Halfway Report on Building a Tesla Coil and Driver

Posted on by

I was trying to beef up my Tesla Coil driver and there were a few challenges:



I have built my driver, which I may update depending on my tests, but with what I have I learned a good deal to know where my shortcomings are, which I summarize here.

The coil is basically a transformer nothing too special by today’s standards. The output voltage to input voltage ratio is equal to secondary winding to primary winding number of turns. Some general properties of a Tesla coil are as follows:

  • The primary is made of thick wires wound around 10 times, because much higher current at very high frequencies (1MHz) run through it. Skin effect increases the resistance of wires.
  • The primary diameter is usually bigger than secondary, or grows as it goes up. One of the reasons is that the secondary voltage is higher at higher heights of the tower and more clearance to it is needed to ensure the energy doesn’t discharge to primary halfway, unless the primary has good insulation around it.
  • The secondary is made of much thinner wire as way more windings is needed, usually 1000 turns or more.
  • The winding on secondary only goes one way, and not back and forth. Because as we get close to the end of the winding the voltage significantly rises in terms of 100s of kilo volts. So if the windings go back and forth, low and high voltage sections of secondary will be overlapping, breaking the insulation and arcing between the turns.

As you see for my first secondary winding I used 28AWG wire and above 700 turns. It was wound on a black PVC pipe with about 4.2cm diameter and 27cm long.

The top load toroid size was pretty much randomly picked. I decided to make it like a spring. It ended up being around 5cm inner diameter and 11cm outer diameter (3cm toroid diameter) made out of 12AWG tin-plated copper wire. I placed it on a cross made of the same wire and used a threaded metal rod to hold it with some nuts. As you can see I passed the metal rod through a round piece of plywood and fastened the rod and wood with more nuts. The rod is cut close to the wood and doesn’t go through the coil to the bottom, because it will have a very large voltage that would arc to lower voltage sections of the secondary. The output of the secondary is tied to the rod and toroid.

I was able to wind with the help of my drill. The wire was thick enough not to break under the tension applied, but I couldn’t do the same thing with the 38AWG wire and had to softly wind it with hand, which also takes forever.

But why did I decide to increase the winding? It was because I was not able to get the length of arc I wanted. Remember the following diagram from Slayer Exciter article? With my new beefed up circuit I still have similar to Q1 MOSFET transistor switching the primary, but my driving circuit is different. When my project finishes, I will provide the final circuit I will use to drive the high power version

Slayer Exciter with MOSFET and gate driver

Slayer Exciter with MOSFET and gate driver

See, when we switch on the transistor, the primary is charged with high current and when Q1 turns off, the energy stored in primary causes great voltage pulses on primary that get transferred to secondary with respect to the transformer winding ratio. This is similar to my discussion in Making a Taser article.

Now this great voltage spike on the primary falls over the transistor. So it needs to be able to handle such high spikes. Otherwise it will enter an operating range called avalanche mode, where the transistor will start conducting high currents and at such high voltage, it would mean great power across the transistor which quickly heats it up and will pop like you saw in the video.

Now my transistors were rated at 1200V, which means that if everything went as planned, I could get a maximum of 1200 x 70 (transformer ratio) = 84kV. But you saw what my best arcs looked like at such high voltages. Still impressive, but I wanted more.

To satisfy my greed, I had two ways:

  • I had to raise the MOSFET rated voltage. But after some searching they would end up very expensive each.
  • I had to change my secondary to one with much higher turns so that the secondary to primary ratio would be higher. I’m planning on over 200:1 ratio.

So the second plan was much cheaper to achieve and I’m doing that. We will see how well it goes later. I am also hoping for some additional benefits which are:

  • Increased inductance should lower the coil’s resonance frequency and hopefully a better control of the output.
  • Should be able to use the 170VDC more safely as the higher inductance should mean lower current.
  • And of course triple the output voltage and triple the arc length

Of course there is a third way which is shown in the image below. This was original Tesla’s way. He was actually inducing tens of thousands of volts over his primary which would mean over millions of volts on secondary. The way he did is, he was using a first transformer to raise an AC voltage to many thousands of volts, which he would charge across a capacitor. The capacitor was connected to the primary through a spark gap. When the capacitor voltage would rise high enough, which would happen in every cycle of the AC, the air in the spark gap would break and an arc would short the contacts between the spark gap, throwing all that high voltage across the primary which would oscillate at high frequency until the energy was depleted.

This is one of the ways Tesla did it

This is one of the ways Tesla did it

Now, that capacitor had to be in tune with the coil’s resonance frequency. The sparks would fly out of the toroid on the top.

This method creates great arcing noise as the spark gap breaks with a frequency equal to the AC. and also the arcs are much more lethal, as they pulse out at AC low frequency.

But I like the solid state method better, as you can make it virtually noiseless if the feeding power supply doesn’t have low frequency noise or ripple. And also you can touch the arcs with a piece of metal and they won’t zap you (skin effect)

I will discuss all these and my driver design later when I’m done making it. But for now, stay tuned!

84 thoughts on “Halfway Report on Building a Tesla Coil and Driver”

  2. Sir i it’s not working, I used 550 turns in secondary and 5 turns in primary coil.
    And I am using D718 transistor it good for that.

    Please guide same like your circuit
    I am doing this for exhibition.

  4. It’s in fact very complicated in this full of activity life to
    listen news on Television, so I simply use web for that reason, and obtain the most recent
    news.

  5. Pingback: 会唱歌的闪电——特斯拉线圈 - 完整

  6. Hello Medhi. For the life of me I simply cannot understand why the IGBTs you used would not be able to handle 170VDC, considering they are rated for 1200V. Could you help me resolve this?

  7. When you decide to revisit your tesla coil, you should abandon your ‘turns ratio’ calculation. It does not work for resonant transformers.
    For tesla coils the calculation for maximum secondary voltage is the input primary voltage times the square root of the ratio of primary to secondary capacitance. Or alternately the input primary voltage times the square root of the ratio of secondary to primary inductance.

  8. Regarding trying to boost your tower’s voltage to 70,000V; would you not be able to have cascaded multiple transformers together. Much like designing an amplifier circuit to get a Avo = 125 with 3 small gain stages of Avo = 5 cascaded together, would the same concept not apply to tranformers? My only instinct is telling me there’d be a lot of signal pollution but if its sitting at the same resonant frequency there’d be some destructive interference (assuming there’d be some phase shifting due to the giant inductor we call a transformer)

    If it were a feasible idea you could just sit on a stage_1_primary of 10 windings, stage_2_primary of 100 windings (10x gain) and then stage_2_primary of 5 windings and a stage_2_secondary of 200 windings (or maybe 10, 400) to get 68000V from your 170VDC supply

    Love your videos Mehdi! Keep up the good work

    P.S. I know this is late for the beast 🙁

  9. Pingback: Music, Magic and Mayhem with Tesla Coil | ElectroBoom

  10. Hi
    I made a simple tesla coil using a bc337 and I tryed using it with a 22kohm resistor but it just won’t work I tryed everything even swiching the primary polarity
    Can you help?

  12. I just does not understand the top most round cycle connection ? What is it please explain the round top most cycle of the tesla coil

  13. Hey, so I’m a 17 year old senior in high school. We have a science fair. last year i won first with a arc furnace, this year I plan to leave with a bang. Your Tesla coil caught my eyes and i cant resist doing it. I have taken a normal high school physics class and am taking Physics 1, as well as a circuit design and building class. I have the potential to build it. I just had a few questions. How much of the 38 gauge magnet wire did you use for your secondary coils, like the length you needed? I was also wondering where you picked up your wire because all i can find is 100 ft of 38 gauge? I will probably have one or two questions.

    Thanks Mehdi, you are an inspiration to me.

    • Hi Cameron. For my original tesla coil I used 24 AWG wire. around 700 turns on a 11 inch tall tower of around 1.65″ diameter. You can calculate the required length I’m sure. I haven’t used the 38 AWG wire coil yet. But that one is around 2000 turns. The resonance frequency of second one would be around 1/3 of first one. making this slayer exciter circuit is simple. But there are no big arcs if you are looking for those. You need much more power and circuitry. But you can get the lights glowing and electrons moving inside a incandescent bulb.

      • Thanks Mehdi. The reason i was asking how much was because i calculated the turns with the 38 gauge wire and there was ALOT. I wouldn’t have the funds to make it. But That’s fantastic that you used 24 gauge wire, i want to get mine as close to yours as i can so that way problems may be solved easier. And I’m not looking for big arcs, just musical ones. If there are any tips you have for me with the musical arcs I would love that!

        Thanks Mehedi, best of luck in your future endevours.

        • Yes 24AWg should be good. The musical Tesla is in the next article. The circuit I designed is a bit hard to make and operate, so good luck!

      • Hello I need help I’m making a SSTC and I have used 27 AWG and a primary coil within 5 turns but I need help how to connect the rest of the circuit

  14. So the only way you can increase the arc length right now is by increasing the number of secondary turns? Are there any downsides to doing this?
    Also how does the width of the secondary affect the arc length or other factors?

    Thanks

  15. Hey I have just put my head into electronics and I saw this video and you wanted huge arcs and for that you could use a Microwave Oven Transformer(MOT) or a Neon Sign Transformer(NST). The NST will increase you voltage up to 12Kv and will give you an output of 840Kv. The MOT will only increase your voltage up to only 2Kv and give you an output of 140Kv. All of this is based on your ration 1:70.

  16. Hello
    I dit a cool thing to great a BIG ARK
    3 v and 500 amp
    I put that in my slayer exciter an BOOM
    that ARK WAS so BIG

  19. Hello:

    Im trying to make a VERY HIGH voltage tesla coil using solid state method.
    I do not want very bright arcs, but VERY LONG arcs, using 500W of power.

    I know that i need a lot of turns in the secondary (around 6000-8000)
    I will use 38 AWG wire, but im thinking that the pipe would be too long, and the resonant frecuency too high. Am I right?

    Can I use WIDE pipe (around 50cm diameter) for secondary coil?
    Will the magnetic field of the primary coil dissipate on the air before i reach the secondary due the wide of the TWO coils? Primary and secondary.

    Thanks. Your videos are great.

    • More number of turns will reduce the resonance frequency as the inductance increases. It might get too long though, which is what you want for longer arc. Remember your arc won’t be longer that the distance between your top load and the secondary bottom or your primary winding, whichever is closer. But you don’t want thin and long. I would say roughly the diameter shouldn’t be less than 1/10 of length (just how I like it). Wider coil also causes larger inductance and lower frequency. 50cm sounds huge for diameter, especially since your length will be around 1 meter.
      It is generally better for energy coupling for primary and secondary to be close to each other, but then too close and your coils will arc to each other, unless you have good insulation between them.

      • Ohh, Thank you so much. Understood.

        I think the best would be Cone-type secondary, this way, the primary coil would be wide with few turns, and close to secondary.

  20. Dorod Mahdi jan
    Man komak niaz daram lotfan
    Shoma chegat Volt va ampere be solid state driver tesla coil midi? Man 16 volt midam va 1,5 ampere vali transistore man misoze. Lotfan komak kon

    • The voltage is good but it doesn’t need that much current. if it draws that much, then probably something is wrong. Have you tried to flip primary connections? Does the lower power driver work?

  22. Hello,
    I am a TA from somerset communtiy college and we have built a slayer exciter
    using TIP31C transitors and have issues with them heating up and not lasting. We want to try it using your circuit using the mosfet transitor. What size resistors do you recommend and diode? Thanks so much for any information. Our kids love your videos. Keep up the good work!

    • Hi there, the same size resistor and diode as the simple one would work. But try to see what is wrong with your simple circuit, because you may have the same problem with the new one too. Flip the primary winding connection as a start.

      • i have also made it with TIP31c but i have had no problems yet maybe there is something wrong with your circuit or just like he said try to flip the primary but we could also just change the polarity of the coil. Try it and see if it works !!

  23. Hi mate, i’m a tesla coil lover like you and i think i can contribute for your beautyful coil.
    First thing, the size limit for the lenght of the secondary coil is four times the diameter of the tube, for an example, if you have a 4inch tube your coil lenght can be 16inch maximum lenght. This helps the energy of the primary be more absorved by de sec.
    Second thing, Your mosfet usually can take the voltage you put on that before. The problem are the reverse voltage peaks we got on this very fast switching. But you can put a 2kv 1nF ceramic capacitor in paralel with the mosfet, connected to the drain and source, and put a reverse fast diode to conduct de reverse peaks out the mosfet.
    And Third thing, try to make your primary with a copper wire. Its better for de coil because conducts the energy more efficiently
    I hope i help you with this comment, Beauty coil, good luck

  25. Hi,
    What is input of primary coil ? In other words if I apply 80V DC on primary is it still work?
    Thanks

    • The primary coil doesn’t have any problem, it is the transistor that may not take that much voltage and power.

  26. سلام اگه میشه اموزش ساخت یک تسلا کویل رو به فارسی برام بفرست
    ممنونم

  27. Hey Mehdi,

    I’m pretty new to making electric circuits and the like and I’d like to ask some questions.

    First off, for your MOSFET gate driver version, where are the grounds connected for the diode, secondary and transistor? Also, is the battery negative connected to the primary or the resistor?

    Sorry for asking dumb questions but I can’t seem to find answers anywhere else.

    Thanks!

    • No questions are dumb questions. Let me read them…

      Man! They are some dumb questions!! Just kidding! They are good questions.

      The diode and MOSFET and the gate driver all have the same ground, which is your supply ground or your battery negative. In this circuit though, the secondary doesn’t have ground connection and basically connects to the input of the gate driver.

      The input of ICs generally have diode protection on them to clamp the input voltage to the power rails to avoid excess voltage and breaking the device. Diode D2 does that, But I wish I had another diode added with anode to the BATT line and cathode to the IC input to support the IC more against high input voltages. yet, the circuit works as is.

      • Hey Mehdi,

        Thanks for the info but i have 1 last dumb question. Would the battery be connected to both the gate driver and MOSFET, or would you use 2 separate batteries?

        Thanks again,

        • Here, yes, that’s what I did. The voltage would be limited to below 20V. But you could supply the coil separately with a different higher voltage for more output, but that may hurt your MOSFET if the voltage is too high.

  30. Dear mehds, im an avid fan i hope u could send me a picture of your sxhematic… i mean an actual photo im a noob on electronics though i made a lot of creations already if possible i need pics cuz schematics are horoglyphics toe as of now hehe thanks love your vodeos

  31. Finally! I always wondered why in tesla coil people don’t overlap turns like in a transformer to increase the number of turns! Love your videos!

  33. I made a coil motivated by your videos, but using 30 AWG magnet wire. Can I still hope to achieve such awesome arcs you show in the later video with this gauge?

    Here’s my setup with the simple slayer/exciter circuit (no MOSFET driver) driven with an input of 12V DC.

    http://i.imgur.com/SFQdco9.jpg

  34. Could you give us the schematic of your tesla coil , using the full bridge rectifier , thanks

    • I ended up not using the full bridge rectifier for this one. That high of a voltage breaks my transistors

  35. Iwont to make small transformer give me ahighe. Voltage using amegnatic power can you help me. And pay the way I wish ican help you your ideas

  36. Dear sir!
    I have made the circuit . i used 2n2222a transistor. i have made the coil with 1000 turns and secondary coil with 8 turns . but its not working. 🙁 when i increase the vcc then a large current is drawn into transistor . what should i do now. please guide me. its my semester project and until now i can’t done this 🙁

    • Did you try flipping your primary connection, that usually does the trick. Do the simple Slayer circuit and make sure your connections are correct.

  37. In the slayer exciter, the circuit automatically keep its working frequency same as the second winding’s resonant frequency.And you built a PWM driver with comparators. I’m thinking that you can make them into one circuit. You can pick up the second winding frequency and stick to it, then try to figure out a delay when the MOSFETs are on. Frequency might be changing, but it won’t change far away from the original frequency. You can check some induction cookers’ circuit, they use the same idea aiming to reduce switching loss. (More, I suggest that you can use IGBT s instead of MOSFETs, they are more difficult to break down and work better than MOSFETs at high voltage. You can also try to make primary winding work in resonence, that’ll increase the power a lot. Maybe you have already discovered that when you use square waves or triangle waves to modulate the circuit, sparks grow much longer, I don’t know how this work, but it really is a skill for bigger sparks 🙂
    PS: I’m just a senior school student andI don’t know much about elec. Also I feel sorry for my poor English 🙂

  38. I would love to build this. Can i use the coil drivers found on ebay and build the coil myself. Love your vids and sense of humor

  39. Hello friend could you pls email me some pictures of your tesla driver ive been trying to build again the first succesful one i built was from a bug zapper. And i had no success after that after downloading lots of schematics. Can u send me yours with pics pls…. Thanks more power heres my email dexdirecto@gmail.com

    • Making a tesla coil is not too hard, but driving it is tricky. How are you driving yours?

  40. Mehdi, I am insanely curious to see the videos you’re posting to facebook. Sadly (for me who wants to see the videos) I deleted my facebook account because I think facebook is awful on the whole privacy thing. Anyway, for those few of us who shun that form of social media, will there be any other way to see the rest of what you’re doing?

    • 🙂 just make a throw away facebook account with no information of yourself on it!

      • Well, my facebook account has been zapped. Congrats on finishing your secondary winding, looks awesome!

  41. Hello, I am an amateur Tesla coil builder (as you can see in the link above). I was always told Tesla coils, being air core transformers, do not perform on the basis to turn ratios, but on resonance frequencies. Is this only true for spark gap devices, such as mine, and not for solid state TCs?

    regards,

    Paul

    • I am even more amateur than you in tesla coils, but I’m pretty sure number of windings play a big part. Of course resonance frequency is a huge factor in maximizing the output power. The Tesla coil is like a high quality factor LC filter, which means it is a pretty sharp filter. Now there if an output filter, with secondary and output stray capacitor, and in case of spark driven coils, another one with primary inductance and capacitance. Unless you tune the primary LC filter to match the same frequency as secondary, your frequency will be greatly filtered. But in any case, the output/input ratio remains the same as number of windings. If you probe the primary while you are tuning the frequency, you will see that when you are approaching the resonance frequency, your primary voltage also rises.
      With solid state drivers, you don’t need to have an LC circuit on primary. It is your driver’s job to have correct switching frequency, to switch the primary on/off.

  42. Hey you can find high voltage NPN transistors in CRT TV’s. I have got 3 of those and they are around 1500V ( it’s next to the flyback transformer in the big circuit)
    You can also buy a step-up transformer to step up the mains to 240V so you can get 311VDC with your FULLLLL BRIDGEEE RECTIFIERRR! lol
    I might build a spark gap tesla coil in the future. It won’t be big as I have very little resources. I have lots of wire I got from a CRT monitor.

    Good luck on your SSTC!

    • NPN transistors are slow switchers compared to FETs, that’s why I can’t use them. Otherwise they are cool!

      • You can try IGBTs. They usually work under frequencys of more than 100 kHz. If you make your primary winding renosant, they may work under the freq of 1MHz or more.

  43. Really looking forward to updates on this. Been wanting to build a SSTC for a long time and will definitely follow this.

Comments are closed.