, but figured it wouldn't happen if there wasn't much leakage. Would sodium bicarbonate work to lower the resistance of water? I really don't want to deal with trying to line the inside of a container with foil, or spending lots of $$$$ on fancy capacitors. I doubted the suitability of as an insulator too, after reading a very technical page about it liking to absorb current between the plates.

I'm rather awful at understanding these mathematical/algebraic equations people seem to pull out of thin air(not implying they're false), so I was wanting to start with something simple and possibly substitute parts with javaTC. While I haven't bought anything yet, I would also prefer the design to be use the variable power supply, if possible. How about this tutorial? http://www.instructables.com/id/How-to-build-a-Tesla-Coil/?A...

EDIT: for some reason the post copied itself

[Edited on 30-8-2015 by TheAlchemistPirate]

IrC - 30-8-2015 at 08:43

What is the difference between them if they both provide 25 mA at 10000v and 60hz?

Maybe one of them is not all that concerned with spikes blowing semiconductors. Go to google type in tesla coil and start reading. There are so many good sites out there you will have no problem finding a design which is within your reach and hopefully not excessively life threatening. NST's are great after you are familiar with the safety aspects. Start with some smaller design that operates on say 12 volts, like a 555/power mosfet driving an old car ignition coil as your high voltage source to drive the primary circuit of a small coil design. After you have the learning experience of getting the crap shocked out of you a few times in a fairly non lethal setup you will have a more refined sense of caution when you graduate to dangerous designs. Maybe with these smaller designs even your liquid/glass capacitor is a reasonable approach in the interest of staying low budget. Eventually you will need to find and invest in some decent capacitors but until you know the parameters needed there is no point in worrying about that yet. Also study stun gun circuits of the spark gap type as this is actually not a bad source of primary power in a small coil.

Just keep in mind with little knowledge in high voltage and electronics it isn't all that wise to begin your studies by playing with lethal voltages and currents.

TheAlchemistPirate - 30-8-2015 at 09:52

I still don't understand why power pulse modulators can't be used in place of normal transformers, is it because the negative terminal receives huge voltages that it wasn't designed for? I am now looking into solid state coils since they use lower currents and don't require hard to find transformers. They all have massive circuit schematics however. I'm not looking to produce 2 foot arcs right now, I just want to actually understand how the circuit works and why each component is used. There are so many terms on java TC and such that I have never heard of. I am currently doing all the projects in the book Make:electronics but only understand basic electrical math and terms. The projects are on hold for now however since I accidentally fried all my 2n2222 transistors. The most important thing to me is to understand exactly why everything works, then I can begin my plans to rule the world.

The stun gun circuit looks interesting, and seems similar to some solid state schematics I have seen. Something that feels surreal to me in electronics is seeing all these same components over and over again, like the 555 chip, 14007 diode, 2n2222 transistor, it creates the illusion that I know what I'm doing :) . I wonder how long the stun gun circuit can run, maybe a long time if it is ventilated... So, if a blundering sixteen year old were to replace a stun gun button with a switch, replace the leads with a primary winding, and throw a wire-wrapped pvc pipe with a toroid on top(from a failed sstc kit from christmas-soldering sucks long story short), into the middle of it, how would it go down? What if that sixteen year old managed to calculate the general size of the required primary and secondary coils?

EDIT: I saw a project on instructables where a kid threw together a 12v power source, which went to a makeshift 555 timer which powered the leyden jar and spark gap. He indiscriminately wound a primary and secondary, and attached some sort of metal ball to the top. At first, the high voltage current went to the negative terminal of the timer circuit and blew up the 555 of course. But, to fix that, he attached a photocoupler! This kept the high voltages from affecting the rest of the circuit and still provided a ground for it. What if I used a photocoupler on the negative terminal of the power pulse modulator? Then I could drive a circuit that would normally require an NST with an accurate and multi-use power supply.



[Edited on 30-8-2015 by TheAlchemistPirate]

aga - 30-8-2015 at 12:19

What is it that you would like to actually Do TheAlchemistPirate ?

Most of this thread is really confusing, so just lay it out flat, please.

TheAlchemistPirate - 30-8-2015 at 12:34

To be honest, I have simply been writing what I was thinking about, and my mind is pretty chaotic sometimes. I just want to build a tesla coil, and understand it. Maybe this would be more suitable on an electronics forum. That's about it.

aga - 30-8-2015 at 12:51

You need to drink more.

Anyway, the Tesla thing is quite simple : you got 230 0r 110 V AC coming in.

Either recitfy it or just chop it up raw.

Feed it into a big-ass coil of copper pipe after it is chopped up.

This is called the Primary coil.

The Secondary coil is usually wound on a PVC pipe using much finer enamelled copper wire, around 32swg size, and LOTS of turns.

What is happening is that the electrical energy is being fed into the Primary coil, creating an Electromagnetic field.

The Secondary coil of wire picks this energy up and amplifies the Voltage (consequently reducing the available Current, as the Energy remains constant).

It does that by having a lot more turns of copper than the Peimary coil has.

2N2222 is ancient. Look for HEXFETS instead (if using DC) or Triacs if chopping up AC.

http://www.teslacoildesign.com/

WGTR - 30-8-2015 at 19:40

I'm not sure what you're referring to with the photocoupler. I'd have to see a schematic of it. IrC has the right idea, of course, about starting small. I second the idea of using an automotive ignition coil (the round cylindrical ones). Those are available at any auto parts store and salvage yard. They normally have a 1:100 voltage step up ratio, and are made to switch up to 1kHz or so. Don't plug this directly into the wall. It needs a MOSFET to switch it in a flyback converter configuration. I'm using an MSD 8202 in a flyback converter, and it works well. I've gotten up to about 30,000V from of it without much trouble.

As an aside, high voltage sparks can do funny things to circuits. When high voltage breaks down an air gap, it can do so very rapidly, delivering nanosecond pulses to whatever it hits. If the air gap is discharging a capacitor, it's possible to deliver 1,000's of amps in that brief period of time. If the spark hits a ground plane, it's possible for the ground to "bounce" briefly, destroying nearby components. That may be why they are saying not to use solid state converters. If care is taken, though, it should be possible.

TheAlchemistPirate - 30-8-2015 at 20:38

It was a rather shoddily built design that used the photocoupler, I'm just going to forget about that for now. I feel extremely overwhelmed right now with all of these solid state designs, so I'm welcome to anything that I don't need an electrical engineering degree to understand. I have read many a legend of these magical ignition coils that are the key to all that is high voltage and dangerous. This may sound dumb but is a flyback converter similar to a flyback transformer? I considered using one of those with a solid state power supply but I have no idea how it works...

I would simply use a transformer and make a "normal" coil if I could find one under 60$(including shipping). By what do you mean "care should be taken"? Do you mean I should isolate the spark gap from the solid state power supply? I don't think I would be able to solder any of these 40+ component circuits but I also can't afford to spend 100$ on a part I can only use for one device.

What I tried to say in one of my above posts is that I got a simple sstc kit for christmas, and spent 5 months just trying to get the secondary wound. I then started working on soldering only to totally screw it up, only to use my spare parts to try again, and screw it up again. And it only had 8 components! I still have the 4" sphere toroid however.

IrC - 30-8-2015 at 21:17

"I don't think I would be able to solder any of these 40+ component circuits"

Hopefully the day will come you realize you only learn by doing when it comes to any mechanical skill which is inherently an art. Study Marx banks maybe that will be simpler if playing with sparks is what you want to do. Or better yet build a Van De Graaff not much soldering involved there (battery holder to switch to motor).

TheAlchemistPirate - 30-8-2015 at 21:36

Don't think that I am not reading about any of these things you guys refer to. I only say "I don't understand this at all" because the way they seem to glance past something like what causes a flyback transformer to work is very frustrating to me. I have soldered simple circuits with 10 components on perfboard before, and am slowly improving. I actually threw together a Van De Graaf generator before. It worked for about five minutes and made half inch sparks until something shifted and it didn't generate any more sparks after that.

If you think I could build one of these circuits I may try, but I will only understand the general function of the circuit instead of why each component is there, which isn't what I wanted.

I will look into the Marx banks though.

IrC - 30-8-2015 at 22:36

"It worked for about five minutes and made half inch sparks until something shifted and it didn't generate any more sparks after that."

Then you went on to something else, soldered your kit, tried to wind a coil, and stopped yet again. Now you are out asking people questions in an effort to move on to something else. This is a problem. One if you fail to conquer it is unlikely you can ever advance. Why did you not experiment to see why the Van De Graaff failed, and get it working again. The effort would help you learn. Why did you not rewind your coil, re-solder the board in the kit with new parts? If you wrecked the foil build it on a perfboard. IIRC you said 8 parts, how hard can this be. Likely not very expensive either. This I think is your real problem, always half trying, giving up and moving on. Fix it. Along the line you will learn perseverance, patience, and likely how it is things work.

This is how I and likely virtually all the members here learned. Yet do so starting only with fairly safe projects. I have seen many who were far too young and lacking in experience with this drive to take on the dangerous projects first. In the interests of full disclosure I must say I also began this way and if I think back a half century many NDE's come to mind. All of which I survived but many only barely. Since the advent of the internet I have read the stories of many who did not. Involve a parent and/or teacher in your projects, this will help you learn while providing a margin of safety. Also if you have trouble winding a secondary you may luck out by talking one of them into doing it for you. This is how a kid got 1st place in a science fair I visited in 1969, he built all but the coil getting his dad to do that after several failures on his own. Something to consider at least.

Edit to consider this: "because the way they seem to glance past something like what causes a flyback transformer to work is very frustrating to me"

You, me, and all who study. In virtually everything. Something you will always deal with. Sometimes the writer assumes others know more than they do so they gloss over important points. Sometimes it is being lazy and more often than one may realize they just do not know themselves. You will always deal with this and getting good at searching, spending time in study, asking for clarification from others (as little as possible don't ask for spoon feeding in lieu of effort on your part), are a few ways to overcome this obstacle.


[Edited on 8-31-2015 by IrC]

aga - 30-8-2015 at 23:38

If you're new to electronics, then a useful way to learn the basic soldering skills is to go and buy some small copper pins/nails.

Knock these into a plank of wood and practice soldering wires to them to connect them up.

No components get broken if you screw it up, and it's easy to see a 'good' soldered joint when you make one.

Once mastered (an hour ?) you can then draw a circuit on the plank of wood, hammer copper nails in at the connection points, then solder your components to the nails.

All very easy and much clearer as you can see the actual circuit.

like this : http://www.stormwise.com/vlf.htm

WGTR - 31-8-2015 at 06:16

Quote: Originally posted by IrC  

Edit to consider this: "because the way they seem to glance past something like what causes a flyback transformer to work is very frustrating to me"

You, me, and all who study. In virtually everything. Something you will always deal with. Sometimes the writer assumes others know more than they do so they gloss over important points. Sometimes it is being lazy and more often than one may realize they just do not know themselves. You will always deal with this and getting good at searching, spending time in study, asking for clarification from others (as little as possible don't ask for spoon feeding in lieu of effort on your part), are a few ways to overcome this obstacle.[Edited on 8-31-2015 by IrC]


Yes. This. I've been immersed in electronics for 25 years. These assumptions about others' knowledge are almost unavoidable.

Part of the self-study process is learning how to piece together information from different sources. It is rare that one source/author presents an entire topic in a sufficient way. You pick up nuggets of useful information from multiple sources, and piece them together, learning how to filter out the chaff. It's not easy, but eventually you learn how to recognize good information quickly, and not waste time on the fluff.

I try to write things that generate leading questions, if one chooses to ask them. I don't really have time to explain an entire topic start to finish, because that means I have less time in the lab. I'm a very slow writer. However, I'm happy to answer questions.

A flyback converter contains a flyback transformer. If you look at a schematic of a points-based automotive ignition system, you are essentially looking at a flyback converter that uses a mechanical switch (points) instead of a MOSFET. You can even replace this switch with a mechanical vibrator, if you can find (or build) one. Personally, I do things the solid-state way.

When you build things for the first time, it is helpful (and normal) to test things in sub assemblies. When I build a coil, I test it with a meter to verify its inductance, that it doesn't have turns shorted, etc. I'll check the MOSFETS to make sure that they're switching, that they have proper gate drive, etc. Then perhaps I'll wire up all the inductors and MOSFETs, and bring up the supply voltage slowly, looking for possible over-voltage problems, or current fault conditions. For a high voltage supply, then I'll characterize its output, see how it handles short-circuit conditions, etc.

Explaining something so that someone else can build it is actually very difficult. The one doing the explaining has to consider a wide variety of skill levels in the audience, every possible thing that can go wrong in assembly. Something that one person already knows how to do, and does without thinking about it, can take an hour or two to explain to someone else. For similar reasons, kits can be good ones or poor ones, depending on how much care the person designing the kit put into it. A poor one may take considerable debugging to get it to work for the first time.

Some people, like Tesla, can visualize entire projects in their heads, do some calculations, and then build something that works the first time. Most of us, however, need to spend time in the lab getting our hands dirty. Don't be afraid to blow up components, if it helps you learn. Small signal transistors are cheap. I usually buy them in 100 piece quantities. Things like resistors and capacitors are easy to reuse, if you stay within their voltage and power limits.



[Edited on 8-31-2015 by WGTR]

aga - 31-8-2015 at 09:04

Quote: Originally posted by WGTR  
Don't be afraid to blow up components, if it helps you learn.

That's a very good point, also the one about buying in quantities of 10 (at least).

Too often a lot of time is wasted doing careful calculations when the detonation of $0.05 component would give you exactly the figure you were looking for.

If explosion is likely to happen, wear your lab goggles, as bits can fly around.

https://www.youtube.com/watch?v=3b7mjukhTyQ

WGTR - 31-8-2015 at 10:19

Might I also suggest, that if you want to learn how transistor circuits work, then 2N3904's (NPN) and 2N3906's (PNP) are good generic choices for low power/voltage experimenting. They are usually cheaper than 2N2222's. A quick check on eBay showed 100 x 2N3904 pieces available from Tennessee for $4, with free shipping, for example.

When I started out I did small assemblies on wood, like aga suggests. If you have heavy solid copper wire, drilling small holes in the wood and hammering pieces of the wire tightly into the board to make mounting posts works as well. I built several radios on stiff cardboard. The connections were made by wrapping copper wire around the leads and then soldering.

aga - 31-8-2015 at 11:22

I made some radio circuits on flattened out beer cans to get a great earth plane, and also screen the circuit when finished simply by folding the metal into a box shape and tack-soldering the edges together.

The surprise was that the solder flowed into the can material, which i assumed was Steel as it is magnetic.

Whatever it is, it takes solder really well, and it certainly isn't just Tin.

Edit:

Found some of those circuits in the junk box.

Here's some radio stuff on beer-tin ground plane with the donor material in the background :

[Edited on 31-8-2015 by aga]

IMG_1511.JPG - 176kB

WGTR - 31-8-2015 at 12:06

Why am I not surprised that this project of yours involved beer in some way? :cool:

I've cut up metal food cans and found them to be solderable as well. They seemed to work OK for shielding purposes.

TheAlchemistPirate - 31-8-2015 at 12:09

Wow. I have a lot to think about lol. The van de graaf generator is completely disassembled at this point, though the sstc has two nonfunctioning circuit boards and a badly wound secondary. I have more patience now, and should soon be getting my replacement parts for an electronics project book. As I said before I blew up(more-so popped) all my transistors. The reason I'm using 2n2222s is simply because the projects require them.
You're probably right that I need to practice my soldering, I'll try the nail thing later today. Are those radio transmitters or receivers, aga? I was intending to do the project around late fall, because my beekeeping hobby winds down around then. In the meantime I was thinking about buying a simple FM transmitter kit. I was considering trying to amplify the signal with transistors, though that might be slightly illegal :D .

This thread has gotten slightly off topic(not that I mind), maybe I should post later when I get a solid plan for building a coil? Or should we still talk about miscellaneous electrical things that I may need help on? You guys are very intelligent, thanks for helping.

aga - 31-8-2015 at 12:28

They were/are radio modules, such as mixers, oscillators, amplifiers etc. that i was playing around with in differerent configurations that didn't work very well.

You should stick to one project and Finish it (as should i).

The feeling of Success is better than just finding a New project to not finish, and teaches you more too.

A 2N3904 will generally do fine where a 2N2222 was specified.

The main things with bisexual transistor compatability (there are so many of them now) is the collector-emitter voltage Vceo and the Current Gain hfe.

2N2222 Vceo = 40V, hfe = 100~300

2N3905 Vceo = 40V+, hfe = 100~300

www.farnell.com is a great vendor with datasheets on all of their products online, which is where i found those data.

Other transistor characteristics become important depending on the specific application.

TheAlchemistPirate - 31-8-2015 at 12:36

I had ordered the 2n2222s long before this thread even started, :D . These 2N3904s must be very impressive to you guys lol. But you're probably right, I should wait until I finish the electronics book (I have actually had it for 2 years). I also just remembered the negative ion generators I ordered from china a month ago...

aga - 31-8-2015 at 12:39

2N3904 is just the newer and cheaper model of the 2N2222LG2.

It's a bit like Windows 7 and Windows 10 - same shit, just that the newer one is cheaper.

2N3906 is the companion PNP version.

2N3904 is the NPN version, which most people prefer as the schematic looks nicer with NPN transistors.

[Edited on 31-8-2015 by aga]

IrC - 31-8-2015 at 14:34

"the schematic looks nicer with NPN transistors"

That is only when they are dead bugged on beer cans right?

aga - 31-8-2015 at 14:39

Nah.

Just that the emitter arrow goes Down to the 0v rail with an NPN transistor.

Looks nicer.

TheAlchemistPirate - 12-9-2016 at 17:03

I have spent the last year trying to become more competent in electronics, so hopefully I will be able to get something done now :D .

I have a few questions I might be asking later, but this is the one which is bothering me most. To put it simply, I don't know how exactly a tesla coil works. I have read quite a few articles detailing exactly this, but they rarely seem to exactly match the actual construction of the device. Here is my somewhat simplified understanding-



There are several "things" which are now called tesla coils, and some are very different than others. What I am talking about is the closest thing to what Tesla himself created, which he did not call "tesla coils". According to what I could discover on the internet, there were several devices Tesla invented which could be called "tesla coils". They all involve two or more resonant circuits, and fairly high voltages and frequencies.

- One was an inductor connected to a capacitor bank (In series, or in parallel???) so as to form a tuned circuit, with another tuned circuit consisting of a cylindrical (air core) inductor and a roundish metal(usually) "elevated capacitance" at the top. The inductor connected to the capacitor bank surrounded the other secondary inductor, usually near the bottom. The primary circuit was the one which was powered by a high voltage and high frequency AC.



The method of how this AC was produced is a mystery to me, since (in modern designs)(and in SGTCs) the high frequency/high voltage is produced by allowing a rectified HV from a transformer to charge a capacitor bank, which periodically (usually only a few 100 times per second) discharges through the spark gap so as to "ring" the primary coil. But in many cases this same capacitor bank functions as the capacitor in the primary LC circuit! Can this same capacitor bank, which builds up the power to cross the spark gap and energize the primary coil, act as the capacitance in the same circuit? Isn't a tesla coil (at least partially) two circuits tuned to the same frequency, which requires two dedicated capacitances for each one? But how can this be possible if the primary capacitor bank is only connected to the primary inductor a few 100 times per second? These systems usually are supposed resonate at hundreds of kHz aren't they?

I hope I am making sense, I'm trying to know exactly what is happening in these devices. Another issue is I cannot determine whether the primary capacitor bank should be in series with the primary inductor, or in parallel. I have seen both configurations, and while I am new to this realm of electronics, I think there could be a large difference in operation.

What if a pair of resonant circuits could be built so as to resonate at a frequency such as 10 kHz? I know this might be rather difficult, but perhaps the dimensions of the secondary inductor could be radically changed from the normal thousands-of-turns-using-tiny-wire type to something else. Maybe the thickness of the inductor could be increased as well as the form diameter? Anyways, if the coil resonated at 10 kHz, and a rotary spark gap was built which could achieve that switching speed, and the primary capacitance was dedicated solely to LC duty, what changes would be observed? I would like to try.


[Edited on 13-9-2016 by TheAlchemistPirate]

wg48 - 13-9-2016 at 03:23

Ok lots of questions. I will answer some of them. I hope you have already read the many explanation you will find on the internet. I can only answer in general terms as you have not given a particular schematic your having problems with.

For a spark gap impulse type Tesla coil the primary is a tuned circuit consisting of an an inductor L and capacitance C. That C is charged by the power supply until a sparkgap SG breaks down.

While the SG is conducting the L, C and SG form a loop, a circuit with each element in series ie a tuned circuit. Near the end of the ring down the SG stops conducting and the loop is open circuit so no tuned circuit but then its not required. In fact it must open circuit to allow the C to be recharged.

Note 1: The sinusoidal ring down has a lower frequency sinusoidal profile as the energy moves between the primary tuned circuit and the secondary tuned circuit.

Note 2 : Deciding if a circuit is parallel or series must be done with respect to two circuit points so depending on which points you choose the circuit can be parallel or series.

Yes you could build a 10kHz Tesla coil but the L and C of both the primary and secondary would have to be 15 times larger than a typical 150kHz Tesla coil. Yes you could power it from in effect a 15kHz supply. I believe Tesla experimented with alternator powered coils at those relatively low frequencies.
But it would be very difficult to reach the peak powers of typical impulse coils.

The used to be apps like wintesla and probably now modern versions that allow you to design an impulse Tesla coil by inputting various parameters. In particular they calculate the approximate resonant frequency of the secondary. They would give you a good feel for how large a secondary, its number of turns and the size of the top load you will need for 10kHz.

I think you should build a pair of these LOL


90L10K.jpg - 162kB



[Edited on 13-9-2016 by wg48]

TheAlchemistPirate - 13-9-2016 at 17:36

I forgot to mention what I am actually trying to achieve, which may affect your suggestions. I am trying to verify or dismiss some of the effects usually associated with Tesla's various devices, such as transmission through Faraday cages, or wireless power transmission. I know many of these ideas haven't actually been demonstrated (recently, at least) , but I think they are worth experimentation. These kinds of effects (according to Tesla) are disrupted by the breakdown of the surrounding atmosphere, which form the sparks many TC builders desire. My goal is not to create sparks, but to maximize the energy which is stored in the device. Another idea I have is the cooling of the device to super-low temperatures using liquid nitrogen, which might increase the power to a large extent. Tesla reportedly did this too.

This is the main reason I want to stay as close to the original designs as possible.


Back to the actual discussion--
So in a normal SGTC the primary tuned circuit doesn't have to exist constantly for it to work, and the primary capacitor can function as both a part of the switching mechanism and the primary tuned circuit? After the primary tuned circuit "rings down" , and the spark gap is extinguished, does the secondary tuned circuit store the remaining energy? Or is the resonant energy of the secondary tuned circuit dependent on the primary tuned circuit being present? What I'm trying to ask is if the secondary LC can store energy without the existence of the primary LC.

But this might be where my plans differ from many other TC plans. Since most TCs are designed solely for the creation of large sparks, the peak voltage would probably be the main (or sole?) goal of those designs. This might not require the device to have energy in the LCs continuously, since it would be the peak voltage that matters, not the ability to build up energy. Is this what you mean by "impulse coils", because they only operate during each "push" of energy (the spark gap breaking down)?
If this is the goal of most SGTC designs, then it makes more sense to me. Or maybe the above paragraph is pure speculative nonsense :D . Assuming it is true, I want to try to preserve the energy from each successive "push" of energy, which might require a continuously active primary LC? Wouldn't that require the power supply to work at the resonant frequency? Of course, as we stated before, this would require large inductances and capacitances, as well as (in the case of SGTCs) high speed rotary spark gaps. I won't worry about those design challenges right now.

I will keep trying to dig up the original plans and theories, and will probably update this thread soon. Thanks for helping

Sulaiman - 13-9-2016 at 19:19

look at the Wikipedia Tesla Coil page https://en.wikipedia.org/wiki/Tesla_coil
I think that you want a continuous wave Tesla coil.

in a typical 'sparks'n'arcs' TC most of the input power goes to the arcs, then spark gap and I2R losses.
A secondary on its own looses energy mainly to I2R and corona losses, with some near field absorbtion, and negligible electromagnetic (Hertzian in the far field) losses.

If you consider a typical 300 kHz TC, e.m. wavelength = 1000 m
so 'wireless power' within a few hundred meters is near field,
energy transfer is generally by magnetic and/or capacitive coupling, non-Hertzian.
I have not come across proof that N.Tesla ever efficiently coupled his TCs to the ionosphere for his world power distribution scheme.

P.S. for high voltage without sparks you could consider a magnifying transmitter https://en.wikipedia.org/wiki/Magnifying_transmitter
basically a TC primary and an extra coil as a step-up transformer, base-feeding the 'secondary', (with various interactions)

electronic inverter base-drive of just a high 'Q' secondary would be another way to c.w. ehv

or magnetically drive the secondary as in the very popular Continuous Wave Solid State Tesla Coil (google cw-sstc)

There are so many websites with lifetimes of TC information.
[Edited on 14-9-2016 by Sulaiman]

[Edited on 14-9-2016 by Sulaiman]

[Edited on 14-9-2016 by Sulaiman]

wg48 - 14-9-2016 at 14:28

Below is a link to a video of an mechanical analog of the coupled primary and secondary of a classic impulse Tesla coil. Its a demonstration of two coupled pendulums. The initial lift of one of the pendulums corresponds to the charged primary capacitor when the spark gap conducts.

A mechanical analog of a magnifier type Tesla coil would have three coupled pendulums not necessarily with the same type of coupling between them.

A driven CW inductively coupled tesla coil would have one pendulum with the man rhythmically tugging on the spring.

https://www.youtube.com/watch?v=CguKKl9mX2s

Here is nice Java simulation of coupled pendulums you can play with. The graphs are equivalent to the voltages on the primary and secondary.

http://www.walter-fendt.de/ph6en/coupledpendula_en.htm

And even better, a page from Richie Burnett's Tesla Coil site. Its got the pendulum analogy too. An excellent site. Its British of cause LOL
http://www.richieburnett.co.uk/operation.html


[Edited on 14-9-2016 by wg48]

[Edited on 14-9-2016 by wg48]