Kaizer SSTC II


I built this is a modified version of mt first SSTC, the Kaizer SSTC I. It uses the same secondary, topload and driver board. New things is a full bridge of IRFP460 MOSFETs, audio modulation, shielded drivers and a new casing.


WARNING!: Working with electricity is dangerous, all information found on my site is for educational purpose and I accept no responsibility for others actions using the information found on this site.

Read this document about safety! http://www.pupman.com/safety.htm


In the quest for longer sparks I decided to use a full bridge to take advantage of the full voltage on the bridge.

Firstly the MOSFETs will be mounted on top of a heat sink so its easy to change them by only removing the secondary platform and solder them off.

Secondly the drivers will be shielded in order to avoid the EM field generated by the Tesla coil itself to inject noise into the drivers.


Bridge4x IRFP460 MOSFETs in a full bridge configuration.
Bridge supply0 – 260 VAC from a variac, 8 A rectifier bridge and 1500 uF smoothing capacitor.0 – 365 VDC on the bridge.
Primary coil115 mm diameter, 1.78 mm diameter isolated copper wire, 8 windings.
Secondary coil110 mm diameter, 275 mm long, 1000 windings, 0.25 mm enamelled copper wire.
Resonant frequencySelf tuning at around 250 kHz.
Topload100 mm small diameter, 240 mm large diameter, toroid.
Input powerContinues Wave mode: 2000 – 4000 Watt at 200 VAC input voltage.
Interrupted mode: 100 – 2000 Watt at 260 VAC input voltage.
Audio modulated mode: 300 – 400 Watt at 150 VAC input voltage.
Spark lengthup to 475  mm long sparks.

 Full Bridge SSTC Schematic

The UCC3732X are MOSFET driver ICs, one non-inverted output and the other inverted, in order to get a push-pull drive of the gate drive transformer. A gate driver IC can deliver the high peak currents needed to drive MOSFETs efficiently.

The 74HC14 is a inverting hex schmitt trigger, it is used to get a proper solid 0-5V square wave signal from signals that are not perfectly square, the antenna feedback can vary a lot in waveform and amplitude, the 74HC14 converts this to a clean drive signal for the MOSFET drivers.

I tied all unused input pins of a 74HC14 to ground, floating inputs and a noisy environment is a recipe for trouble. The noise can couple between the gates internally and make the whole IC not work properly.

The music modulator works by amplifying the audio signal in the LM741 and at the BC547 transistors. The 555 timer ensures that the signal length of the generated square wave is much shorter than the audio signal, in order to not have too long on-time and thus damage the MOSFETs / IGBTs from over-current.

Full Bridge SSTC Construction

15th March 2009

I took apart a 19″ LCD monitor and a 24″ CRT monitor, from these respective computer parts I salvaged a good piece of acrylic from the LCD monitor and a fairly sized heat sink from the CRT. I cut the acrylic in half for a 2 level platform and the heat sink was cut in 4. It is necessary to isolate the MOSFETs from each other as their housing is also a conductor.

19th March 2009

I installed both driver electronics and audio modulator under a metal casing from the CRT monitor. I wanted to shield it from the heavy EM field surrounding the Tesla coil. This is to avoid problems with the driver being interrupted by its own EM field.

I made the bridge out of four IRFP460 MOSFETs, four MUR1560 diodes, four 5R resistors. The power supply is a 8 A rectifier bridge with a BHC 1500 uF/450 V smoothing capacitor, a 27K 7W bleeder resistor is added in the final build.

I later removed the audio input jack, due to it making a short through its metal housing to the ground rail, I had overlooked that the audio in negative was not common with the ground rail, but there is a capacitor in between.

The secondary is held in place by a crate for ventilation on houses, its an easy and quick way of taking the coil apart for transport or storage, and it holds the secondary firm and tight.

A acrylic tube is added to support the antenna, in this way it is possible to adjust the coupling of the antenna to the secondary simply by pulling the wire.

The new shielding of the audio in signal is made from a piece of shielding from a industrial cable pulled over it and grounded.

The secondary with terminations. 110 mm diameter, 275 mm long, 1000 windings, 0.25 mm enamelled copper wire.

The complete coil looks, except maybe the electrical tape used to hold the topload together.

 Full bridge SSTC Sparks

Interrupted mode

At 250 VAC input voltage, 350 VDC on the bridge, it was possible to reach 475 mm long sparks. This was in interrupted mode, to a grounded object.

More pictures of sparks in interrupted mode, it is running at about 4 – 5 BPS.

3rd May 2009

Continues Wave mode

At 200 VAC input voltage, 280 VDC on the bridge and a power consumption around 10 A, peaking at 20 A, the coil was drawing somewhere in between 2000 to 4000 Watt. Surprisingly resultung in a very hot, thick white arcs punishing the dead iPod shuffle which remarkably left the player relatively unharmed considered what had just taken place.

These flame like sparks are 250 mm in length.

18th August 2009

I constructed a new topload from two cheap aluminium frying pans from Ikea. Firstly I removed the handles and the screw from handles was grinded away, in order to have a smooth surface. Secondly I fixed the two parts together with a long screw through both of them.

6th September 2009

During a run of CW at full input voltage, the full bridge blew apart completely, with a loud bang.

I suspect it was something with the new topload, it was smoother, but also a bit smaller. The resonant frequency did maybe go up too much. The higher losses caused the MOSFETs to exceed their maximum die temperature.

Tesla Coil Audio Modulation

I use a audio modulator made by the user Reaching (Martin Ebbefeld) from 4hv.org.

For sound input I use a cheap children’s keyboard from a toy store. It is far from perfect for the job, especially because its waveform is highly distorted. It is not clean tones, but seems to involve a lot of modulation inside it to simulate different instruments. But it is cheap and expendable.

Watch the film and look at the schematics for more about the audio modulation.


Upgrading the SSTC I with a full bridge was a absolute must. It is small changes compared to the better performance and the driver have no problems at all driving four MOSFETs instead of just two.

Getting sparks at 475 mm length in interrupted mode and white power arcs at 250 mm length is truly satisfying for this little coil, the secondary winding itself is only 275 mm in height in comparison.

Enjoy the demonstration.

 Full Bridge SSTC Demonstration

Demonstration of different modes.

Interrupted mode with the new topload.

New topload, running in CW mode.

New topload, running in interrupted mode and closeup of sparks.

488 thoughts on “Kaizer SSTC II”

  1. We would like to order SOLID STATE TESLA COIL for about 500 watt

    please let us know if it possible and how much it will be


    Maxim Pryymuk
    Skype: exclusiv-store

  2. Why do you need two IRFP460’s? The second one really doesn’t do any thing! Would it not be closer to the 555 fly-back driver?

  3. I know how a full bridge works and I knew it takes four; I was talking about a half bridge, like the kaizer SSTC I. but useing a single Mosfet, I tryed it and it works just as good as two, ( but the Mosfet runs a little warmer).

    It also works with a 555 timer circut, just not as easy to tune.

    so I guess it really is like a fly back driver after all……… just self tuneing and w/ interupter.

    induction heaters are just like tesla coils, Cool!

  4. You are absolutely right, power electronics are to be found in almost any household item today, its just a question of the power levels 🙂

    Electronics design really only comes down to one thing: Accepted component heating vs. production costs.

    If we want better control, we have to pay more and the complexity of our designs raise too.

  5. I am almost sure that Reaching from CTC-labs.de used it for his first DRSSTCs, but if you want a pure MIDI interface I would find something else 🙂

  6. Hello, I’m Japanese high school students.
    I want to make this sstc.
    But I have some uncertain things.
    So please answer two questions.
    1.Does it work if IRFP460 is replaced with IRFP260?
    2.How many Voltage is Rated Voltage Range of the C8 and C9?
    I’m sorry my English is unskilled.

  7. Hey Imai

    1. IRFP260 will work at lower bus voltage, it is only rated for 200V where the IRFP460 is rated 500V, what is important is to keep a certain headroom, I would advise on atleast 30% headroom, so I would not apply more than 140VDC to a halfbridge of IRFP260.

    2. The capacitors I used for C8/C9 are 400VAC MKP X2 capacitors, you can use any DC blocking low ESR capacitor as long as it has a higher voltage rating than your busvoltage.

    I hope this help 🙂

  8. Thank you for answering.

    After all, I decided it to use of IRFP460.
    Because it is more certain than IRFP260.

    And I bought 400VAC capacitors.

    Your advice helped me very much.
    Thanks a lot!

  9. Dear mads,

    The capacitor in series with the + audio input, is it’s polarity shown correctly? negative side to the audio signal input (+) and positive side of the capactor to the 2K resistor?

  10. Hi, Is there a way to run this SSTC from 250VAC in modulated mode?

    If so, what modifications should I make?

  11. Also I want to ask: If I use 12V (60A) instead of 350V for the primary, how many turns in the primary coil do I have to make?

    I’ll use different mosfets of course…

  12. Hi Jogosa

    To run this coil at higher voltages you should decrease the primary to secondary coil coupling coefficient and of course insure that all components can withstand the higher voltage.

    The thing is that with this topology you would need a very low primary impedance in order to get a current high enough with only 12 Volt through the primary, so a loose guess would be something half a turn…. If you want a low voltage Tesla coil I suggest you take a look at Skori’s Tesla coils: http://skory.gylcomp.hu/tesla-ml/index.htm

    Kind regards

  13. Thank you Mads. Already have both PCBs made, wondering whats the pulse width range on the timer…
    By the way do you know an easy way how to make the topload?

  14. Will it be OK if I use 110mm diameter PVC pipe for the secondary and 160mm for the primary coil?

  15. The coupling coefficient is a number representing the proximity of two coils, the lower the number the further the coils are from each other, at a coupling of 1 the two coils would occupy the same physical space, which is of course impossible in practise.

    Read my article in order to see a easy to make topload.

    The coupling for a SSTC will be too low with 30 mm between the coils, again look at my article.

  16. Hi again, Mads

    I have already made and tested the driver circuit, and everything is working except that the transistor Q6 goes hot after a few seconds. Is that because it dissipates too much power when its switched on? Maybe I should use higher value for R14 ?

    I tested it with 1kHz 900mV sine input the transistor gives me a square wave on the collector…

  17. Hello,

    I will be using the same components that you have listed but in North America we only have 120 VAC. What can I do to get the same HV output as your SSTC? ie increase the number of windings or increase the base frequency of the 555 timer?

  18. Hello James

    Increasing the number of windings on the secondary will only have a slight effect compared to higher bus voltage.

    Increasing the bang per second rate of the interruptor will for a SSTC only result in thicker sparks, not much longer.

    You should look up how to make a voltage doubler and add that to your circuit.

    Kind regards

  19. Hi Mads
    What amplitude of audio input woltage should I give to get any effect for audio modulation? I have maked this, but after giving it about 0.5V audio signal(max) from mp3 player there are no signal from it. I have builded simmilar audio modulation for SSTC on TL494, but with same effect.
    Sorry for my poor english.

  20. Hey Radek

    I used a toy keyboard with the volume at max, you might need a amplifier inbetween to get good results for a unamplified signal.

    Kind regards

  21. Hi good work very nice coil, My question is what is the diference between SSTC and DRSSTC? because I see to bridge LC system in the primary coil similar to used in DRSSTC? and when you used interrupter is very similar the sparks form, I will like make DRSSTC but I dont have IGBT do you think is possible used bridge of IRFP460 with low power? thanks.

  22. In your audiomodulator you have a .1 uF cap between the audio signal ground and the schematic ground. I understand why its there but isnt it better to have a common ground?

  23. Hey Den

    You could risc to ground the circuit back through the audio input with the capacitor, it will also take care of DC components in the signal.

    Kind regards

  24. Hello Nikita

    The layout is made for veroboard, not a etched PCB. I suggest that you draw a PCB layout yourself, it is a small circuit and it should not take you long to do so.

    Kind regards

  25. Hello.
    I wan’t to build the SSTC following your guide, but something is just not clea.
    I see a little red wire inside a tiny tube near to the coil.
    What’s that for, and where does it come?
    Thanks in advance.

  26. Nice to meet you I am not Japanese so English well. I saw this site want to make the SSTC for the first time. But I get work but I’ve set up tentatively. The size of the pulse transformer are using here or how much will I question name? Do you want to operate 100V input power is tentatively but after?

  27. Thanks for the answer.
    Just one more thing, what heatsinks did you use for mosfets? Did you buy them on internet? If yes, could you link?

  28. Hi Unleashed, those four heat sinks are all made from a long piece of heat sink from a audio amplifier. Many of the materials I use are salvaged from old electronics. A great way to get electronics for free.

  29. 私は日本人です。この三端子レギュレーターは%Vに下げるものでいでしょうか

  30. Tell me the value of a resistor R1 and R4 and R5.Please tell me the value of the voltage conversion of 3-Terminal regulator.

  31. Hello tぬい

    All the values are written in the schematics. R1 is 4R7 Ohm, R4 is a 10K potentiometer and R5 is a 2M Ohm potentiometer. The reason for this large value it to be able to get down to 1 spark per second.

    Kind regards

  32. Hello Mads.
    As you probably know, i’m building this sstcc, and it’s almost done but unfortunately i couldn’t find a 11cm tube, so i had to take a 10cm one instead. Is that ok as well?
    Same for the alluminium tube on the topload, i’ve used a 80mm diameter, made up as a toroid, of 25cm of diameter.
    Will this do?
    Then i wanted to know, since i’ve made the logic part and i’m waiting for the mosfets that will take a while to come, is there any way of testing it?
    The last question is about the supply.
    Since i don’t have a variac enough big for that, will i get to run only in interrupted mode?
    What happens if i run it in continuous wave mode at 220Vac?

    Thanks in advance,
    Best regards

  33. Hey Unleashed.

    Your 100 mm secondary tube will be fine.
    The smaller toroid will also work.
    But remember that both these factors will result in your secondary circuit having a higher resonant frequency, unless you add more turns to the secondary coil. Its properly not much to care about, but do check your numbers.

    You can test your driver circuit by having all but your power MOSFETs in place, especially the GDT, without it you can risc to set your driver ICs on fire.
    Use a frequency generator or a simple 555 circuit to simulate the feedback frequency and you can check out the GDT switching with a oscilloscope.

    Running the coil directly from 220VAC without ramping the voltage up slowly can either result in some nice big sparks from your switch it on, or a boom, puuff and smoke. You get no small weird switching noises when going full voltage and will likely only see catastrophic failure if something is not made correctly.

    Running CW at 220VAC will only pull as much power through your coil as it possibly can, be aware! This can easily be up to 4kW!

    Kind regards

  34. Hey x-ray

    If you look carefully at the pictures, you can see that this is laid out on veroboard, so there is no PCB files to give out.

    Kind regards

  35. Hello Mads.

    Thanks for the answer.
    Since i didn’t have a oscilloscope, i tried my driver circuit with a simple LED+resistor.

    To trigger the antenna, i wound it to the 220cable.
    The 1N60 diodes made a weird noise like “bzzz” i guess because the 50hz 220 is hearable frequency.
    When i unwound from the cable anyway, the led turned off and the diodes didn’t make noise anymore, so i think the driver is alright.

    Other problems i was having are about the grounding.
    I would be grateful if you check this topic on 4hv.
    And please answer to my last question.

    Best regards.

  36. Hello again Mads.
    Today i had the first test.
    I obtained just some corona on the break out point.
    Of course i did a soft start with the variac.
    Then it started to make a weird noise… Some second after 2 mosfets literally blew up.
    I’m gonna change it now.
    Have any idea of what could have been^?

  37. What is all the capacitors? What is the variety of capacitors that I also use the ceramic capacitorsfilm?

  38. C1, C7 and C11 are electrolytic capacitors
    The rest are MKP or MKT capacitors, you can also use ceramics, but it will be hard to find the values used in a package that is still small and cheap.

  39. Hi,
    ich habe soweit alles fertig aufgebaut, alles funktioniert auch und bei 30V bekomme ich schon 2cm Entladungen.
    Das einzige Problem ist jedoch dass ein Strom von ca. 60% des Maximalstroms fließt, selbst wenn die Primärspule nicht angeschlossen ist. Das spricht doch dafür, dass die MOSFETS falsch angesteuert werden.
    Was kann ich dagegen tun?
    Vielen Dank

  40. Hey Stephan

    You will have to use a oscilloscope to check your driver signals and go through your power section to be sure there are no faults.

    Kind regards

  41. Hi, im just wondering about the diodes conmected to the entanna, is the part number only 1n60? And would a 1n4148 diode work??

  42. Hey Chris

    There are multiply manufacturers of 1N60 diodes, but the 1N60 part should be a part of their part number.

    You can also use 1N4148 diodes for the feedback clamping.

    Kind regards

  43. Thanks wards i appreceat the tip. I have another question, how many volts are your decoupling caps for your GDT circuit?

  44. Tesla coil works great now but my only problem is as soon as i put a toriod on it i cant get any sparks out of my tesla coil even if i put a break out point , all i can hear is a buzz noize. Would you know what im doing wrong??

  45. Well i bring it up to 70v ac and i finaly got a break out point but as soon as i go higher my mosfets blow right away and im not sure why? And when i bring it to 70 v ac i get almost a foot of arc.

  46. Hello,

    I want to build a SSTC for a student project.

    What is the main reason for the capacitors C8 & C9 in series to the primary coil?

    Can this capacity be left out?

    Kind Regards

  47. Hey Manu

    C8 and C9 are DC blocking capacitors so the AC waveform can work on the primary coil without any offset. If you want further explanation look up some information on this subject.

    Kind regards

  48. Hey Mad,

    We have successfully build up the Kaizer SSTC II.

    We have only tested it with lower Input voltage (with an external power supply) yet, but there are 10cm lightnings visible. 🙂

    We have only one question.

    If we earthed the buttom of the secondary coil (grounded over the external power supply), the antenna circuit delivers no a signal anymore, so the tesla coil doesn’t work.

    Without grounding it works well.

    What could that be?

    Thanks and Kind Regards

  49. Hello.
    The question on the audio modulation. R14 goes hot after a few seconds. Tell me, what is it? Q5 and Q6 workers, the scheme was built without errors.

  50. Hi Mads, I was wondering about the height of the secondary; according to the specifications it says that its height is 275 mm long, is it the secondary coil’s height or the PVC pipe’s height? Thanks! ^_^

  51. Thanks again, Mads 😀

    All I need was the secondary. The driver, audio modulator, and the bridge is working perfectly. I’ve even made a pcb so that I could preserve this design of sstc of your’s.

    Kind regards,
    your friend Charles ;D

  52. Like many other people out there, I have been looking at the different kinds of coils out there, and I really wanted to make a coil that makes a somewhat decent spark length and to be able to play music/midi. Which one would you suggest that is somewhat easy to build, and yet is able to play midi and have a decent spark length? I am still fairly new to all this, and I want to learn all I can about these coils!

  53. My answer will be that you should look at a audio modulated SSTC, like the one written about on this page. It is a fairly simple design that can be pushed without suddenly exploding. A DRSSTC modulated with a midi interrupter is a advanced project and it is also a lot more expensive.

  54. You can read it from the schematic. The driver section, all resistors are just normal 1/4W resistors, most capacitors are MKP type, rest is ceramic. Just be sure to get components that is within the voltage ratings. C8/9 in the bridge section is just normal 275VAC X2 MKP capacitors found in most power supplies today.

  55. Ok thank you. If I happen to have anymore questions, is there an email or forum I can go to and ask away? or is here fine?

  56. This will probably be my last question here before I start posting on 4hv.org — For the audio on this design, I know that you have to convert the signal to light (fiber optic) and then reconvert it on the other end for it to work so you don’t have an electrical connection between the two devices. Do you either have or know where I can find a simple design to make a converter? (would 4hv have a design?)

  57. You do not need it, but you will of course experience noise issues without a proper shielding / grounding of a wire connection.
    I am not familiar with transmitting a analogue signal over fiber, but research that if you want to use a fiber connection.
    Be sure to read the rules on the 4hv forum, asking directly for designs or schematics is not allowed.

  58. Thank you for your support Mads! I do have one last question for sure — I have been looking around at different schematics, and I am probably going to make Steve Ward’s since it was already set up for 120v. My dad confirmed that this is possible, but I just wanted to make sure with you. Is it possible to take the audio portion from your circuit and merge it with steve’s?

  59. Hey Keegan

    This is already based on Steve Wards mini sstc, so you should have no problem, as you only want to do what I have done here 🙂

    Kind regards

  60. Hello Mads, ‘m still trying to make this thing work, all i get is mosfets blow up, i’ve bought an oscilloscope, but i get no output from the GDT, also, i get a weird noise from the diodes (1N60) and from the the GDT ( like, bzzzzz).
    I would also apreciate if you could tell me what’s the GDT core you used.


  61. Hey Vladislav

    The GDT is a 15:15 turn, which you can see in the schematic. If you have the wrong phasing you will switch on the MOSFETs wrong and in the worst case this will make a short circuit through them instead of leading the energy through the primary coil. Read about full bridge switching if you want to know more.

    Kind regards

  62. Hi there Mads!

    Really nice work!
    Been trying to figure out the numbers with this, single/half and full bridge.
    How do i calculate the peak current and voltage over the mosfets? di/dt?
    How big capacitator do i need (on rectifier) to be safe? usually oversized i guess but how much?
    The real problem for me is not the “drive” part (gdts etc) this is working really nice and this is my strength, one the other hand my knowledge about power electronics is somewhat poor. Most people just slam this together but i feel there is no point to do so when i do not fully understand what happens in the bridge when it switches this kind of load, ofcourse there is a sinusodal ring down but at what values? is there a fancy name or equation for this? i have a a few st-micro mosfets well suited for this but i want to do the math first.

    The classic nst/mot ones is just not that fun anymore… 😀


  63. Hey Jont3

    When you want to get this in depth with the theory behind switch mode, you properly need to get some literature, I have myself enjoyed the book “Power Electronics Handbook” by Raschid.

    But as you say yourself, we are working way outside the safe area of operation for most of our components and thereby we mostly keep to best practises and experiences as most books does not treat the subject “How to overdrive components at 500% for 10 minutes” 🙂

    Kind regards

  64. Hi. I want to know how i can mod this to eat 6 kw to 10 kw peak using igbts.
    IXGX120N60B These. At 220 volts. I will be using a full wave bridge rect which is capable of 1000 volts at 50 amps. BY329X diodes. Can those diodes be used in the whole circuit I mean substitute every diode for this nice little diode . I can use a 12 volt 3 amp little adapter and throw away the regulator part of the circuit and connect it directly? and for these extremely high powers. Would i have to do a full bridge of igbts or can i just leave them as a half bridge? I will be running under continious mode. The sec will be a 6 inch or more wide with 26 awg wire will be 30 cm tall. Or will i be better with 30 awg and a 30 cm long secondary same width? I hope for some 17 inch ++ flame arcs min.
    And by the way. Can you tell me the best source for gigantic heat sinks. Were to find or were to buy? And i hope you reply.
    Whatever the email is for while i am posting. I entered random stuff

  65. Hey BOoom

    Remember that IGBTs have a constant on-state voltage drop in form of the Vce saturation voltage, this will give you a higher dissipation and this will for sure be the limiting factor in your design, running these devices in continues wave for flame like arcs will properly make them explode within 10 seconds.

    You can use the BY329X diode where power diodes are used, signal diodes should not be replaced by this!

    12V 3A adapter should be good enough for the driver. Add some more capacitance, some 10000uF to be able to deliver some peak current if you fear the driver could pull the adapter down at heavy loads.

    I would advice you to use a full bridge to spread the load out on 4 switches instead of only 2.

    Make a secondary using 26AWG, you need some cross section to deal with the heavy current you are going to push at 10kW. Be sure to get resonant frequency down around maximum 100 – 150 kHz, your conduction losses simply gets too high above that.

    Old amplifiers, frequency inverters, motor drives, industrial equipment. Search scrap yards or surplus stores for cheap heat sinks.

    The ferrite core from a flyback can be used, but depending on type they are only good to around 50 kHz or 100 kHz.

    Kind regards

  66. Would mosfets go until 10kw.
    Would it work to use 22,000 uf instead of 10,000

    I heared about a cw tesla coil with igbt bricks. How so?

    If cores of flybacks wont work. Would the core of cfls or radios or power supplies work?
    Thanks for replying 🙂

  67. Hey Mads, just a quick question. Im building a drsstc with two cm300dy 12nf bricks but i just bought some 26 awg wire for a 10″ tube thats 48″ inches long would my wire work? Or should i bite the bullet and get bigger?

  68. Hey BOoom

    If you find the right device, be it MOSFET or IGBT, that seems to be able to run within your specifications, they should both be able to be used.

    Adding more capacitance means nothing more than you have more storage for energy, when it is just for smoothing a supply voltage. How good is your basic electronics knowledge? a 10kW SSTC could seem like a steep project.

    Do you mean QCW Tesla coils with bricks? They run long on times, but heavy water cooling is required, IGBT bricks running in pure CW, I doubt anyone made one that could run for long, but ask the ones who made them, not me 😉

    Be sure to get some high permability ferrite cores, around or over 5000 Al, you will not find such in cfls, radioes or power supplies.

    Kind regards

  69. Hey Chris

    I think you will be fine with 26AWG, you should watch the secondary temperature though, but honestly I think it is okay, thinking about that I used 30AWG for my 2,5kW DRSSTC 1.

    Kind regards

  70. I just don’t have a 10,000 uf cap. I have a 22,000 uf. And i just like to make sure. I also have a 52,000 uf cap.
    My electronics knowledge is a.. Little advanced beginner. I built the zvs driver with out problems. And i understand most of the sstc circuit. By the way. What is permability If 10kw is that steep. Then what about 6 kw. Would those mosfets go with me till 6 kw. And what would i want to change?

  71. Thanks Mads 🙂 sorry to bug you but i Appreciate your help. I was just wondering, im thinking of getting two 5600 uf at 350v for filter caps, or should i use different caps over all? ones in like a series and Parallel circuit for voltage doubling? I wanna get high output with my two cm 300’s

  72. Thanks! Just wondering could you send me a schematic of your full bridge circuit or link me to a web site please.

  73. BOoom: Read this http://thedatastream.4hv.org/gdt_index.htm and adding more capacitance is not a bad thing, use the 22000uF.

    Chris: If you look at my DRSSTC3 you can see I use 4x 6600uF 350V for having 6600uF at 700V, so if you have 1200V switches, I would make sure my smoothing capacitors could handle a higher voltage than just 1 phased rectified.

    Kind regards

  74. Oh okay, so the only thing id be changing on the diagram for cm300 12nf igbt is the zeners being 20 volts for the gate, or do i use 33v?What value should the resistor be on the gate? Sorry i just wanna make sure i get this all right, i have built many other tesla coils but this is the biggest one and first time i use the big bricks.

  75. Hey Chris

    Depends on the voltage you use to drive your gates, if you use 24V it makes no sense to use 20V zeners, gates are usually rated 30V, so 33V zeners is usually used there.

    Gate resistors is something you experiment with until you find the right value from looking at the waveforms, you do not want the signal to ring up at turn on and neither have too round corners. For a first, try something between 5 to 10 Ohm.

    Kind regards

  76. After taking a really close look at the schematic for getting parts i realised something. The control circuit consists of 3 parts The upper most part
    The part on the right and the part on the left The part on the right is for CW and for the left is the interrupter then why is the cw wave part so much harder than the cw one? And i do not want my coil to modulate audio if it will cause me to need all the components on the right. I want a coil capable of cw only .. Also i cannot find the ucc37322 and the ucc37321 at my electronics store. I can only find 555 timers :/
    and are the 74hc14 the signal diodes? those big arrows? All i could find on the web site is a
    7414 “Hex Schmitt-trigger Inverters” [74HC14] Is this the one?
    What about the lm741. I bet it is a part of the audio modulator which i will be leaving out. But still i cannot find it
    And the 1n60. I cannot find that. Any alternatives?
    Is it better to use a 12 v 3 amp little adapter.? or a 12 v 3 amp smps? the adapter is like 3 times cheaper. but if the smps equals better performance.
    So simply i want to simplify the circuit that much to only run cw and be hell good at it. And also substitute a few components. Thanks

  77. Hey BOoom, order your parts grom Digikey, they have everything you need and they offer great customer service 🙂

  78. Hey BOoom

    For only running CW, you only need the upper part of the schematic. Lower left is the interrupter and lower right is the audio modulator.

    This coil is based on Steve Wards minisstc, take a look here for the original simple verison: http://stevehv.4hv.org/SSTC5.htm

    The UCC372X is a normal and inverting MOSFET driver IC, you can also use other MOSFET driver ICs, just be sure to get some of atleast similar rating.

    The 74HC14 is the correct IC you found, it is just the individual gates that are drawn for simplicity in the schematic.

    The LM741 is only used for the audio modulation.

    Find a small signal diode with similar ratings of the 1N60 to replace it. You should check out buying such parts from Hong Kong ebay sellers, it is often cheaper than finding a alternative in your own country.

    It makes no difference to the driver if you use a iron core transformer supply or smps, I would buy what is cheapest if they have the same ratings, if I had to buy one.

    Kind regards

  79. No credit card :P. Thanks anyway chris.
    Mads those are the inverting mosfet thing ..
    i found
    Would these work?
    Also what wattage would the resistors have to be? ( power mosfet section)would the usual 1/4 watt one be okay? How much ohms too? Cause a main problem with my current project ( igbt zvs driver) is getting high wattage resistors. Tho once i finish it i will start the tesla coil project. Tho if i even find the sstc parts more available than the igbt zvs. I might even postpone the zvs.

    I just finished the new schematic for me

    Is it right? The kaizer power electronics part got erased by mistake while using the rubber on the photo. If that annoys you i can remove the pic. But is the schematic right? God it is so easy. XD And the antenna. I can make a copper antenna. A aluminum zinc coated antenna and a iron antenna( that in a radio) Which would be the best? I guess the copper. And is a topload really THAT necessary

  80. I read some stuff about how bad pvc pipes are at serving in the cw sstc duty. And that page recommended acrylic tubing. Now there is a chance i get get a 10 inch wide ( The air inside the tube) and i decided 14 inch length Tho original plans were 12. But i added 2 more inches for.. Because the same website recommended leaving a inch before and after winding just for it being easier. But i will wind on as much of those 2 inches as i could :D. I dont know if making something to wind on is worth it. i mean it is only my first coil. There is even i shop that might wind it for me. And also the same website recommended putting something to close the hole at the top and the end of the pipe. Like a piece of plastic. I think it would be nice cause i would attach a screw and easily put- remove toploads 🙂 if i really need one tho.
    And lets say i put all the cicuit in a aluminum box. Is it fine if i connect the part of the secondary that is to be grounded, on the aluminum box. And ground the box along with the lead of the secondary. And at the schematics the 2 12 volts positive polarity inputs near the ucc. Booth positive ? I don’t know if its my eye or the picture quality I see booth positive. but just like to make sure. So nothing goes boom. 😛
    And if i made it into a pcb. ( much cleaner ) Should all the traces be thick? Are there parts of the circuit were high currents are flowing are parts were we do not have high currents. And how wide should i make the traces be on the pcb?
    And for the filter capacitors. Should the capacitors be rated for the voltage Before or after the filtering. i will be filtering 220 v ac. Sometimes goes to 210. And when that is filtered will go to some 300 something volts. my caps are 250. If those could be used i can only use 6 of them. Plus another weird looking 450 uf cap that does tolerates up to 15 kv or something. Tho it is rated for 450 volts. It easily eats the output that a flyback with a cfl gives. And gives loud discharges. And anyway. If i should use 300 volt ++ caps for filtering 220 volts. I will have nothing to do but series 2 for half capacity double voltage. And then use 10 of those pairs to reach 1000 uf. And those caps aren’t cheap. Talking about capacitors.
    What is the voltage rating of the caps your using in that circuit. And should any caps be electrolytic apart from filter caps. And in the power electronics part. Those C8 and C9 Can be replaced with 0.50 uf caps ? Thier rating is about 2500 volts. It is simply a large cap bank. And since those 2 in parallel then we have a total of 1 uf at 2500 volts of large ceramic! capacitors. Those would work? There are 18 of them in series-parallel in that bank. Tho i can add some more caps i have laying around. to make the capacity 1.36 not 1 uf. They just will be ceramic Again i just like to make sure so when i first power it up there are no booms. And on the control circuit side. How much volts are those little capacitors. If you read my 2 large posts. Then thank you ALOT and if you didn’t Still thank you alot for helping me :).

  81. hey Mads… one question, can I increase the output voltage of the secondary? I want to have longer sparks, can you help me with this? Thanks ^^

  82. Hey Mads, yesturday i just finished winding my coil and now im trying to figure out what size of toroid to build, i was thinking of using 4inch aluminum flex pipe like your drsstc 1, but idk what to use in the middle? Also because i have a low budget , i found 4 caps rated at 6800 uf 250v which give 500v at 6800 uf. Is this going to work?

  83. Hey BOoom

    You can use the IR2112, but you will have to use some logic gates to split your signal from the antenna into a inverted and noninverted signal to input to the IR2112.

    You can use normal 1/4 Watt resistors for the gate resistors.

    The schematic is what I described, but prepare so that you can connect the enable pins to either positive or negative, I do not recall which makes then be enable all the time, I just remember that I left them floating for CW. It might not work stable for you.

    I just used some standard copper wire for antenna. Stay with materials that are good conductors.

    A topload is necessary to lower the resonant frequency of the secondary circuit, to provide a capacitor to the secondary circuit that can supply peak currents. It will stress your secondary coil less with a topload, it will provide a protecting field that prevents breakout from the top windings of your secondary coil etc.

    Acrylic will withstand higher temperatures than PVC.

    I wound my first 3-4 secondary coils on a brown handle by hand.

    Connect the ground from secondary to your ground with a thick proper cable, then ground the rest of your stuff to a common point that then all leads to the ground, do not let secondary current flow through your driver box, it is just a bad practise.

    The inputs to the UCCs are both positive, you can see that the GND pins go to ground.

    If you rectify 230VAC you get rougly 320VDC, so your capacitors should be rated for more than that, 350VDC will do fine, but get 400/450 if you want long run times often. Do a calculation on the size of the filtering capacitance you need, you know the input power you desire and the voltage, so you can google how to calculate it.

    Everything is expensive if you buy it from new, salvage stuff from old electronics to keep your costs down.

    C8 and C9 are just regular MKP X2 capacitors found in my power supplies, rated for 275VAC, so some 500VDC rating thereabout. Capacitance is not so important, can be between 0,33 to 1 uF with too much trouble, they only act at DC blocking capacitors.

    All the decoupling capacitors are properly 63VDc rated on the driver side, but just look at the voltage you use in the circuit and then use a capacitor with a higher rating. You have a lot of basic and fundamental things to learn about electronics, please be careful!

    Kind regards

  84. Hey George

    If you think you have optimized your Tesla coil in all the ways possible, by building it or properly sized materials and components, then there is only the input voltage to the bridge that will give longer sparks. A larger topload for a higher stored energy could also be a possibility.

    What is your current spark length? Input voltage? Dimensions of secondary coil and topload?

    Kind regards

  85. Hey Chris, if you are asking about DRSSTC, could you please do it in the DRSSTC article?

    You can use wood like I did for the inner circle and use elastic robe through the tube to fix it towards the circle.

    The capacitors will do fine, good voltage rating and capacitance is enough.

    Kind regards

  86. On the control electronics side. I will use 50 volt ceramics. Should work great 🙂
    Here are the details of my hopeful build

    6-8-10 inch wide. 12 inch long former. If i find acrylic i will use it. But if not pvc pipes. To be precise UPVC 😛 Using 0.4 wire.

    4 IRFP460 with large pc processor heat sinks. With thermal compound to stick it. and 2 fans for each heat sink.

    2000 uf of 500 volts ( HOPEFULLY :D!)

    50 amp 1000 volt bridge rectifier

    Would the 1n60 be replaced by 1N4148 with no problems?

    “”You can use the IR2112, but you will have to use some logic gates to split your signal from the antenna into a inverted and noninverted signal to input to the IR2112.””
    Please explain more? I can’t understand that with my small electronics knowledge 🙁 Sorry :/
    What are Logic gates? Logic inverter? logic gate mosfets?

    And at the control electronics. A 12 volt 3 amp little adapter with 30,000 uf but no regulators. 🙂
    I heard that a smps would work alot better for a tesla coil. But a adapter is 3 times cheaper than a smps and with a smps i have less money to spend on making the actual coil 😛

    GDT using ferrite from flybacks. (What can i do?? :P)

    And for a pcb? Where would high currents flow in the circuit? And What would the minimum dimensions i would want that piece of copper to be? I would suspect no more than 3 amps would flow near the control electronics. So atleast 1 mm thick traces? And at the power electronics part? Some 5 mm thick traces?

    Everything seems fine?

  87. Hey BOoom

    You should be able to use the 1N4148 instead of 1N60.

    As you can see if you look up the UCC37222/1 they are inverting and noninverting output, so the input signal to both is the same signal. The input to the IR2112 need one signal for Hi and one signal for Low, so you need to make a inverting signal yourself for the input. F.ex. with a 7414 inverting schmitt trigger IC.

    Whoever told you that SMPS is better than a iron core transformer for a Tesla coil has clearly no idea what he is talking about. A Tesla coil present some heavy RF noise that without problems can knock out sensitive control circuits as the ones regulating a SMPS. A iron core transformer, with a filter circuit, on the other hand will hardly notice being near. 30000 Uf filtering on a already filtered power supply is overkill, 1000 uF would be enough, but more is not bad 🙂

    Try to use the flyback ferrite, measure waveforms with a oscilloscope to see if the core saturates or it delivers a good sharp edged signal.

    I usually make my traces as large as possible, might aswell take advantage of the copper when its there! Signal paths can be small and it will never draw 3A continuesly.

    Your list of components etc looks fine, now go for it! Get it built, experiment, learn, fail, start over, fail, blow some MOSFETs, make some sparks, enjoy it 🙂

    Kind regards

  88. Will do. After a day or two i will order my components
    And i do not know if the adapter uses a small iron core transformer or a little inverter circuit with a tiny transformer. And happy to know my filtering is fine 🙂 Anyway. I decided on my adapter. 🙂 Good news is that i found pcb ports that can withstand 35 amps and 60 amps peak. and they look cool. Better than the 10 amp exploding thing 😀

    And i guess until i can get the ucc or something like it i will be having trouble. But hopefully a friend of mine will be going to another electronics shop today. And if they have the ucc or a substitute it would be GREAAAT ! And eh. Mosfets are expensive. But what can i do XD? Thanks for helping me with all my probably stupid questions. You will be the first one to know of my tesla coil when it is hopefully completed. and i will give you credit 😀

    Would a double layer 20×30 cm pcb do good ? or should i use single layer Booth are the same price
    And 500 ML of etching acid. And the laser printer way 😀 using a iron

  89. Hey BOoom

    Sounds good 🙂

    I only use single sided, it can be a bit hard to align the two opposite surfaces in parallel. But do what you find best, when you use double sided board, the one side should be kept for negative/ground only.

    Kind regards

  90. Booom(using another pc)

    Okay. Cool 🙂 i will use a single sided board 🙂 And it is a very simple board 🙂 Best thing is i\d use port connectors 😀 no more problem of burning mosfet by the heat of the soldering iron. I actually destroyed a good 70$ by that 😛
    Or with replacing mosfets thanks for the help with all my stupid questions
    I hope to reply in a few days telling you my results

  91. The resistor at the power electronics section? R1 , R2 , R6 , R7
    Those are 100 kohm right? And that booom that wrote above is me

  92. Hey mads. Can the GDT be wound on a stick? i read that it must have no air gaps. a stick does not have any? it is a ferrite stick

  93. I did not find the ucc. So can you explain how to use the IR2112 ? A schematic would be preferred Sorry for the 3 comments. And sorry for bugging you 🙁

  94. Meh. 4th post. Sorry :/ there is no editing. I saw some GDT winding techniques And as i understand i get a ferrite core wind . 15 windings 5 times. And connect each of them to a bridge and one to the control electronics. And if i get it right? it basically throws out all what the bridges give to the control circuit or the opposite while the GDT provides isolation? And basically preventing the control circuits from blowing up? And i seen numerous ways. Some trifilar windings with all the 3 windings in parallel And some people insulate the whole thing and cover it with foil? And also which AWG of wire should i use?

  95. Hey BOoom

    A ferrite rod got a huge air gap, you can look at it as a ferrite toroid that has been cut open and folded out into a straight line.

    There are other alternatives to the UCCs that are much more like them, f.ex. MAX4420/4429. If you want to use the IR2112, split your input signal to the HI/LO inputs on the IR2112 with a 7414 hex schmitt trigger inverter. Simply by letting your normal signal from the antenna circuit go to the HI input, but the signal for the LO input must first go through a inverting gate to get the same signal in inverse. I will help you, but not do all your work by supplying complete schematics to your own specifications.

    You can use normal CAT5 network cable to wind GDTs with, wind 15 turns of it on a core, then pair up all the white+a colour wires to two wires by paralleling them, this will be your primary to connect to the driver circuit. Each of the full coloured wires will be twisted together to be the lead for one gate on a MOSFET, this way you easily get wires for 4 gates. You could look at some of the pictures from my DRSSTC 1, 2 and 3 to see the GDTs I made.

    Kind regards

  96. I have EVERYTHING but the uccs or whatever in the place of them. the pipe. And the base. But i discovered that nasty little adapter that i had hope into him. Was a little inverter. So i looked at what i have. And there is was. a 1200 atx power supply :). And i found a 22,000 uf cap laying around in a box. So now 50,000 uf of filtering. Plus the atx needs much lower capacitance than the 50 hz. That means almost no ripple. And the atx has it’s metal shielding so no problems with a tesla coil and i can use the 3 v rail to run leds 🙂 and i made a decision to put the 10 fans that draw about 1.75 amps ( metered) in parallel with the control electronics part. I hope everything is fine. That is just letting you know my progress. 🙂

  97. Hey Mads!
    Very nice and tidy work! Great, I love it!
    I have a question regarding the gate-drive IC’s. Don’t they get hot in your SSTC II set-up? Do you use any heat-sink or a cooler for them?

  98. Hey Alex

    Thanks for the kind words 🙂
    They world run warm at CW duty, but never so hot could burn your fingers, I actually think I might have added some RAM heat sinks at some point.

    Kind regards

  99. Thanks Mads!
    100kHz is not a problem for my UCC’s but for some reason they get hot if I run them at 250kHz (H-bridge with 100V on it). If I feed no current to the H-bridge, UCC’s stay cold at any frequency. Wonder what that means.
    I have a couple of construction questions. I see you don’t use any pipe for your primary coil, it seems you wind it directly on the secondary with some sort of insulator in between. Could you please explain how you did it?
    Another question I have is: how do you connect the secondary coil to the toroid? I am afraid of arcing to the primary/Hbridge/drive circuit if I put a metallic screw at the bottom of a toroid. It will create such a nice breakout point directed to the bottom where all the drive electronics are situated. Or do I fear something I should not?


  100. Hey Alex

    It could be that the UCCs can not supply enough drive current fast enough at 250 kHz and thus spend most of its time in the linear region for its output transistors, this will give excessive losses.

    I used like 6-8 layers of thick plastic sheet, found in LCD monitors, but you can use anything you could find from plastic ring binders etc.

    At the top I make it like this, except now I have the wire on the outside! https://kaizerpowerelectronics.dk/wp-content/gallery/2008_11_06_-_kaizer_vttc_i/1220130959_1403_FT51774_billede_001.jpg

    At the bottom I use a nylon screw and a ring shoe for the wire, this way there is no metal inside the secondary and the ring shoe only got round surfaces. The bottom of the secondary will not arc if grounded correctly 🙂

    Kind regards

  101. And those plastic sheets you put right onto the secondary coil with no gap in between? How do you affix them to the secondary coil?


  102. Hey Alex

    There is no gap and they are only fixed by a tight winding and secured with electrical tape, there is really nothing complex about it, just some plastic sheets wrapped around the secondary 🙂

    Kind regards

  103. I see. I am just concerned about flashovers from my primary to the secondary coil. I mean I could wind the primary directly on the secondary but people don’t do that exactly for flashover reasons, right?
    On a somewhat related note, have you adjusted the number of primary turns for optimal performance? How do you know that 8 windings is optimal?
    Thanks again in advance and sorry for so many questions.


  104. Hey Alex

    You are absolutely right about the dangers of flashover, I actually did have flash over as the layers did not extend outside of the area the primary covered.

    The reason we almost have the primary directly winded on the secondary is that we need the very high coupling in order to transfer as much energy as possible.

    The fewer primary turns you have, the higher primary current will flow and should hopefully also result in bigger sparks. It is all a experiment where you look at spark output vs. input power. At some point your MOSFETs will blow up 😉

    Kind regards

  105. Thanks for your advice, Mads! I have built my SSTC, successfully ran it for some time but was never able to run it at full power. Here is what happens.
    So I have a clone of your circuit, except for the fact that I have 4 pairs of zener diodes for each output of the GDT. Also, my GDT has 12 turns instead of 15. So when I go up to around 100VAC in the CW mode one leg of the H-bridge blows up due to cross conduction. Also, most of the zener diodes die at this point. (By the way, I use 10 turns of 12AWG wire for the primary) This whole picture looks like the notorious Miller effect to me. Have you ever had similar problems and do you have any ideas on how to resolve this issue? Have you blown any MOSFETs up to now in your set-up?


  106. Hey there Mads 🙂

    One question, can I increase the coil winding as well as the PVC’s length?
    I’m really doubting that I should never do this because it might affect the drivers and the power supply of the sstc :S

  107. Hey Xenon

    If you increase the number of secondary coil windings, this will lower the resonant frequency of the secondary circuit, as it is a SSTC, the primary circuit is not resonant, so you can lower the secodary resonant frequency without changing other parameters of the setup.

    A lower resonant frequency is almost always better as switching losses are reduced. I would say go ahead and do not worry about your driver and power supply in this matter.

    Kind regards

  108. Hello ILoveStorms

    You can use both the MOSFET and both type of ferrite cores that you linked to. Take a look under my DRSSTC 1 where I calculate the gate driver transformer specifications.

    Kind regards

  109. Hi ILoveStorms

    The diode you linked to is classified as a fast diode, the MUR1560 is a ultra-fast diode, if you look at the datasheets there is difference of 200ns vs. 60ns. A fast diode is needed to catch the reverse energy from the switching to not damage the MOSFET. The internal body diode of the MOSFET is often very slow and thus we add a external.

    If I were you, I would find a ultra-fast diode to use.

    Kind regards

  110. Hello mads.
    Can you please explain how many amps are passing through the diodes? And can you point me to a diode that will function in a 15-8 KW bridge?

  111. Hey Kaizer, I’m a rookie in this kind of things and this is my first time I build something like this, can you give me a pice of advice.

  112. HI. I have a question about your design. How the oscillation are triggered ? I saw that Steve Ward used an extra 555 to initialize the oscillations but this design seems working without it. I don’t understand the feedback system at all. By the way, GG!

  113. Hello Ogata

    The oscillation can trigger because of leakage throughout the circuit, so just the energy transferred through the GDT is enough to have a small current ring through the secondary and the antenna can pick up a signal and it starts up.

    The antenna feedback is just a piece of wire close to the coil, coupled through a capacitor, clamped with diodes to get a useful signal level, put through two inverting hex-Schmitt triggers to get a nice square wave signal.

    Kind regards

  114. Hello. Sorry for my bad english, because im from Slovakia. Im build some vacuum and spark gap tesla coils and now I want build tesla coil with semiconductors. Is this kind of SSTC good for first project ? I never work with MOSfets and IGBTs. I have many experiences, but only in vaacum technology. I tryed build mini SSTC with one MOSfet and it isnt work. But this is me. I never know build mini tesla coils 😀 only medium or big coils. Thanks for early answer. Pavol Baran.

  115. Thanks for your answer. I´d like to ask if I can use the coil I used in my last project. The coils is 18,5cm long and 7cm wide. Resonant frequency with topload is about 230kHz. And topload has capacity of 10pF. Thanks for help.

  116. Hi Mads, finally i’ve found time to re attempt building this coil from scratch.
    Once it was done i imediately went straight for testing, but everything is acting really weird.
    The only output i get is a small hot arc to grounded wire, but no discharge in the air.
    The power input gets really high even at low voltages, so i can’t go any further since it’d burn my variac (12Amps).
    The weird thing is that in all this, the GDT is making a weird buzzing noise, and if i ground the antenna the coil actually works better(i’ve experimented a lot with it, out of desperation).
    What are the steps you’d go through to troubleshoot it?

    Thanks a lot.

  117. Hey Unleashed

    Check out your wiring for the transistors and GDT, a high rising current at low voltages could indicate that your switches turn on at the same time and thus short circuits the DC supply, hence the rising current.

    You can also try to reverse the polarity of the primary coil, just switch the two leads around.

    Power electronics do buzz and make a lot of noise, you learn this by time but do not worry, I am also worried when firing up my large CM600 bridge, that really buzzes! It could also have something to do with first mentioned cross conduction problem.

    Grounding the antenna gives it a direct path to the grounded secondary coil, so you are basically giving it secondary base feedback instead of antenna feedback. Try to move the antenna closer to the coil.

    Hope some it can be a help in trouble shooting.

    Kind regards

  118. Hey mads.

    Thanks a lot for the answer.
    First of all, about the wiring of the bridge i think it was right, since a badly wired GDT would lead it not to work at all right?
    Instead, i got a little output as i said before, but only to grounded wire.

    I’ve also tried reversing the primary coil, but it doesn’t seem to make any difference.
    Also, i’ve tried moving the antenna but it doesn’t make any difference as well.

    I have no idea what to try anymore, the coil looks like it’s totally out of tune, or maybe the mosfets are spending too much time in the linear region due to slow switching?

    How can i verify that, considering i don’t have an oscilloscope?

    Thanks a lot.

  119. Hi Mads, I have a question, The capacitance next to de audio input, Why is the negative terminal after the positive audio input?? We have to chance the directon of this polarity??

  120. Hey Unleashed

    When it is not working, double check, double check, take clear pictures of the setup and show together with schematic, then we can start trouble shooting. There is no such thing as a SSTC being out of tune. The primary circuit dumps energy into the secondary and uses a feedback signal from the secondary to switch the primary circuit, so it is self-tuning.

    I honestly think you have a problem with your switches cross conducting, be it driver, GDT or bridge layout that is the problem…

    Kind regards

  121. Hey mads, i’ve just noticed that i have used 470ohm damping resistors instead of 4.7ohm…
    Yeah, it deserves a huge facepalm indeed…
    I’ll replace the mosfets and test again, and let you know 🙂

    Thanks for all the help.

  122. Hi Mads.

    I’m sorry to bother with those continuous questions, but i just can’t get it to work :/

    Let’s go step by step.

    I’ve noticed that the resistors where wrong value and replaced them.
    Today, after replaced all the mosfets that burned yesterday due to that mistake, i went to testing, but nothing changed…
    Only thing that changed actually is that i get much higher current drain and no output at all.
    It’s like the variac’s ends are shorted out, but i’ve checked several times the bridge connections and everything’s right, the GDT phasing as well.
    Even weirder is that even with the coil primary disconnected it still acts like it’s shorted out, as well as if the driver board is turned off.

    I seriously don’t know what to think anymore, it’s getting pretty frustrating that i can’t get to work a simple project like that.
    Thanks in advance.


  123. Hey Unleashed

    That was a very important detail to know that you had blown MOSFETs, you now have to check every single component that you have between GDT and primary circuit. Measure every single resistor, diode, zener, capacitors and MOSFETs.

    Do not get frustrated, we have all been there, I spent days looking for faults in the wrong places, after a bridge blow up it took me quite a while to find out it was just a gate resistor that had failed open circuit that was giving me problems.

    MOSFETs will most of the time fail to closed circuit, also even without it exploding into pieces, be sure to check all the switches, especially when you describe the high current conditions like that.

    Kind regards

  124. Hey Mads.

    Thanks a lot for the help, you were right… My bridge was totally shorted out, and 3 MOSFETs out of 4 were totally shorted between gate and source.
    How can that be?
    When did they fail so bad?
    (they’re not the ones with the wrong resistors)


  125. Hey Mads.

    Sorry for double post.

    Today, after i replaced all the burned mosfets (again) and checked things 4 times, something finally worked.

    All i got though was a small corona at the breakout point and some 4-5cm arcs to a grounded wire, at around 45V, i can’t go over that because there’s a huge current drain from the variac…
    What could cause that?


  126. Hi Mads.

    It’s me again, i managed to borrow an oscilloscope from a friend, it’s not the best around but i think it’ll do.
    I went ahead and made a couple of tests… which were… a total disaster.

    that’s all i got that looked like a gate drive waveform..


    Unfortunately the wave gets totally unrecognizable when i start raising the bridge voltage.

    Got any ideas?

  127. Hi another time Mads, I’m asking about if a simliar Tesla coil as your Tesla coil, Can be a problem if we connect the gorund of the Tesla Coil with the ground of the house, or we have to make our own grown?

  128. Hey Jesus

    I have used the house ground before without problems, but there is no way to be sure if it is a good idea to do so, unless you know how the grounding in your house is made.

    Also be aware that high voltage in the return path from the secondary to ground can result in arc over to the phases in a normal plug in the wall.

    You can also make a artificial ground plane from a metal sheet under the coil, or simply just use lots of aluminium paper/wrap used in the kitchen.

    Kind regards

  129. Hi Mads.
    Sorry to bother again.
    Thanks for the answer, according to this site the gate drive is insufficient, how can i fix that?
    Also, shouldn’t the gate drive wave be at around 10-12volts? Because the ones you see in the pictures are with a base of 1-2V, so the wave is actually at a really low voltage.

    I’ve also noticed that when the oscilloscope is connected the coil actually works better, how can that be?


  130. Hey Unleashed

    I think that you should disconnect your bridge from the GDT, now connect the oscilloscope to the GDT outputs and see if they are correct.

    The probes usually add some capacitance to the circuit and can thereby interfere some, it is also possible that you are grounding your circuit through the oscilloscope which can damage your oscilloscope if you are not careful.

    Kind regards

  131. Hi Mads sorry for ask stupid questions, but I have checked the electronic circuit two times and I don’t know why the out put voltage between the two pins numer 7 of UCC it’s cero. I have tried with the tester, with AD and DC, and in both cases the same, but when I test the seven pin with the five pin of one of the UCC, there are 5.8 Volts. And my question is, Why can this happen and how can I solve the problem?

  132. I asked you the last question because the H bridge don’t run because the Transistors aren’t activated because the voltage out of the GDT is cero cause’ the gate of the transistor don’t activate.

  133. Hey Jesus

    Are you using a normal multimeter to measure these voltages with? You need a oscilloscope in order to be able to measure these voltages at the high frequencies a Tesla coil operates at.

    Do you have a feedback signal from the antenna?

    What is the DC voltage between pin 5 and 8 on both UCCs?

    On the schematic, only lines crossing with a dot on it should be connected, can you confirm that is how you followed the schematic?

    Kind regards

  134. Hi Mads,

    I’m using a normal tester because I don’t have an osciloscope and that’s why I ask you because when I did the first try, the SSTC didn’t work, and I checked out step by step. I saw with the tester that in H bridge didn’t pass any current, and I supose that because the transistors aren’t activated cause’ the GDT don’t supply enought current to the gate of the transistors, just because the electronic circuit works well except this part, the GDT.

    The DC voltage between pin 8 and 5 is 12.07 V in both UCCs, and that’s why I don’t understand why the SSTC don’t work, as I don’t have an osciloscope I just made exactly 15 turns, and I checked out the electronic circuit and the H bridge, I have to add in that when I connected the electronic circuit and the 220 AC, the switch power control don’t “turn off”, this can be because there isn’t any short circuit, so Can I supose that there is any part wrong excepting the GDT?

    What Can I do?

    Sorry for my bad english.

    Kind regards

  135. Hey Jesus

    You need to buy a cheap old oscilloscope or borrow one to properly fault find on these circuits. A tester is no good for other than DC or 50 Hz AC.

    Your best bet to fault find it to have the complete Tesla coil built and see if it works with proper antenna feedback from a signal from the secondary coil.

    Kind regards

  136. Hi Mads,

    Thanks for all, finally I descovered that in my electronic circuit there was one wrong thing. I had put the same UCCs in the circuit and in your schematic there is one UCC finished in 1 and the other in 2.

    Really sorry for have been bodering you these days.

    Kind regards


  137. Hey Jesus

    I am glad to hear you found a mistake, maybe we can see some progress with the right ICs in 🙂

    The difference on the two MOSFET driver ICs is that one is inverted and the other non-inverted, so together they work as a push-pull pair driving the GDT.

    With the same type of IC on both ends you were putting either plus or minus potential into both leads of the GDT and thus nothing happens.

    Kind regards

  138. Hi Mads! I was wondering if I could directly use the mains (110 V) as my power supply. Thanks in advance 🙂

  139. Hey Shoz

    If you look at the schematic with the bridge and mains supply in it, you can use 110VAC on the input where 0 – 260 VAC is marked.

    We need a DC voltage to feed the bridge with, so you have the choices of using half wave rectified, half wave rectified with smoothing capacitor, full wave rectified and as in the schematic, full wave rectified with smoothing capacitor.

    Kind regards

  140. Hi Mads,

    It’s me another time, Jesus. After being testing and thinking that all problems were solved, I did the finall test with the H bridge and the secondary coil. First I connected all the circuit to 12V and all was fine. Then I disconnected the 12 V and I tested all the circuit with 220 V, the result was the same, the circuit seem to work properly, but when I tested the circuit with the 12V and 220 V the house lights blinked for a few seconds and then all the circuit seem to be another time without any problem excepting that form the Tesla Coil I didn’t get any spark. After all I disconnected the electronic board from the GDT to check it out if there was any chip broken and in fact there were both drivers (UCC’s) broken and here is my question. Why did the two UCC’s break?

    Another question I have, Can be posible or I have to have caution of the harmonic injections to the electrical network, I ask you this cause’ I have connected the 12 V supply with the 220 V supply in the same “extension cord”. Do you connected the 12 V supply in a different place that the 220 V supply. I have to be awared of the harmonic injections? Why did the two UCC’s break?

    Kind regards


  141. Hey Jesus

    If you house lights blink, that means you put on a load heavy enough to pull down the voltage, which means you created a short circuit or that filter capacitors after the full bridge rectifier was filling up from empty. A empty capacitors looks like a short circuit and have a large inrush current.

    Do not connect mains power to the bridge without the driver section is running, you risk short circuiting the bridge as you do not know which state they are in, on, off or somewhere in between.

    The UCC driver ICs will likely fail due to parasitic oscillations if there is no load connected on the output.
    If the supply voltage is too high they will also burn, measure your voltages at the 5VDC and 12VDC rail, when your 220VAC is also connected, maybe you connected something wrong.
    If you drive a very heavy load at continues wave, no interrupting, they might also be damaged from heating as they are pushed over their limits. Did you notice if they died cold or warm?

    I have never had problems with using the same extension cord for both control and power supply. The noise from the sparks and magnetic field is the biggest problem with noise injection to the control electronics.

    Kind regards

  142. Hi.
    I’ve been working on the present schematics for quite a long time. Yesterday it got to work in the fixed frequency mode.

    I have routed a small board for audio modulation but as I can observe the 100 Ohm resistor after second BC547 transistor is getting soon extremly hot after the power is connected.

    I have found one mistake already, I got 10k resistor in place where 1k should be, but it made no difference.

    So I assume my board is not working even for a short periods of time.

    I hope for your help.
    Best regards.

  143. I’m sorry for being that lame 🙂
    I thought I got to the point of no hope on myself. Now the original schematic by Martin Ebbefeld is found and I’ve been looking throught the comment section so some ideas showed up.

  144. Hi Mads,

    Terrific site!!

    I would like to ask about the audio circuit in your schematic.
    I have built basically the same coil as yours (following Steve Wards SSTC5), but only a half-bridge, and the damn thing is absolutely fantastic right off the hop. The first time I fired it up, i was able to get at least 10″ sparks. And I have never built an sstc before!!
    What a great little schematic. I would thank him personally for sure, but unfortunately, he has no option of doing so on his site.
    Anyway, now I would like to add your little audio modulated circuit for some more fun. At the top of this page you state that you overlooked the negative side of the audio in jack, and that you have corrected it. But I can’t find where you show this correction. If I follow the schematic at the top, will I not be making the same mistake? How does your correction change the schematic? Could you please provide some clarity?
    Does the switch that you use to change modes have three positions? ie; 555-off-Audio, so that when off, the coil is in CW mode. The way it is drawn on the plan looks like only a two way to me. But I think it would make sense to be three-way??? Correct?
    If this is a stupid question and the answer is supposed to be obvious, or worse, you have already edited it and it works the way you have shown!……please excuse my ignorance. I have hardly any knowledge of electronics at all, and I’m just winging it and learning as I go. It is not so difficult to follow a schematic, but it is an entirely different thing trying to figure out exactly how it should be modified. My plan is to learn that too as I go as well… 🙂

    And most importantly: You are an excellent source for guidance to noobs like me, the info and tips you provide are priceless. It needs to be stated that you have already helped me a great deal, and I’m positive the same is true for a great many others.

    Cheers to you mate…..and thank-you..

  145. Hey Dean

    Thanks for the nice words, you can find Steve Ward on the 4hv.org forums from time to time.

    The audio ground is not the same as the circuit ground, but is instead coupled to ground through C12 in the schematic.

    You are right about the switch being three way, for clarity I should have added a third circle for the off position, for CW it was simply to disconnect any interrupter.

    I started out just as you are doing know, reading about what others made, build it, learned along the way and build more and more. I made this site to give my experiences back to the community so I am very happy that you find it useful 🙂

    Kind regards

  146. Hi Mads,

    Thanks for clarifying for me. I had assumed that was the case, but because I was unsure I just wanted to confirm it. Don’t want to mess up my new shiny awesome coil!!! LOL.
    I would just use a standard 3.5mm headphone in jack for that right? And then sheild the cable.

    How do all you guys take pictures of the coil in action? Every time I try, it is just a blurred mess!! I know a video camera would work well but I don’t have one. I am just wondering what the trick is for a regular snapshot. Does it require a high end expensive fangled camera? I don’t have one of them either!! 🙁

    I will sign up at 4hv.org to thank Steve then too…..

    Thanks again,

  147. Hey Dean

    I just used a regular stereo mini jack cable with 3.5 mm plug, connected to the speaker output from a toy keyboard.

    You can use any camera to take pictures of sparks, depending on the camera quality and options you have to experiment with adding background/foreground light so the sensor have enough light to work with.

    Manual focus set to somewhere in the middle of the spark branches seems to work the best for me.

    I started out with Canon 500D DSLR and that was not even enough for me, so I have a Canon 5D mk3 today, which is as you described a high end fangled camera.

    I do however also use a old Canon IXUS 70 compact camera with custom software installed on the SD card, which gives it more options than my expensive camera, f.ex. it can run scripts with the sensor chip as detector etc. If you have a old compact camera to use for that, you should really check out this software: http://chdk.wikia.com/wiki/FAQ

    4hv.org is a good forum with a lot of knowledgeable people on it, if you do your research and ask questions that show some insight, there is no limit to the help you can get.

    Kind regards

  148. Hi Kaizer

    I’m Jesus an old user that tried for first time to build your SSTC but unfortunately I left this project and now I want to continue from where I left off.

    I did all the steps you said I build it very similar to yours but I didn’t get any spark. And now I asked you my questions that could be the problem, I have in the secondary coil 500 windings, the primary coil I have the same as you and the H bridge works fine because it’s hasn’t broken after tried lots of times.

    Why do you think it doesn’t work, Can you say me your e-mail to send you photos of the Tesla coil and the microcontroler board and have if you see there is anything wrong. I have tried everything but finally I desisted I wish you could be my salvation.

    Kind regards


  149. Hey Jesus.

    I will try to help you all I can, please upload your pictures to http://imgur.com/ and share the links here, that way others can also gain knowledge from the trouble shooting or give you advice 🙂

    Kind regards

  150. Hey Mads! Nice job! I have some doubts about audio modulation, i’m kinda new at SSTC because i have only made SGTC before, and i pretty want this one for playing music. So in your schematic, there is an audio input, this audio input is an 3.5mm audio input?
    And if it does, that mean if I only connect an mp3 player I can be able on playing music from it?
    As i said sorry for my newbieness but i really want to understand how does the audio modulation works!

  151. Hi Dimitri

    The audio input is meant for exactly s battery powered music player. If you use a regular amplifier fed from mains there could nr problems with the common grounding.
    Just turn up the volume on your audio player for a bigger spark output.
    Use only a audio player that you can risk damaging, there can be transient voltage spikes in the circuit from the Tesla coil.

    Kind regards

  152. Allright! Thank you very much!! Now i’m working on the schematic to print it and make it pretty in a board, and waiting for components to come!

    Thank you again, Mads, and Merry Christmas! 😀

  153. One more question Mads! In your schematic, in the left-top side you have a LM7805 voltage regulator, I’d like to add before it a LM7809 voltage regulator to have a source of 9v for audio modulation site. Can you advice me what kind of capacitator I have to use to get a proper work of it? As i saw in other schematics it usually adds some capacitanse and resistor if needed.
    Thank you very much!! 😀

  154. One more question, hehe, can I use OPA2604AP as operational amplifier instead of LM741?

    Thanks again! 😀

  155. Hey Dimitri

    You can add a 7809 regulator in front of the 7805, just like after the 7805 you should add a 1000uF capacitor. You have have the output from the 7809 feed the 7805 or have them parallel to the 12VDC supply, that is your own choice.

    The OPA2604AP is a much faster op-amp, but you should be able to use it, with this you no longer need the 9VDC supply from the 7809.

    Merry Christmas and please come back with pictures or link to a video of your Tesla coil 🙂

    Kind regards

  156. Hey Mads! 😀 Thank you for reply!! So now i can feed the Audio Modulation circuit with +12 VDC? 😀
    I’ll make videos and photos as soon as possibile, but I think it won’t be soon… Because I will start my exams and I wont be able to end it soon 🙁
    Anyways, when it will be done i’ll make sure sending you videos ^^

    Best regards! Thank you very much! 😀

  157. Sorry one more question xP! You said me that OP2604AP is a much faster op-amp. Is this one better? As slower better? What it makes it better? 😀

  158. Hi Dimitri

    Yes you could feed it with 12VDC IF your 555 can handle the higer voltage.

    LM741 is a old op-amp design, The new one can just switch faster, but it does not matter to us at audio frequencies.

    Kind regards

  159. Thank you very much Mads!! Soon i’ll order all the components and i’ll make the plate. At this moment i’ve got done one SGTC modulated with a flyback transformer so, the sparks are not so big (about 3-5cm) and in my phone the sparks are not well apreciated. So I hope SSTC will be dealing bigger sparks and I can film it properly 😀

    Best regards Mads! A pleasure talking with you!

  160. Hey Mads.
    After being busy i’ve finally decided to take back this project.
    I have noticed that i wasn’t using a proper GDT core, so i bought one that’s identical to yours, but still can’t get the thing to work.

    I can’t go above 50-60 volts because the power consumption becomes insane, even 10-15amps, and my variac can’t handle much, also i have the feeling that if i go further i’d probably blow the mosfets, all i get is some corona from the breakout point (and from the antenna, which is very weird).
    However, i’ve also noticed that making the coil discharge on a grounded wire, lowers the current from the above mentioned 10amps, to 2-3 amps, which allows me to raise the voltage a little more.
    Got any idea what the problem could be? I’ve literally tried everything.


  161. Hey Unleashed

    Are you sure about the phasing of your GDT? You might be turning the MOSFETs on at the same time, it certainly sounds like there is a short circuit / cross conduction what current raising that fast.

    Have you tried to reverse polarity of the primary coil? It could also be a feedback polarity problem.

    Have you taken your MOSFETs out and tested them one by one, out of the circuit, to see if they are really okay? They could have been damaged from being driven through your old improper GDT core.

    Kind regards

  162. Hey Mass, thanks for the reply.
    For the gdt phasing I’m kinda sure it’s right, but if it were wrong, shouldn’t the coil not work at all?

    For the primary polarity, yes that’s sometching that had occurred to me, but my coil is not working at all if I reverse it.

    For the mosphets, yeah it could be, I honestly haven’t tested them, but could it be that they don’t work well? Shouldn’t a mosfet be broken or working? Or there’s actually a state in between?


  163. Hey Unleashed

    You could be driving them both on and all they would do was short circuit the supply, but limit it through their on resistance, up until the voltage is high enough and where you see the current rise fast. Are the MOSFETs heating much when this happens?

    Primary coil phasing is correct if something just happens, when its wrong you can get a tiny bit of break out at full voltage.

    MOSFETs can fail in any state between working and exploded, they can just have half working gates or very high on resistance due to heat damage. If you have had any accident with burned or blown parts, its best to take it all apart and see if there is more damaged components than what meets the eye.

    It is a classic fault that the gate resistors have failed open or have higher resistance and thus does not drive the MOSFET fully on, or on at all.

    Kind regards

  164. Do you think this would be a decent replacement for your set of mosfets UCC37322?

  165. What are your audio+ and -? I am starting by using a signal generator with a signal of 1khz at 1V. If we measure the signal after the Q6 or c15 on your diagram we get a swuare wave centered at the peak of our input sinwave that stretches between 9v and 0v. Is this what we should be expecting?

  166. Hey Arasmas

    The ZTX450 transistor only have a pulsed current rating of 2A, I am not sure if that is enough, it might be. You should however use some base resistors to limit the current so you do not burn them down.

    Are you sure about the two 10R, 100n and 10uF on the output of your transistor stage and how it is connected to the MOSFET gates, B and C both goes to the same MOSFET.

    Audio+ is the normal audio wire and Audio- is the shield/ground. It sounds like it should work with the signal you describe, you could connect a small speaker to the output and just listen if its correct.

    Kind regards

  167. Honestly, I’m not amazingly certain. I have been having trouble getting the lm741 stage working. Why do you keep the antenna stage when you add in the audio modulation?
    The square waves I mentioned were after the BC-transistors. Was that planned?
    What input was this designed for? It seems like 1VPP is needed but my circuit seems very inconsistent.
    Thank you for keeping this site alive. It has been incredibly helpful.

  168. Hey Arasmas

    The coil was made to be able to run in both CW, interrupted and audio mode, I just wanted to have all the options.

    The analogue audio input is chopped into square waves where it will turn on/off the coil to the frequency of the received audio, this is how it works. It might need a relatively high input voltage, I remember having to turn up the volume on input devices.

    I will keep my site up for as long as I can, hopefully that is many more decades 🙂 I am happy you find it useful.

    Kind regards

  169. Hi Mads,
    So this should work if I just use the audio modulation? Even when the switch is flipped it looks like the antenna stage still contributes.
    You have several 10k pots in the diagram. Why are they there instead of a normal resistor? Are they on a particular setting?

  170. Hi Mads,
    Sorry for this stream of posts.
    What is the point of the 2 UCC39322 mosfets? I just modeled both your original circuit and the circuit I posted earlier in ltspice. They have the same behavior after the 555 timer but there isn’t a model available for the UCC39322.

  171. Hi Arasmas

    The antenna feedback is always needed, it is what ensures that the primary circuit switches to the resonant frequency of the secondary circuit. Since the primary circuit is not resonant, this is the way to match the secondary frequency.

    Potentiometer R4 and R5 are for adjusting bangs per second and on-time for the normal interrupter.
    Potentiometer R17 adjusts maximum pulse width of the audio signal, that is the same as on-time.

    You have mistaken the UCC37321/UCC37322 ICs, they are MOSFET drivers, it is a small integrated circuit that has the capability to drive a MOSFET gate with a high pulse current. You can replace them with other MOSFET driver chips or try to build a small transistor driver as you tried, but I can not help you in detail with that without having tried to build it myself.

    Kind regards

  172. Hi Mads,
    That all makes sense now. Thank you.
    Is there a difference between the ucc37322 and the 321 that Im missing? The datasheets make me think they are identical.

  173. Hey Arasmas

    There is a version with inverted output and one with non-inverted output, that way they work as a push-pull driver on the primary winding of the GDT.

    Kind regards

  174. Hey Mads,
    Ah! I see now. So I have built everything up to the controllers and the transformer but am seeing some strange behavior. The output of the 555 timer in the audio modulator is producing square waves with very odd edges. They alternate between large waves and then smaller ones( with about 50% of the width of the larger). Is this what yours look like? Do you see any output waves from you discharge? Spice Predicts the one I have uploaded but I don’t see anything.
    Finally you have 15:15… above your transformer. How is this actually wrapped? Is each output and the input wrapped around the toroid?
    Thanks again for the help.

  175. Hello Mads,
    Final Question, I think. For your bridge circuit I don’t see any ground at all. Is that correct?

  176. Hello Mads,
    I wish I could edit my older comment because I do have a follow up question about the Potentiometer. They look like they could be replaced by resistors because I don’t understand the configuration they are in. Is port 7 of U6 attached to the wiper of R17?What about R4 and R5?

  177. Hey Arasmas

    You can find a detailed describtion of how the audio modulator works here: http://www.ctc-labs.de/html/body_musical_drsstc.html it is in German, so use google translate to read it.

    Scroll down to 25th September 2009 and you can see some details on GDT construction: https://kaizerpowerelectronics.dk/tesla-coils/kaizer-drsstc-i/

    The bridge is not supplied through a transformer, grounding the negative rail of the bridge supply should only be through a small capacitor in the nF range. But that is not necessary with a SSTC.

    I explained in a earlier comment what the different potentiometer do, the wipers are all connected to one of the other legs on the potentiometer, it only has to work as a variable resistance and not a voltage divider. If you replace them with resistors, you can not vary bangs per second and pulse widths.

    Kind regards

  178. is it a must that i have to build a coil [secondary and primary] of the parameters that u have specified or can i just build a coil using Tesla map or other software and use the driver circuit and bridge u have displayed on the page
    and yes i have already built the coil its secondary is about 6″ long and 2″ in diameter
    i am 16 years old and i am searching for all the info i can get
    and i hope i can get the answer here

  179. Hello Aditya

    If you have used a small enough wire so that you stay below 400 kHz resonant frequency of the secondary with topload, then it should be perfectly fine to use a smaller secondary with this circuit. A frequencies above this you will have problems with excessive heating of the switches spending too much time in the linear region.

    Kind regards

  180. my secondary coil with its topload has a frequency of 350 khz
    thank you for your help and time sir

  181. and sir i also wanted to know if i could use the the same gate drive circuit for a drsstc
    instead of a sstc
    according to what i know the difference between a sstc and a drsstc is that a drsstc has a capacitor in series with the primary to increase the current and to prevent overload of current it is interrupted and driven in pulses
    so if i have a sstc with an interrupter and i add a capacitor such that the resonant frequency of primary matches the secondary it would become a drsstc
    please do correct me if i am wrong sir

  182. Hey Aditya

    You are correct, you can convert a SSTC to a DRSSTC like you described.

    However, you do not have the precession switching of current transformer feedback or over current protection that is the two largest differences between the two drivers. Without these measure you are most likely just going to explode your transistors from hard switching too high currents or from excessive heat dissipation.

    If you want to build a DRSSTC, at least use the simple universal driver 1.3 as the one I used for the my DRSSTC 1.

    Kind regards

  183. Hello Sir.
    Thanks for the great article. I am working on this coil and I had a question. Did you use any sort of inrush current limiting in this circuit? I am planning to use a 680uF 450V capacitor and since it will act as a dead short when discharged, will I need any inrush current limiting? I have started a thread at 4HV here: http://4hv.org/e107_plugins/forum/forum_viewtopic.php?172579 Could you please take a look?
    Thank you.

  184. Hey SK1701

    I use a variac to bring up the voltage slowly when powering on most of my Tesla coils, this makes it easier to spot any fatal errors from watching the current draw. This slowly ramping up of the voltage also does that I never have to deal with inrush current to the smoothing capacitors.

    You could use NTC resistors to limit the inrush current.

    Kind regards

  185. Hi Mads,

    I was thinking if I can use a IR2104 and IR2101 instead of the UCC3732X drivers. I couldn’t find then…


  186. Thanks for the reply. I have a couple more questions. I did not realise that I need a length of PVC longer than the secondary so I cut mine to 30cm and am almost done winding the secondary. However, I have realised that there is not space now to wind the primary below the secondary. Can I wind the primary directly over the bottom part of the secondary (with some insulation)? Would Kapton/ duct tape work as insulation or do I need a good plastic sheet?
    Also, how would I go about attaching the toroid to the top of the secondary? Can I make plastic end cap over the pipe with a bolt to attach the toroid like you did? What is that plastic ring used for? And how do I connect the bottom of the coil to ground?

  187. Hey SK1701

    The primary is wound directly on top of the bottom windings, it is not alone on its own piece of the secondary form. Kapton tape, electrical insulation tape or plastic sheets work fine as a isolator, just be sure to wind on a few centimeters above the primary coil, I had some flashover because I did not do that. Duct tape could contain some metallized form of plastic, so avoid that.

    The plastic end cap with a screw through is the easiest solution for a secondary termination where you can also connect a topload. If it is possible, do not have any metal on the inside of the secondary, try to isolate it with another plastic disc, epoxy or what you can find. The ring I have on mine is just for making some distance to the topload, so that the bottom of the topload aligns with the top windings of the secondary.

    You should connect the secondary ground to the same ground as the rest of the coil is connected to and if possible have it all grounded to its own earth ground rod. You can also use house ground, but there is a risk of introducing large amounts of RF into your other household items.

    You can also make your own artificial ground from sheets of aluminium foil covering the table the coil is standing on. As I did here: https://youtu.be/URfGo21smqA?list=PLCDCF7ED07B7628B8

    Kind regards

  188. Hey Vinicius

    The IR2104 is a single half-bridge high side driver. You would only need one, and not the IR2101 also, you would also have to omit the gate drive transformer. But beware that you no longer have the galvanic isolation between the high voltage on the bridge and the low voltage on the control logics, now it is only a small IC and a diode that separates the circuits.

    I think it was necessary for me to order the UCCs from USA to find them at a reasonable price or even just find them 🙂

    Kind regards

  189. Hey im a Fan of your work and I´m building your SSTC II Tesla Coil with my Teacher. Im very impressed by Teslacoils and I always wanted to build one since im 10 Years old (im 19 now). Do you got any Tipps or Tricks for an Newcomer to get a quiet impressie Outcome?

    PS: I added the actual Version of the Driver Board.

  190. Hi Hidan

    Thank you for the kind words and I am happy you use the documentation here to build your own.

    I think you should have the traces from the MOSFET driver ICs larger, the whole output circuitry to the GDT could use some more copper, it is afterall 9 Apeak drivers, just so you do not waste some of it as heat.

    The trick with SSTCs are to use as large as possible topload, high coupling between primary and secondary coil without flashover, antenna in a position where it does not arc and as high as possible input voltage to the bridge. It all requires experimentation.

    Good luck and please post your results when you have some 🙂

    Kind regards

  191. Hi sohaib

    I do not have any experience with low voltage supplied Tesla coils and I am very uncertain on what you are really asking?

    Kind regards

  192. good day sir i was wandering what can i use instead of a variac because i live in south africa and they very expensive if u can find them

  193. Hello Wikus

    For SSTCs a simple voltage regulator as you know it from fan or light bulb damping can be used. You can either buy a triac regulator from ebay or build your own like f.ex. this: http://danyk.cz/triako_en.html

    Search google for sstc triac voltage regulation to get more projects. I have not tried it myself.

    Kind regards

  194. Thanks Mads will check it out also my tesla coil is nearly done hope to start up this week end will post pics if all goes well
    thanks wikus

  195. hi mads
    also i used a 1.5 uf in the place of c8 and c9 is this okay
    and does c5 and c6 have to be ceramic or is poly okay
    here are some photos of what i have build so far the gdt doesnt work i found the toroid in a motor drive
    again thanks

  196. Hi wikus

    Your PCB layout is very bad, honestly.
    – you have sharp edges.
    – large empty squares not connected to anything, if anything have a ground plane or at least those parts connected to ground.
    – decoupling capacitors for the ICs should sit as close to them as possible across the positive and negative supply rails.
    – you have traces and pads that are so close that will have trouble with short circuits.

    You are fine with 1.5 uF in place of C8 and C9, C5 and C6 can be MKP plastic film capacitors without problems. It is tantalum capacitors and not ceramic capacitors that I used.

    I assume the picture of your GDT does not show the finished one that does not work, because the one you show me is not done. If the toroid comes from around a cable for noise suppression it is most likely no good for use as GDT.

    Keep reading, asking and building, you will get there sooner or later 🙂

    Kind regards

  197. Hi Mads
    Thanks for the input will fix the pcb and the gdt was only to test if it works
    Will keep posting and keep you up to date
    have great new year

  198. Hi wikus

    If you look at my PCB designs from the SSTC’s to the DRSSTC1 and DRSSTC3 you can see that my own PCB layout have improved vastly over the years.

    Kind regards and happy new year

  199. Hey there again 😀

    I wanted to ask you if you have got a Block diagramm of your SSTC II you can share with us? 😀

    would be great 🙂

    MfG Hidan

    PS: I addet a Example in the Attachments

  200. Hi Hidan

    You already have a block schematic that shows exactly what is going on, just add a switch between music and interruptor for the osc. box and write full bridge instead of half bridge 🙂 You should also add a antenna from near secondary coil to go back into gdt driver box.

    Kind regards

  201. hi Mads

    Me again my coil is done and works ok but some issues have poped up.
    Firstly it doesnt want to run on CW but works fine interupted mode
    and secondly it keeps burning one 4k7 ohm resistor. I think mabe i lost count on the GDT that mabe its one turn more than the others.
    Please give me some input
    also i will post picks as soon as it is running fine and neatend it up some

    thanks again

  202. Hi Wikus

    My method of leaving it open for CW mode is not reliable at all anyway. Just pull the enable pin of the UCCs high with the +5V and it should run in CW mode.

    If you have a dodgy GDT and you suspect that one gate is driven with less voltage than the other, that could have lead to that MOSFET only being driven in linear mode, it could have damaged it and is not completely dead, but damaged enough to draw so much current from the GDT drive signal that the gate resistor burns. Check that all your MOSFETs are okay. Does it happen on another MOSFET if you switch the GDT wires around between the MOSFETs?

    Kind regards

  203. Good evening friend, very good his post on the SSTC, I’m trying to make one too and I have many doubts, the first starts at the bridge inverter, it must be done with MOSFETs or can be made with IGBT transistors, in my case here IXXH100N60C3 I can use this transistor in place of IRFP460 you used.? I hope that this will translate correctly need a translator because my language and Portuguese. From already thank you very much .

  204. Hi Alexandre

    Do not worry about having to use translator, I understand your question perfectly fine.

    You can replace the IRFP460 MOSFET with IXXH100N60C3 without changes to the control circuitry. But you need to add a external anti-parallel diode as the IGBT does not have these compared to MOSFETs.

    Kind regards

  205. Thank you for your answer, but I will put the diode, it would be the collector and emitter of the IGBT.? These types of IGBTs can be used also in DRSSTCs why came to me doubt if I could use in a SSTC,

  206. Hi Alexandre

    Just like the diode is between drain and source on a MOSFET, so should you place it between the collector and emitter of the IGBT.

    Kind regards

  207. Ok thank you, was watching your bridge diagram reverse and has no zener diodes in the gate to the source. What about the driver needs to be within a metal box to the field coil does not damage the components.? It must be grounded as well.?

  208. Hi Alexandre

    The driver is in a grounded metal box to avoid noise problems from both the switching of currents and mostly the strong interference from the secondary coil. I would not go as far to say it will be damaged, but you might experience that it will run with spurious faults to maintain oscillation.

    Kind regards

  209. Yes I understand, so it works like a Faraday cage to protect the driver.? And one more doubts I have about it, to land and just call the negative rectified or needs to be grounded to earth electrical network.

  210. Hi Alexandre

    You should only ground the negative DC bus to ground if you feed the input rectifiers through an isolation transformer.

    You can add a 1 – 10 nF / 2kV MKP capacitor from negative DC bus rail to ground for decoupling and protection against primary strikes. That is however not necessary in SSTCs as in DRSSTCs.

    Kind regards

  211. I understand, if in my case I use a variac not poso land, only if I were to put an isolating transformer before variak, but I will do this, I will put this capacitor also to help protect the system, thank you for everything,,

  212. Hey Mads

    Back in November 2015 you told me my circuit would need more copper but in my case i don´t need it at all because its galvanic isolated from the bridge so i wont get an 6A peak at all.

    But still thanks for your Advice

    Yours Hidan

  213. Hi Hidan

    Thin traces, magnetizing current of your GDT and so on will all limit the possible peak current and you might end up with a too weak gate drive. I suggested thicker and aligned traces could be a good idea when we want to conduct high peak currents, in order to minimize losses and parasitic inductance.
    I recently uploaded a new calculator to my site, a gate drive calculator, you could try to play a little with it with your numbers and see how it checks out 🙂


    Kind regards

  214. Hey Mads

    Since im new to this tesla coil stuff i would like to get some informations about it.

    May you can explain me what the UCC 37321/37322 and the 74HC14 do?

    and may you can explain how the Modulator works? 😀

    would be nice if you do 😀

    Yours Steph

  215. Hi Steph

    The UCC3732X are MOSFET driver ICs, one non-inverted output and the other inverted, in order to get a push-pull drive of the gate drive transformer. A gate driver IC can deliver the high peak currents needed to drive MOSFETs efficiently.

    The 74HC14 is a inverting hex schmitt trigger, it is used to get a proper solid 0-5V square wave signal from signals that are not perfectly square, the antenna feedback can vary a lot in waveform and amplitude, the 74HC14 converts this to a clean drive signal for the MOSFET drivers.

    The music modulator works by amplifying the audio signal in the LM741 and at the BC547 transistors. The 555 timer ensures that the signal length of the generated square wave is much shorter than the audio signal, in order to not have too long on-time and thus damage the MOSFETs / IGBTs from over-current.

    Kind regards

  216. Im not sure if i build myselfe a teslacoil

    i just want to understand how they work 😀

    by the way i recognized that you answerd to a german post, do you speak german did you use a translator?

    Yours Steph

  217. Hi Steph

    I speak can speak and write German well enough to work with technical matters 🙂

    I used to work for a German company and we have it as a obligatory language to learn in the Danish schools.

    Kind regards

  218. Nur mal so ne frage:

    Wofür ist das Potentiometer im Modulator gedacht?

    MfG Steph

  219. Hi Steph

    Your English was fine, why not keep it in English for others to understand the questions too? 😉

    Steph asks what the potentiometer in the audio modulator is for.

    The potentiometer adjusts the on-time of each tone, depending on the music or how long sparks you want, this is a way to adjust that.

    Kind regards

  220. Hey Mads

    I once asked you about an Block diagramm and you said my Example is good to go but someone Told me the coils are wrong because they are connected to the ground. actually i dont get it. where should my coil end in?

    I hope you can help me.

    Yours Hidan

  221. *Edit: is it caused because you used a fullbridge instead of a halfbridge? if yes do i have to connect the bottom ends of the coils to the bridge too?

    Yours Hidan

  222. Ah ok I´ll try my best to keep writing english 😀

    Uhm i got another Question:

    which type of Modulation is the Modulator?

    AM or FM?

    MfG.: Steph

  223. Hi Hidan

    Your block diagram is not wrong, it just represents a half-bridge. In a half-bridge you would have the coil connected between the two switches and the other end to ground.

    In a full-bridge the coil would naturally be connected between switches of both legs and have no relation to ground.

    The negative rail of the bridge supply should not be grounded, but you can decouple it to ground with a low capacitance capacitor.

    The secondary coil should always be connected to ground!

    Kind regards

  224. Hey Mads

    If I want to test out my Driver i have to Simulate the Antenna with an Oscillator but how should i Attach it to my Driver?

    I mean i got 2 plugs from the OSC but only 1 Antenna input.

  225. If I attach the Signal between Antenna and ground i got the sinus from my OSC on the Driver Output (I simply put a 330 Ohm Resistor on the Output of the driver)

  226. okay i got some messures:

    OSC = 10kHz Input Sinus
    +Uin = 12V
    Iin = 15mA? (i think it was higher on the version that works)
    Output = 13kHz Sinus (if i disconnect the OSC / Ground Connection i get a 22kHz Signal but it has no clear form its just random)

    I already tryed to replace the 74HC14 but with no effect.

  227. Just the 74HC14 alone should results in a nice square wave from a irregular input waveform of varying voltage and shape. As long as it is above its threshold level, there is no possible chance you should get a sinus wave output, unless its defective.

  228. Well i replaced all of my IC´s and it seems that my problem are not my IC´s because for about 5 seconds after i turned on the voltage my Driver works perfectly well but after 5 secs its only a sinus.

    Btw i removed the connection OSC / Antenna input to see if it shows the Sinus again and it does. So i think the Sinus comes from the GND connection. is this a possibility?

  229. So you have a square wave output from the driver ICs for 5 seconds and then it is suddenly sinus?

    Beware that you can not run the driver ICs unloaded, they might self-oscillator and burn out, do they get warm?

    Just looking isolated at the 74HC14 you have an input and output, with a common ground to measure input and output against. Following the schematic you should measure a square wave coming out of the 74HC14 when feeding it a sinus. Your circuit is not correct if you get a sinus on the output.

  230. On the 74HC14 Output i got an Continous High output and the ICs dont get warm at all.
    It seems the last thing possible ist that the UCC37321 and 37322 are dead because I got no IC´s to replace them anymore.

    Im 95% sure the Sinus comes from the GND attachment.

  231. A continues high output on the second 74HC14 output is either caused by it being defective or your oscilloscope is not fast enough to see the switching.

    You should get a square wave with the same frequency as sinus wave input.

    The 74HC14 is a inverting hex schmitt trigger and if you are only using a single gate of the IC your signal gets inverted and will properly result in a non working circuit.

  232. Phil Taylor

    Hi Mads

    I am new to sstc’s and would like to build one. I have searched the Internet for a suitable design and finally discovered your site and as far as I can see your Kaizer SSTC II and Kaizer SSTc bridge are exactly what I am looking for.
    However, I would be very grateful if you could please give me a little advice regarding the construction of the GDT. Also are the mods you can recommend before I start construction?
    Kind regards

  233. Hi Phil

    The SSTCs I built are based on the design of Steve Wards SSTC5, so you could also take a look at how he constructed his coils.

    The easiest way to construct a GDT is to find a f.ex. N30 material ferrite core, or other with a permability around 5000, and use CAT5 network cable to wind 15 turns on the core. Now you can connect all the white/striped wires in parallel to make a primary coil of the GDT and each of the four colours can be made into two or four secondary windings, depending on you using a half- or full-brigde.

    I have not yet finished my own GDT guide, so you should check out this excellent guide: http://thedatastream.4hv.org/gdt_index.htm

    Kind regards

  234. Phil Taylor

    Hi Mads

    Thank you very much for your reply and advice. I feel I can now have a go at building the GDT and will certainly check out the 4hv site.
    Once again, many thanks

    Kind regards


  235. Hi Phil

    4hv.org is a great high voltage forum and community, if you show interest, give a lot of details, show pictures and ask specific questions, you can get a lot of help and experience from there 🙂

    Kind regards

  236. Phil Taylor

    Hi Mads

    Once again, Many thanks for your help.
    From what I have found out about building sstc’s, particularly music devices, any help, advice, or assistance I can get will be very useful to me. I do appreciate that many of my queries are probably very basic so I do thank you for your patience in taking the time to answer them.

    Kind regards

  237. Phil Taylor

    Hi Mads
    I have now built my GDT as per your details. When checked out with my ‘scope I was very satisfied with the resultant waveforms.
    Your idea of using two frying pans for the topload is really very good and I shall be building mine in the same way (frying pans arrived this morning!!).
    One query I have at present is that I am not sure what the purpose of the Antenna is, where it should be placed and how to adjust it.
    Finally – for the moment, having read through the above I understand that the SSTC should not be started on full mains power if running in continuous mode due to the how power consumption, however, is it OK to start on full mains power if running in interrupted mode or audio mode.

    Kind regards

  238. Hi Phil

    I am glad to hear that you get good performance out of your GDT.

    The frying pans are perfect, with their smooth surface and often low cost in stores like Ikea.

    The antenna should be so close that it runs stable, this changes on how long the sparks are. It should also be so far away that there is no corona forming on top of it, so it takes a bit of experimentation to find a good spot for it. Try to start with a antenna that goes 1/4 of the secondary length up and maybe at a distance of 15-20 cm.

    Full mains in interrupted mode should be no problem, unless your coil have issues with flashovers in the first place, but you will find out about that if you first ramp it up with a variac, or some people have used dimmer controls for small coils.

    Kind regards

  239. Phil Taylor

    Hi Mads

    Once again many thanks for your help.
    Thanks also for the links to your new Youtube and Google + chanels. For a budding sstc experimenter they are very useful, particularly with their associated comments and links.

    Kind regards

  240. Phil Taylor

    Hi Mads
    Yes subscribing has certainly given me access to a lot more information that I otherwise would have been lacking.
    I am sorry but another silly question. Having read back through this forum it is clear from your answers on the operation of the various parts of the circuit, that for only running CW, you only need the upper part of the schematic. Lower left is the interrupter and lower right is the audio modulator, my question is this. Is the switch that selects either audio or interrupted mode a centre off switch, with the centre off selecting CW mode?
    I have also seen mention that some people seem to use a heat sink with the GDT, is this necessary with this circuit?

    Kind regards

  241. Hi Phil

    To run it in CW, you should tie the ENBL pins to +12VDC (ENBL input is limited to Vdd+0.3), mine worked for some reason when it was disconnected, but that is not stable.

    There should be no heat sink on the GDT, what do you mean? Could you give me a link to a example?

    Kind regards

  242. Phil Taylor

    Hi Mads

    Thanks for that, it does clear it up as I did doubt if the enable should be left floating. It makes much more sense to tie it to +12VDC.
    I will send you the link regarding the heatsink for the GDT as soon as I can find it again.
    Again, many thanks.

    Kind regards

  243. Hi Mads

    I am a little confused regarding the earthing of the secondary coil. I have had a look on 4HV but am still not sure. Is it OK for me to use the mains earth or should I use something like 4.0mm Stranded Conductor (7/0.85) attached to an earth rod.

    Kind regards

  244. Hi Phil

    The best solution is without a doubt a separate ground rod, this way you are sure to have as short as possible connection to earth and there is reduced risk of damaging your other plugged in house hold items.

    I have however used mains earth for secondary ground at up to 2 kW input power and did get some arcing in a distribution bar between ground and phase/neutral. No house hold items was damaged.

    Kind regards

  245. Hi Mads

    Many thanks for that, I think that I will defiantly use a ground rod as I do not not want to risk damaging any house hold items. I also understand that a ground rod is even more important when running drsstc’s.
    Kind regards

  246. Hi Phil

    The ground peak currents are higher in a DRSSTC, so it would result in higher amplitude noise on the mains ground than if it was a SSTC.

    Kind regards

  247. Hi Mads
    I have now accumulated all of the components and hardware for my SSTC II and have just completed the construction of my coil winding machine.
    I have also catered for an optical link between the audio modulator and the MOSFET driver.
    My question is:- Is it possible to have the ground of the schmitt trigger / MOSFET driver circuits connected to the equipotentiality point (-ve) on the interrupter / audio circuits.

    Kind regards

  248. Hi Mads

    Would it be possible to use 1.5mm PVC covered copper wire for the primary coil. If so would I have to alter the number of turns.
    Kind regards


  249. Hi Phil

    Do you mean 1.50 mm^2?

    I used 2.50 mm^2, even that could get warm at high power, so you would be limited by a smaller wire.

    You could use a few more turns to lower the primary circuit current through higher impedance from more wire. There are no resonant frequency matching to be done in the primary circuit, so you do not really need to worry about number of turns, only in regard to primary current.

    Kind regards

  250. Hi Mads

    Many thanks for that.

    Yes I did mean 1.5mm^2
    However, I will now use 2.5mm^2 and keep it to 8 winding’s
    Kind regards


  251. Hi Mads
    i’m interested in D1 and D2 (1N60), what’s there role in this circuit and can you recommend any other diode to replace this? would any other schottky diode do the job?
    I’m also interested in the amplifier, LM741, i can’t seem to find one in my area ( local stores), just like those 1N60 diodes, so will TL071 be a good replacement for it?

  252. Hi George

    D1 and D2 limits the antenna feedback to be within the voltage range of the 5 VDC supply rail, you could replace them with some other fast small signal diodes like 1N4148 or 1N4448.

    Compare the datasheets to see if you can replace one op-amp with another. Too see if supply voltages, pin-out and amplification, drift, noise etc are fairly the same.

    Kind regards

  253. Hi Mads
    I think I’m gonna go with the 1N4148, mostly because my local stores have that one, i don’t know about the other one. about the amplifier, as far as i’v searched, TL071 seems to be the ideal one for replacement.
    I have another question, or two:
    1) about the reverse diodes(D4, D5, D8, and D9, (MUR1560)), as I know(Data sheet says so) those mosfets have built in reverse diodes and why are these for? additional protection against reverse emf?
    2)about your primary coil, judging from the resonant frequency it’s inductance should be around 298.003 nH right?
    3)and antenna’s there for the drivers to self tune at the resonant frequency right?
    4) would you recommend to use voltage regulators? i mean other then the 5 Volt regulator(U1), would it be better to use 9V and 12V regulators to feed the interrupter and the audio modulator?
    5)would it be better if I use zenner diodes on the power mosfets? not that they need one but, just in case…

    kind regards

  254. Hi George

    1) You need faster and larger external diodes as the internal in a MOSFET is just a bi-product of manufacturing the MOSFET itself. It it not very fast and thus we need to bypass it.

    2) There is no resonant primary circuit, see 3.

    3) Correct, the antenna picks up the resonant frequency of the secondary circuit and switches the primary circuit accordingly.

    4) All control voltages in my coils uses voltage regulators.

    5) It is better to add snubber capacitors to take care of switching transients than zeners, TVS could also be used, but for the very fast transients, only capacitors really do the job.

    Kind regards

  255. Hi Mads

    Thank you for all the answers, but i still have one more question, just to be sure.
    judging from the 3rd answer(about antenna/self-tuning) i’m not restricted in choosing the capacitance of C8 and C9 right? changing there capacitance will mainly result in the change of the resonant frequency and that won’t be a problem because of the self-tuning nature of there circuit right?
    what if i also tune the primary circuit to the secondary’s resonant frequency? i would end up with the DRSSTC instead of just the SSTC right?

    Kind regards

  256. Hi George

    C8 and C9 are a part of a voltage splitter, they are substitutes for the missing transistors in a half-bridge, but also really only necesary if you wanted to feed the coil from a voltage doubler circuit. You could actually just omit the capacitors and tie that end of the primary coil to the negative rail.

    The primary circuit in a SSTC is NOT a resonant circuit, it merely oscillates to the resonant frequency of the secondary circuit!

    You can not tune the primary circuit of a SSTC. If you however do add a series capacitor with the primary coil, it becomes a DRSSTC and you will have to tune it.

    Kind regards

  257. Hi Mads

    Thanks for all the answers, again.
    Could you please tell me, or give me a link, how can I calculate correctly the values of R5 an R4? (On and off tume depends on the mosfets right? And how much they can or can not handle?)
    And could you recommend some website about GDTs?
    And what possible problems can there be if I tune the primary circuit and chenge this into DRSSTC? Some additional changes may be required right? Like current transformers on the primary an mosfets with greater current rating?

    Thank you for your time and Kind regards

  258. Hi George

    You could use many different 555 calculators found on the internet, f.ex. http://www.talkingelectronics.com/projects/555/Page2-555.html it also explains about the diodes.

    This is the best GDT guide I know of: http://thedatastream.4hv.org/gdt_index.htm

    MOSFETs are not suitable for DRSSTC use as their conduction losses are very high due to their on-resistance R(ds), whereas IGBTs only have conduction losses due to their CEsat voltage drop. The MOSFET has squared higher losses than the IGBT.

    Current transformers for over-current detection are needed to protect your IGBTs from exploding 🙂

    Kind regards

  259. Alexandre Kreuz

    Good evening friend, and there must use these resonance capacitors in the primary there . ? You have two connected in parallel .68uf . to be this or capacitance may vary. ?

  260. Hi Mads

    Thank you very much for all the answers
    this will be my first SSTC so i hope everything goes well(if not i’m gonna return with questions) 🙂

    Kind regards

  261. Hi Mads! First, thanks for documenting and posting your projects online. You are a great deal of help, and the fact that you are so active in helping others means a lot.
    I am currently in the “planning phase”, and i had a question on how one would build an effective GDT

    From what i understand:
    you first take 5 tiny issolated wires, and wind them together into one wire. Then you wind the wire 15 turns around the core, and then i am lost..
    How do you determen the polarety? Plis explane in details 🙂

    King Regards

  262. Also, forgot to add, i was wondering if i could maybe use a 300uf 450v as a smoothing capacitor as i cant say i have a big 1500 uF/450 V capacitor around. And i dont have a variac laying around, so would it be just fint to run 220v without slowely increasing the power? 🙂

  263. Hi Magnus

    Thank you for the kind words 🙂

    The easiest way to wind a GDT is to use regular network cable, here you already have the wires colour coded.

    Mark all the wires in one end of the cable with a black marker, this is how you know the polarity later when its winded.

    Wind the cable 15 turns around a ferrite ring, strip the outer insulation of the cable and now you can take the two wires of same colour and twist them into a pair. Black mark for polarity. Do this for the remaining 3 coloured wire pairs.

    Now take all the white/coloured wires and twist them together into one big wire, this is the primary winding and with its larger gauge perfect for the higher currents flowing there.

    About the DC bus capacitance, 300 uF might not be enough to make long sparks at high BPS, but it will work fine nonetheless.

    A variac is a great help to avoid catastrophic failures. Lets say that you adjusted the antenna too close to the secondary coil, with a variac you would be able to observe corona forming on itand stop before failure. With full voltage on from the start, you would risk huge sparks, flash over and possible a ruined secondary, primary or blown MOSFETs from bad feedback. So a variac just makes life easier 🙂

    Kind regards

  264. Thanks Mads for your detailed reply 🙂
    Here in Norway, we use the same power system in the USA, so the power insted of 220v out of one wire, we have 110 on 2 wires, comming together and making 220v. So i can test the circuit on 110v first 🙂

    I wonder, how long is your antenna? And where do you place it to recive the best signals without disturbence?

    I will try to get it finished in 2 months, but cant be certain because of work, but if it workes i will send you a picture 🙂

    Kind regards

  265. Hi Magnus

    You can see the antenna placement in the pictures, it is hosted inside a small acrylic tube.

    It would be a good idea to make a switch for 110 and 220 VAC operation, the low voltage will give you a good indication of how well its running 🙂

    I am looking forward to see your results.

    Kind regards

  266. Hi, can I use this ferrite core for GDT? 🙂 Or can u recommend me something better? But it must be on GES.cz website, because im from Czech Republic and import taxes quite kill me when buying UCCs 🙂

  267. Hi Leni

    I have used the N30 material before with great sucess, but these ring cores on that website is not very big, it will hard to fit 15 turns of CAT5 network cable on it.

    Look at the material specification sheets from Amidon to see see which material types are used for which frequency region and application. http://www.amidoncorp.com/product_images/specifications/2-02.pdf

    Material type 77 is the best available, permability around 3000, I would advise you to choose one of these, but take one that can actually fit 15 turns of CAT5 cable in the winding window 🙂 http://www.ges.cz/en/electronic-components/cores-chokes/ferrite-cores/e-type-cores-material-77/SAG31.html

    Kind regards

  268. Thank you for reply, it can be small, i can get two cores and connect them in parallel 🙂 I have one question, why E type? Is it better than torroid? 🙂

  269. Hi Leni

    I prefer toroids, no possible air gap in those. E type is much easier to wind, not a particular problem with a 15 turn GDT however. I chose the E type because, in my opinion, the type 77 toroids you linked was very small.

    Kind regards

  270. Hi there, so if i buy 2 toroid cores, connect them in parralel and wind in trifilar it would be ok? I now found that i have not enough space for E type so.. 😀

  271. Hi Leni

    Yes, you can side-by-side toroids and wind the transformer around both of the cores and they will function as “one” core.

    Kind regards

  272. I mean that there will be two cores and one be trifilar with primary and 2 secs for half bridge and the other will be also trifilar with 1 primary with 2 secs for the rest of full bridge 🙂 By the way… will be 470uF enough for bus cap? 🙂

  273. Hi Leni

    470 uF is enough, it is in the low end, but it is enough for it to work, you might see limited performance in full CW mode.

    Kind regards

  274. Hi Mads

    i tested the controls circuit without the GDT and i can’t seem to control UCC3732x, they are both on, at the same time, only difference is the pulse width, one has slightly wider pulse than other, so i think i should replace them both because they might be busted.
    Datasheet says that UCC37322 should be off when theres logical 0 at the enable pin, but my ic doesn’t seem to care. If replacement is the only option here, can u tell me if this IC has any analogs.

    Kind regards

  275. Hi George

    The UCC3732X IC’s should not be run open loop, they need a load. They can enter a high frequency oscillation if not connected to anything and will destroy themselves.

    I think its the MAX4420 that I have also used, try to see if they can be used. They have a weaker peak current drive.

    Kind regards

  276. Hi mads

    So now they are already destroyed right? :d
    Just to be sure, the signal is ok, only issue is they are not responding to logical 0 or 1 at the enable pin, they are running all the time, so i must replace them or can they be fixed?

    and can u breafly exlpain to me the operation of these drivers? for example can’t i break there connection via the primary coil of gdt and give them there own GDTs? will it work that way?

    Thank you and kind regards

  277. and one more thing…
    i think it’s just gonna work only in cw mode, no matter if the enable pin will be connected to the interrupter or to the audio modulator, or am i wrong?

  278. Hi George

    If they do not respond to 1 or 0 on the enable pin, I assume they are defective. Constant high output is a fault I have seen with them before.

    You can not use a CT on the primary for primary current feedback as the SSTC does not have a tuned resonant circuit, but it follows the resonance of the secondary system. You could possibly use a CT at the secondary base to use base current for feedback, but this solution is just as “good” as the antenna feedback.

    To work in CW mode, the enable pins just be activated for drivers enabled all the time, so its always just running.

    Kind regards

  279. Hi Mads

    yes that’s what i meant, they do not respond and they are running all the time, so that is practically the CW mode, so until the new drivers arrive i’m gonna have to let them work in CW mode. interrupter and audio modulator work nice, so i’m just waiting for the drivers and i’ll be able to fully use your circuits functions, thank you.
    what if i ”measure” the resonant frequency of the secondary and by this frequency tune the primary? will the mosfets hold?

    i’m kinda amazed how sensitive is the antenna and self tuning part of the circuit, and these drivers as well 🙂

    Kind regards

  280. Hi Mads

    I have not connected anythings except for the GDT, interrupter and audio modulator both are ok and working nicely, but the drivers don’t react neither to the interrupter signals, nor to the audio modulator (thus i think the drivers are the problem) so i’m guessing that if i connect this to the Bridge and drive my coil, it will run CW mode right?
    sorry for the late reply
    i will look into your long artice 🙂

    Kind regards

  281. Hi George

    It will only run in CW if the drivers are able to respond to input and deliver a oscillating output between the two drivers to drive the MOSFETs through the GDT.

    Kind regards

  282. Hi Leni

    Yes you can use X2 MKP capacitors, they are designed to counter EMI on mains input. So choose as high capacitance (around 4.7 uF) for the lowest possible impedance at the resonant frequency of the coil.

    Kind regards

  283. Hi Mads,
    why that high capacitance? Is big difference between MKP10 and MKP62? 🙂
    Thank you 🙂

  284. Hi mads

    So Finally i have connected everything… For a moment, about 5-10 seconds, it it worked kinda nice.since my UCC3732x mosfet drivers ar always running, i had to test this circuit in CW mode, around 40V or so…. after 10 seconds one of my mosfets (Q1) went out with a bang and nice fireworks. what do you suggest should i replace only the damaged mosfet or others as well?
    what could have coused two mosfets(Q1 & Q2) to turn on the same time and short out? UCC3732x malfunction or maybe the signal from the antena wasn’t good?
    i think it either should have been the drivers or maybe GDT, but mostly drivers…

    Kind regards

  285. and one more thing, drivers were not shielded, hole board wasn;t shielded and at the top of the antenna i also had sparks, guess secondary induced those….do you have a guide about proper shielding?

  286. Hi Leni

    Maybe I was a little too fast, there is not a big difference, MKP62 is just rated for mains line use as a X2 capacitor. Higher capacitances have much lowered working voltage deratings at higher frequencies, so that might be worth looking into too.

    Kind regards

  287. Hi George

    You need to at-least check all the MOSFETs and gate resistors. GDT will almost always survive and driver circuits are also in a very safe place due to the isolation from the GDT.

    Only mis-wiring of the GDT secondary side could have caused them to turn on at the same time, think about it: the magnetic field in a transformer, as the GDT is, can not be at opposit polarities at the same time, this is one of the inherent advantages of a GDT, there is built-in dead-time when the magnetic field switches.

    Bad signal from the antenna would only make it work sputtering or unstable, not explode.

    Do you have a oscilloscope to check out the phasing of the GDT, that drivers drive the GDT correctly etc?

    Shielding should not be an issue, only to make it sputtering or unstable. The part of the guide on EMI and interference is still in the work and far from finished 🙂

    Kind regards

  288. Hi there,
    so I can safefully use 0.68uF MKP62 caps? I found them in my desk and I do not want to buy new (they are little bit “expensive” 😛 ). Can I use 0.33mm thick wire for secondary coil without problems? 🙂

  289. Hi mads

    I checked the mosfets and turns out 2 of them died, not just one, i replaced them and tried again but something went wrong, again, but this time it worked more stably( now I added shielding). I can’t check right now but i guess at least one of the mosfets will be dead 🙂 so my final guess is that i have a bad GDT, signal comming out of it is too distorted. I don;t have good toroidal ferrite cores so i’m gonna go with the E shaped ferrites.
    GDT secondary side is also wired correctly and the resistors are fine as well as the control circuitry,come to think of it, control circuitry is truly quite safe 🙂
    I have the oscilloscope at work(don;t have my own) so at this moment i can;t check the signal but last time i check it, phase was shifted but the signal was distorted.
    Hope you finish that EMI guide soon 🙂

    Kind regards

  290. Hi George

    It sounds like you need to find a proper core material for the GDT if you are having problems with highly distorted signal and you are not sure of the ring core you have is good enough.

    Using opto-couplers and high side drivers to drive the gates is even more complex than using a GDT and it is not better in the eyes of a experimenting amateur. You need to use a isolated power supply for it, you can not just use a 7812 like in your drawing. Remember that it will be sitting at high voltage potential when connected to the gate-emitter.

    High side drivers are only used in the industry to avoid large magnetic components like the GDT, it is simply a matter of energy density in the inverters. GDTs are superior in every way in experiments if you ask me.

    Kind regards

  291. Hi Leni

    It will surely work, but I can not say if it will hinder performance from the high impedance at the resonant frequency of the coil.

    You can use 0.33 mm wire, but remember that the resonant frequency will be higher, so maybe you need to add a bigger topload to lower the overall frequency of the system.

    Kind regards

  292. Hi Mads

    ok then, i will stick to the GDT’s. it’s my first GDT so, it is kinda frustrating 🙂 i guess i need to do more calculations, more accurate ones.
    when testing the circuit, i should give it the approximate resonant frequency of the secondary or it’s not necessary to use exact same frequency as a test?
    can you show me the appropriate output signal form from the drivers( ucc37322 & ucc37321) and the output of GDT?
    i checked mine and turns out there are no square wave outputs, there are like spikes, no way near to the squarewave.
    output of the drivers also was filled with noise or some kind of distortion.
    and what is the exact role of the C4? just to filter out the distortions?

    Kind regards

  293. Hi Mads
    I am building a second Tesla Coil on your chart .First Tesla was on the half bridge circuit with the frequency of the source 166 kH CTC- labs de miniSSTC 2v2.0. circuits work very well. Now I try to start your design and after that burned 2 mosfet and ordering another GDT I had a question about the frequency of steady work full bridge.Thank you for help
    .Kind regards.

  294. Hi Mads
    I will explain, I have Feritte ring mi 2500. can be therefore had to change the minimum frequency at interrupted mode to 5000Hz to get acceptable waveform. I bought another ferrite like you and I hope that Tesla still sing
    Kind regards.

  295. Hi Alekc

    I had the full bridge running stable at both full CW mode, interrupted mode and with musical interrupter, with a resonant feedback frequency from the secondary circuit at around 300 kHz.

    5 kHz interruption signal sounds very strange, maybe you just have problems with too weak feedback, if the antenna is too far away from the secondary coil?

    Kind regards

  296. Hi Mads

    I decided to test myself again and got in after interrupted 555, 120 Hertz. I have instead of a 2 M to 500k .My question whether feedback change so much Tesla coil frequency.
    Resonant frequency from the my secondary coil at around 251 kHz.
    Thank you.
    what are you think?

  297. Hi Alekc

    120 Hz is normal interrupter frequency, that is good. Your resonant frequency of 251 kHz is also good. The feedback can only be from the secondary resonant frequency and it then drives the primary circuit at that frequency.

    Kind regards

  298. Hi Mads
    thank you for your help.
    i have another question, can i change GDT to scheme circuite electronice to control mosfets for Tesla coil.
    Kind regards

  299. Hi Alekc

    High side drivers are much more complex, requires isolated power supply and there are no real advantage over a GDT when it comes to be used in a Tesla coil.

    Kind regards

  300. Hi Leni

    You should look for a ring core with Al higher than 3000 uH, that one is simply too physically small and has a too low permeability.

    Kind regards

  301. Hi MaxEletron

    You can replace it with a old-school transistor pair, but it will not be as good as a driver optimized for correct peak current and driving current. One of the key design parameters in power electronics is to switch as fast as slow as possible, so if your transistor pair delivers a very high peak current it might give more ringing on the gate than if you used a driver IC.

    Kind regards

  302. OK, thanks. Here in Brazil you can not find these ICs even if it rains electronic components. And that’s a big problem for us electronics lovers! Can you tell me some other (gate driver component)?

  303. Hi MaxElectron

    There are numerous different MOSFET driver ICs and if you can find one with a peak current of around 6-9A and one that has a enable pin input, you can use that. I have also used the MAX4420 in other projects.

    Some times it was cheapest for me to order UCC37XXX ICs from mouser in USA and just pay the import taxes to get them to Denmark, some times it sucks to play with power electronics and paying out of your own pocket 🙂

    Kind regards

  304. Hi mads,
    I’m little bit embrased with capacitors next to the primary. They are DC blocking so if i have low esr capacitor with 500VDC and 250VAC rating it will be ok or I need buy new with higher rating? And will be 14 turns on GDT ok?Thank you 🙂

  305. Hi MaxEletron

    You can replace it with secondary base current feedback transformer, but it does not work that much better. CW operation can also risk driving the CT into saturation if you do not prepare it for CW use.

    Kind regards

  306. Hi Leni

    500 VDC should do good enough, you will at least know if it starts to smoke 🙂

    With 14 turns on the GDT, you will have higher peak current and voltage/turn, it will do fine, there will not be much difference on 14 vs 15 turns. You are still above the minimum windings needed to maintain proper operation.

    Kind regards

  307. Some schemes have pin 5 of the CI555 connected to the GROUND with a capacitor of 100nf. Is this capacitor necessary or does it depend on the circuit?

  308. Hi MaxEletron

    Connect it to ground with a capacitor, this will keep the control voltage input at a stable voltage, this voltage is what is used for trigger/threshold limits.

    Kind regards

  309. Hi,
    what is resonant frequency of secondary coil – need to compare? I wound secondary with 0.33mm wire and want toroid with inside diameter 180mm and outside is 370mm – will it be ok? or is it much. Whats output voltage of secondary? 🙂
    King regards,

  310. Hi Leni

    I think the resonant frequency of the coil itself is around 330 kHz.

    You topload sounds like a big one, which can be a advantage at times, it would properly work very well, makes it able to store a lot of energy in the topload.

    Output voltages from a Tesla coil are very tricky and depends highly on the load it discharges into. You can NOT use the general rule of 10 kV/cm. A coil of this size will properly produce around 100 to 250 kV, this is just a rough estimate. The voltage drops significantly when there is a spark flying out as all the energy is being fed into generating the spark.

    Kind regards

  311. Hi Gert

    Thank you for the kind words.

    You can use a off the shelf gate drive transformer, but it will have to match the needs and frequency of your SSTC. The 3 you linked to are all small and meant for SMPS that operate into the MHz range, you need to find some that are bigger and suitable for the frequency range of your SSTC.

    Kind regards

  312. Can you tell me why the CW mode is still active even with the ENBL pins out of the 12v? I can not put it in interrupt mode because of this.
    Even switching on interrupter mode will continue in CW mode!
    my coil actually works and make sparks! But just the ENBL inputs that does not work!

  313. Hi MaxEletron

    I know that my schematics do have a error in them, the ENBL pin should not be left floating, it should either be tied to positive or negative rail for the desired operation. If interrupted mode does not work, you are not meeting the conditions to set ENBL off on the driver ICs, check the datasheets and compare to the output of your interrupter.

    Kind regards

  314. Hi MaxEletron

    It is a Metalized Polypropylene capacitor, so in theory it should work fine as long as you does not exceed its voltage rating.

    Kind regards

  315. Hey mads.
    As I have more free time I’m finally managing to get back into electronics, and looking back at this project, I noticed on mistake I may have made back when I built it.
    What happens if you don’t connect ENBL to +12V, without either interrupted mode or audio modulation?
    Best regards.

  316. Hi Unleashed

    The operation is uncertain, for some reason my coil would operate in CW mode if it was left floating, but that behaviour have not shown to be replicable by other users. ENBL needs a +12V constant/signal input for the drivers to deliver a output.

    I hope you will sign up at our new forum at https://highvoltageforum.net and show us the progress of your Tesla coil 🙂

    Kind regards

  317. Hi Mads,
    I just wanted to thank you so much for helping spread knowledge freely, so that people can learn how to do anything they want. I built the audio modulator for my SSTC and it works beautifully… it was also the circuit I decided to try and etch on PCB for the first time, which also turned out nice.
    I can use the audio interrupter with my guitar!
    Here is a link to a video I posted on youtube:

  318. Hi Adam

    I am happy that you find the information helpful and thank you for linking back to the site.
    Cool that you return with a video of your results 🙂

    Kind regards

  319. Hey Mads,

    I am planning to build your SSTC over this 3 weeks and I have very minimal knowledge and understanding on electronics.

    There are a few questions I would like to ask:
    1.Which are the decoupling capacitors? C5,6,10,12,16 and 17? or is there more?
    2.Is a 200w variac sufficient or should I get a 500w it’s pretty costly or perhaps a light dimmer controller like this: goo.gl/JjCaIn might work?
    3.The resistors used are 1/4w rated right?
    4. Where should I place fuses on the circuit to avoid parts from being damaged as parts are really hard to source here :/
    5. How am I supposed to build the rectifier circuit on the mains voltage (variac), this is the part where I am completely lost. Is there any parts/ diodes I can use or how have you done it?

    My mains provides 240v. I have very minimal trial and error gap which I understand that it is a very vital part of learning but I will be taking my (I)GCSEs this year so I really do not want to spend too much time on this 🙁 It will be great if you could reply by this week as I will hunt for the parts locally this weekend ( sourcing them locally online is very expensive)

    Best regards,

  320. Hi CCX

    C3, C5 and C6 are the only decoupling capacitors, the rest are associated with timing or band pass filtering.

    0,25W resistors is fine

    I have always used a vaiac, but for a SSTC I have seen others use light dimmers for them, but never use them more than up to 50% of their power rating, due to bad power factor in a SSTC.

    Bridge rectifier or 4 diodes in a full bridge configuration, google it 🙂

    Kind regards

  321. Hey Friend,

    my UCC21/22 not work ENBL pin. No possible interrupter mode. I can replace by this alternative?


  322. Hi Max

    No, this will not work. The feedback from the antenna is the resonant frequency of the secondary coil circuit and is needed in order to switch the primary circuit at the right frequency. The interrupter or audio interrupter uses the ENBL pin of the MOSFET driver to turn the drive on/off to the frequency of the audio, so it is two completely different signals that you are trying to mix.

    You could try to insert the transistor at the positive power supply to the MOSFET driver ICs and simply cut the power to those with the audio interrupter signal, I have not tried this before, so it might not run stable.

    Kind regards

  323. Olá Mads,

    qual a função do CI 555 no circuito principal? Sera a mesma função no modulador? Ou ele da suporte ao 74HC14?

  324. Olá
    Existem dois 555 ICs. U5 faz parte do interruptor regular e U6 é parte do interruptor de áudio. Eles não estão relacionados ao 74HC14. Isso é feito com o google tradutor. Atenciosamente Mads

  325. Qual seria a função do u5 no circuito mais precisamente. O u6 voce explicou la em cima.
    Obrigado pela ajuda.

  326. Look at this circuit attached. If you can help me, I would like to know what the function of the CI555 would be. I think it’s an adaptation of what you’ve set up.

  327. Hi vinicius

    The function of U5 in my schematic is the same as the 555 in the schematic you posted. However, as you say, the schematic you posted is a adaption where the interrupter is mixed with the antenna feedback. Here the 555 pulls up the signal from the antenna to be strong enough to drive above the 74HC14 threshold voltage.

    Kind regards

  328. I believe that the 555’s function in the circuit I posted has the same function as your circuit. Thank you for helping me. I used the Tesla coil as the subject of my college graduation. I wanted a more detailed explanation to improve my knowledge. And you helped me a lot. Thank you again.

  329. The MOSFETs switches on and off in sequence, to let power from the DC bus capacitors enter the primary LC circuit where the current will then ring up according to the ESR/impedance of the primary circuit.

  330. Daniyal Hussain

    I’m having a bit problem with MOSFET Gate Isolation Transformer (Toroid) of my DRSSTC; for which i used 24 awg wire with 16 turns on each and a ferrite core. But i haven’t succeed in getting pure square waveform on the other side of Toroidal Transformer ergo does anyone have a better suggestion and could help me over this particular issue!

  331. Hi Daniyal Hussain

    What is the physical dimensions and ferrite material type of the ferrite core?

    If you re-used a yellow/green ferrite toroid core from a ATX power supply, those are not suitable for the job of a GDT.

    Kind regards

  332. Lack of current or RF interference?
    When I turn on the circuit, the LEDs are normal! (flashing interrupter mode).
    When I turn on bridge H and the sparks begin to jump, the LEDs stop blinking and enter static mode. (they stop blinking and stay on). The interrupter mode does not work, only the cw mode.

    Is there any way to test the ENBL pins (ucc37321 / 22) to see if they are working?

  333. hello Mads, how are you? i hope u’re fine

    i’ve bought and searched all the components and studied the schematics, but there is one point that
    i still don’t understand: the feedback.

    could you explain how exactly the antenna make mosfets switch at the right time to make the secondary run at the resonant frequency?
    or send me videos or tutorials about this point?

    i’m asking because i’ll have to explain it to my class, and i searched a lot and haven’t found any good explanation,

    thanks a lot, your tutorials are very good!

  334. Hello!
    I have two questions. What is the required amount of turns on the GDT? (I don’t know if i should look for a bigger ferrite core. It is a little bit too small for 5x 15 turns).
    How should i ground the secondary? (Mains ground doesn’t sound good 😛 )
    Thanks for help and awesome documentation!
    Keep up the good work!

  335. Hi Tymoteusz_K,
    You can use the equation V*s= t*T*(m^2) in that:
    – V: Voltage applied to primary winding in volt, for this application use 15V.
    – s: The time in seconds in which the voltage was applied, use 1/500000 .
    – t: Primary winding turns
    – T: The flux density of your ferrite you want your transformer to run at (in tesla), for safety I would use 0.01T for this application
    – m^2: The cross-sectional area of the ferrite core, you must calculate this yourself using basic geometry. Note that this area must be measured in square meters.
    After you get the turns numbers you had to choose the wire cross-section area, or AWG number. If you can’t wind all the turns then use a smaller wire.
    For high voltage stuff I always use a separate grounding rod, if it was tesla coils then in additional to the grounding rod, you should also connect your garage roof to the tesla coil too. The additional safety worth the cost of building one.

  336. Oi mads tudo bem ? Gostaria de saber se teria problema se eu usar um cilindro de 100mm para a secundária é 110 para a primária?

  337. I was testing the musical interrupter with an led on the output, and the resistor R14 gets VERY HOT, I touched it and burnt my hand, I measures it’s temperature and it was above 100 degree celsius. So is it supposed to happen or is something wrong with mine ? Please help

  338. Will it be fine if my GDT has 12 turns ?? And can I power my Tesla coil without a variac and to prevent the high inrush current can I use a NTC ? And will polyester film capacitors be fine for C8 and C9 ??? Thanks for the help and merry Christmas

  339. Hi Saattvik

    You can try to use a larger value for R14, maybe in the range of 500-1000 Ohm and see if that fixes the heating problem, just beware that you do not make the signal to the 555 too weak.

    If I remember correctly, 9 turns is the minimum for a standard GDT like used here in a SSTC. So 12 should work but if you have problems, try with 15 turns.

    A NTC will help against in-rush current to the filter DC bus capacitor, but not against over-voltage in the primary circuit because of too high coupling. So there is still a risk of racing sparks and flash-over between primary and secondary, a NTC will not fix that.

    Polyester/MKT capacitors will work just fine for the voltage doubler/half-bridge capacitors C8/C9.

    Kind regards and Merry Christmas

  340. I am making the interrupter part of the circuit separate to control it from a distance, so is it necessary to use a earthed metal box for it to prevent from RF ? And do I have to use a coaxial cable for the connection between the interrupter and the Tesla coil ?

  341. I am building it in half bridge configuration, so is a 330uf capacitor enough for the dc bus capacitor ?? And btw they are those cheap capacitors from SMPS’s so are they fine ?
    And is it ok to use 10 ohm resistors for the mosfet gates ?

  342. Should I place the NTC before or after the bridge rectifier ? And will a 5 ohm NTC be enough ?

  343. And my secondary coil has around 835 turns and is 64mm in diameter so should I use 10 turns for primary or some different amount ? Sorry for bothering you and thanks for helping 🙂

  344. How much current will it pull from the wall at 240v AC (Half Bridge and running in interrupted mode) And I noticed that you are powering it with a lower voltage in musical mode, Why is that ? And if there’s a reason to it, I wouldn’t be able to do it because I don’t have a variac.

  345. Hi Saattvik

    You need as much shielding as possible. You could also upgrade the interrupter/driver board with optical input as you see in DRSSTCs.

    With 330 uF you will most likely see some ripple at full power, you can calculate this as I did here: https://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/dc-bus-capacitor/

    It depends on the gate waveform, if its not rising fast enough with a single over-shoot, try a lower value.

    You can place a NTC both before and after the bridge rectifier, doesn’t matter if its on the phase, neutral, positive or negative line.

    How many turns on the primary you need, depends on the current that will flow, if your MOSFETs can handle it etc. You can calculate the primary circuit current and then from your interrupter get an idea of the average dutycycle and calculate the RMS current: https://highvoltageforum.net/index.php?topic=455.msg2763#msg2763

    You should varnish the secondary for two reasons. To secure windings from falling off, a thin wire secondary coil is very sensitive to temperature changes, so you risk having turns getting so loose that they fall down and it becomes a mess. Varnish also improves the insulation to prevent racing sparks and weird breakouts on the middle of the coil. The more varnish you have on a secondary coil, the higher coupling and average power can be pushed through it without flash-overs.

    I used lower voltage because too long sparks generated so much EMI that it got my keyboard to hang up and it would need to be turned on/off to work again.

    You should post your work on https://highvoltageforum.net 🙂

    Kind regards

  346. hello Mads, i am planning on making a tesla and i choose this design, do you have a simple pcb layout without audio ? i am kinda new to this and i don’t want the circuit to be on a perfboard, thank you in advance.

  347. Thank you, i made the PCB and i ordered the components i was missing. Now i have a few problems
    -i measured the 555 timer output, and its range is from 3Hz up to 31Hz, is this the normal value ?
    -how long should the antenna be? does it matter ?

  348. Hi Bob

    I forgot to mention that the SSTC3 is made for very low BPS, as it uses its own enclosure as heat sink. Change R5 and C11 to lower values for a higher BPS.

    The antenna should be able to stand half the way of the side of the secondary coil, maybe some 10-20 cm from it, lower the height if corona/sparks are forming on its tip.

    Kind regards

  349. well, since it’s my first build, i will start at minimum with maybe 12v on the bridge, and after i get everything working, i will step up the voltage and the bps as well.
    Thank you for answers Mads, have a nice day.

  350. I built it, it’s kinda working. When I connect the interrupter to enable pin of the ucc’s it still stays in continuous mode, even when I leave it unconnected to anything. Could it be because of the rf of the Tesla coil because I haven’t put it in a enclosure, it’s all open. And when I connected an oscilloscope to the enable pin it was showing a sine wave of 135khz so it must be from the rf. If there’s any other reason please tell. And will shielding it solve the problem or I must do something else too. Thanks for the help 🙂

  351. Also the tesla coil is oscillating at a frequency of around 130khz but I estimated its resonance frequency to be around 400khz by using some tesla coil calculators. So what seems wrong here? My gdt right now is not good and the driver has some interference from the rf, so could that effect the frequency at which it is oscillating?

  352. I don’t have a variac so when i will power it from rectified mains there will be a large inrush current which could blow a fuse and damage the capacitors. To prevent this could I connect a resistor in series then charge the capacitors then bypass the resistor with a mosfet or solid state relay and after this power the driver circuit. So will this work properly? And if it has some disadvantages to it what are they? Sorry for making so many comments, thanks for help

  353. Earlier I was using a core with a low permeability for the gdt, it was working but not properly. So I bought a new core with permeability of around 9000, but now its not working at all. I have kept a cfl near it, when I turn the tesla coil on, it just flickers a little very faintly. So what could be the problem.

  354. Hi Saattvik

    Check the phasing of the GDT, it could sound like you are driving it out of phase with the feedback from the antenna.

    Kind regards

  355. I fixed it, now it’s working. But it’s only working in continuing mode. Even when I have the interupter connected it still stays in continuous mode. The Tesla coil drivers are not shielded so could that be the cause of the problem? Or something else. Thanks for your help

  356. Hi Saattvik

    Try to shield the driver, SSTC’s does have a powerful electromagnetic field. So inbetween the interrupter pulses it could be driving itself from the induced current.

    Kind regards

  357. I tested the ICs separately by connecting them to a gdt and providing the test signal from a function generator, but still it’s not getting interrupted by the interrupter. So could the problem be the ICs themselves? I bought them from AliExpress

  358. The problem was with the ICs, their enable pins don’t work. So I ordered UCC27425 and will use that and hopefully it will work. Loneoceans had used it too. If there are any other changes in the circuit that are required for it please let me know. Thanks 🙂

  359. James (BD594)

    Would I be able to subsitute C8 and C9 with a mylar capacitor? Also is there a reason why you are using two caps instead of one?

  360. Hi James

    You have to used either MKP or MKT capacitors. Mylar’s high AC dielectric losses may cause excessive heating and losses in radio frequency applications or high duty-cycle pulse applications.

    I used two in parallel to get a lower capacitance value and share the RMS current.

    Kind regards

  361. Hi Andreas

    It is not as fast, but still specified for the upper reverse recovery time of the MUR1560, so yes you can use VS-EPH6007L-N3 instead.

    Kind regards

  362. My friend gave me a Tesla Coil kit from Rainy Lake Electronics -zipcode: 56435 but I do not know their email address and when I called, nobody answered the phone. I can’t find where the instructions are on the web.
    1) Do you know where can get instructions and plans for my DIY Tesla Coil?
    2) Where would I find more people who would help me build it?
    3) I assume you need pictures and the list of parts – I can provide that.
    Thank you very much.

  363. Milos Milicevic

    I am 14 years old.I built some spark gap tesla coils before.I started a new pll sstc with Irfp460 in full bridge. I have 2 questions:
    1)Can i use 440v(220v ac with woltage dubler.I know it wont be 440v because 220v is avrage voltage)?
    2)Is it better to run it interupted?

    I can send some pictures.

  364. Hi Milos Milicevic

    IRFP460 is only rated for 500VDC, a 230VAC voltage doubler would give you 455VDC, that is too close to the limit of your MOSFET. Switching transients should be given a head room of about 33% of the DC bus voltage. So you would need a 600VDC rated MOSFET to use a voltage doubler.

    Interrupted is ALWAYS better, to get a higher expected life time of the various components in a Tesla coil. Especially that MOSFETs needs time between pulses to cool down again. In the topology like a DRSSTC, interrupted is a must, it would simply explode from temperature rise of the silicon die until it melted and short-circuited.

    Kind regards

  365. Prabhat Kumar

    hey !! i tried making the sstc 3 from your site( I know i should have posted there ). I made the pcb from the express pcb file you provided by conventional toner transfer method at home. The pcb is working and I verified with a scope without the bridge supply on ( Only 12 volts was fed for the oscillatory part ( with drivers ics, gdt and the mosfet in place), only thing i didn’t connect the primary wires to the tesla coil . I checked it by running a slayer exciter on the same tesla coil to be used with the half bridge later with antenna in close vicinity ( 3-4 cm from the coil ). The wave forms were as expected just with a slight damping which could be fixed by reducing the gate limiting resistor( i used 5.6 ohm originally ) . But as soon as i connect the supply to the bridge( just 12 volts to check) , bang !! my driver circuity went in the current limiting mode and immediately the top ucc became hotter than normal. ON checking everything i realized this is not the only victim of this crash but also my 74hc14 inverter went bad i.e I dont have any pulses at the base of the driver ics. Now believe me i tried to build a full bridge with ir2110 bef0re and burnt like 10 ir2110 before with the same issue, as soon as i connect the supply i goes bang !!( at least the ir2110 worked till the supply voltage was raised till 25 volts dc).
    So please help me out as i dont have the money to invest on lot of ucc ics they are very expensive and also not available where i live ( I have to order them from mouser).
    PS: Is it because the the primary was not connected anywhere or somehting else which i am not aware of .

  366. Hi Prabhat Kumar

    This circuit is designed to be used with a GDT, since you do not mention a GDT and you have components failing in the logic part of the driver, I assume you are using the IR2110 as a high-side driver and that is where you get your short-circuit path from.

    Kind regards

  367. Irakli khubejishvili

    Hello. I maked driver scheme but with half bridge.
    And when im plugging on 40volts its working but without modulating.
    On 80 volts its working but once mi gate transformer gapped and killed tranzistors. On110 volts it worked 2mins and gapped gate transformer again.
    And can you help me with solution.
    I burned over 16 transistors😞😞😞

  368. Daniel Kowalski

    Hi Mads,
    I’ve built your SSTC (well not quite yours because I have different Coils with 6 turns of 4mm2 wire on primary and about 2500 turns with 0.15mm on secondary and the resonant freq is about 133kHz). My issue is that I get big pulses of high voltage (about 200V peak at 30VDC bus) across the freewheeling diodes and everytime when I go with highest BPS and freq of generator the 2 diodes are burning at about 180VDC bus voltage. My GDT is not perfect because the gates have amplitude of 15 – 7V falling square but the shape is not bad (minimal ringing at 30V) but I will probably rewind this on proper core material (Question: 3E25 or 3E27 material?). In bridge I’m using IGBT’s: IRGP50B60PD1 with MUR1560 diodes.

    I would be grateful if you would help with this problem.

  369. Hi Daniel

    At low voltages on the DC bus you will have proportional higher switching transients due to the output capacitance of the switch is lower with higher C-E voltage.

    How is your wiring done? Maybe sign up at https://highvoltageforum.net and make a thread with many more pictures of the whole setup. More people to help there as well.

    First fix your GDT with a core material that is around 5000 AL, it could be shoot-through or linear operation giving you problems due to bad gate drive.

    Kind regards

  370. Hey Mads, do you think it would be possible/convenient to use an opto-coupler for the audio input?


  371. Hi Unleashed

    Optic isolation is great for improved human safety and noise immunity, it is however mostly left out in SSTCs due to the cost of the parts is close to the rest of the coil. You do however always see optic interrupter in DRSSTCs where the sparks also tend to be longer.

    You just need to drive a optic transmitter with the 555 output, if its powerful enough for that and have a receiver circuit at the driver. Look into Steve Wards Universal Driver 1.3b for inspiration to this.

    Kind regards

  372. Hi Mads
    I am building my first tesla coil and i have a few questions, maybe you can answer them. Because i don’t have a variac I am planning to use a 36v transformer first, then switch it off and switch on 220v mains to help with the inrush current, would that work? Also, I have never made a GDT before, but I plan on using a Cat-5 cable like you previously described, I’m wondering what core should I use, because i dont want to fry the mosfets. Lastly, how tall and how far away did you put the anntena from the secondary coil, if I put it in a bad place am I at risk of damaging the other componens?
    Thanks in advance.

  373. Hi Vilius

    Instead of going through the hassle with a low voltage transformer and switch around, you could just use a manually controlled contactor or with a timer relay on it, so switch on a mains contactor, that has a resistor across it. So when the contactor is OFF the mains goes through the power resistor and limits the inrush current, once you turn ON the contactor, the resistor is “short-circuited” and thereby by-passed.

    Buy a ferrite ring core about minimum 40 mm diameter and a AL value of around 5000.

    Start with antenna placed in level with bottom of secondary and maybe some 10-20 centimetres from it, move it closer for stronger feedback and if you see corona/sparks forming on the antenna, move it further away.

    Kind regards

  374. hay hi, I know this is a silly question but I am really confused.
    does 2k2 resistor means 2.2 ohms or 22ohms or 2.2k ohms.

  375. Hi Aditya Agarwal

    2K2 means 2.2 kOhm
    2R2 means 2.2 Ohm
    22R means 22 Ohm

    The unit is used as comma separator to save space in drawings

    Kind regards

  376. Hey Mads! Thank you so much for the very informational website! I just made a half bridge IGBT but the supply is 120v coupled with a voltage doubler so every transistor switches 170v. I added more topload metal to lower the resonant frequency thus increasing power output and and spark length. It worked but, at some point the interrupter would stop for about 1 second and then burst into CW mode for 1 second and then go back to normal. I thought this was okay (together with the 1 second long, hot sparks) but then when it continued like this, there came a time when the interrupter stopped for one second and when it bursted into CW mode, the IGBTs exploded. I have put the driver module inside an aluminum casing so I think EMI wouldn’t be an issue but I think EMI is coming through the gate connections because the IGBTs are outside the aluminum casing. Could you please share your thoughts on this?

  377. Hi Jhornel

    A antenna feedback SSTC is tad bit unstable when it comes to higher power outputs. You should move into DRSSTC’s to gain some more control over your feedback networks and it is also more suitable for the power levels.

    Kind regards

  378. Thank you for your quick response Mads! I haven’t mentioned that instead of an antenna, I used a torroid ferrite core with a one turn onto the primary connected from the secondary earth connection to earth. And then the secondary of the feedback transformer has some 30 turns. It is like a current feedback. And I then feed that into the waveform cleaning. But I guess I should still switch to DRSSTC for more power? Please let me know your thoughts.

  379. Hi Jhornel

    I think your feedback is too powerful or maybe even phase shifting your feedback too much compared to the original signal.

    Do you use a adjustable start-up oscillator to get it running before the CT can take over?

    DRSSTC is the way to go for better stability, more control and less blown silicon 🙂

    Kind regards

  380. Hi Mads

    I see. Thank you. I will try to change the current feedback transformer with less turns on the secondary and see what happens.

    I am not using an adjustable startup oscillator and I have no idea what it is and what it is for. Should I use one?

    I think I would need to save for the parts of a DRSSTC.

    By the way, do you also save money to buy parts for your coils? I am just curious how do coilers get the funds to buy very expensive parts.

    Please let me know your thoughts.

    All the best,

  381. Hi Jhornel

    You properly need more turns and a more suitable shunt resistor. Hundreds of secondary turns and a small resistor to avoid phase shift.

    Maybe you just have problems with noise induced from the secondary ground…

    I scavenge old industrial electronics for free parts and buy used parts on ebay. I have only ever bought the 60N60 minibricks from new, all other bricks are used from ebay.

    You can build a small DRSSTC pretty cheap with TO-247 IGBTs using prof9dc PCB packs that are available for free, just order the PCB with the gerber files from your prefered pcb manufacturer. : https://highvoltageforum.net/index.php?topic=353.msg2401#msg2401

    Kind regards

  382. Hi Mads,

    Thank you so much for all the valuable information you are sharing. It is really priceless.

    I think I know now what the problem is. My control circuitry negative side is not connected to earth! And because of this, I think that it picks up the small signals although it is inside a grounded metal casing. One question is, can the negative side of the control circuitry and negative side of the power circuitry be connected together and connected to earth? (Separate earth ground from secondary output) Or can I connected them all in the same earth ground?

    This is my last question.
    Thank you so much Mads.

    All the best,

  383. Hi Mads, i have some questions:
    The primary capacitor is two capacitors of .68 µF in parallel (1.36 µF)?
    Haw much is de peak current in the primari coil?
    Haw much time is the primary burst oscilation?


  384. Hi Mads , I have a quick question.
    Is there a time limit for the coil to be working ?
    Because mine is working fine until around 30 seconds and my transistors blow up .

    I have a circuit similar to yours

  385. Hi Mochi

    There is not a set time limit, but it sounds like you have problems with too high heat dissipation in your MOSFETs. You need to turn down the BPS, on-time or input voltage. Ensure there is good cooling of the MOSFETs.

    It could also indicate that you have problems with cross-conduction or very high losses due to a bad design or physical layout. Please sign up at https://highvoltageforum.net and make a thread with more pictures and its easier to help troubleshoot these things on the forum 🙂

    Kind regards

  386. Hello,

    Where can I buy a Tesla coil that works in CW mode? like this:

    Continues Wave mode

    At 200 VAC input voltage, 280 VDC on the bridge and a power consumption around 10 A, peaking at 20 A, the coil was drawing somewhere in between 2000 to 4000 Watt. This resulted in very hot, thick white arcs punishing the dead iPod shuffle which remarkably left the player relatively unharmed considered what had just taken place.

    These flame like sparks are 250 mm in length.


  387. Hi Mads,
    I’m planning on building a SSTC based on this schematic, but there are a few things bothering me:
    1. Why didn’t you add snubber capacitors into the H-bridge ?
    2. I also don’t see tvs or zener diodes protecting MOSFET gates. Weren’t you afraid of voltage spikes damaging MOSFET’s ?
    3. I’ve seen a lot of different schematics and in many of them there were also diodes added parallel with gate resistor to discharge the gate faster. It kinda makes sense to use one, is there any reason why didn’t you use it ?
    4. In order to use this circuit to drive SSTC in CW the only thing I should do is to hook up ENBL legs of UCCs to +12VDC right ?
    5. Why did you use +9V DC in audio modulation circuit instead of +12V ? LM741 can easily handle +12VDC

    Best regards

  388. Hi Rodney

    All good questions that show you have done your home work and the answer to 1, 2, 3 and 5 is: because I built this 12 years ago where I followed others work, primarily Steve Ward and CTC labs. So at the time I did not question much, I was learning as I built it 🙂

    To 4, yes just keep the drivers ENBL to deactivate interruption of the feedback signal.

    I hope you will sign up at https://highvoltageforum.net and make a thread about your project!

    Kind regards

  389. hi Mads!
    I’m designing an sstc, I have 11x28cm secondary, with 0.315mm wire, and with proper topload its frequency is 230khz. I’m using the same feedback on 74hc14 inverter, and some ir2184 chip in bootstrap configuration instead of ucc chips and gdt due to the fact that I can’t get them locally. i want to make half bridge, and the questions are: can i swap the antenna with current transformer, can i replace irfp460 with some igbts, such as 50b60pd1, and what resistor values will i need, gate ones, I don’t know because i haven’t a scope. the driver works good with irf740 transistors and with some flyback transformer as a load.
    kind regards, Eli

  390. thanks for replying!! spoiler: 1:30 ct doesn’t work as intended, the oscillation doesn’t start on its own. though with antenna I’m getting some weaker results from 50v dc. from half wave rectified mains, through 250 watt transformer primary, I’m getting 15cm spark. voltage is 240v ac. there is small corona on the tip of the antenna which scares me a bit. is 15cm spark is good for 250w though? i want to make is as efficient as possible. (capacitor divider value is 11x11uf).i could make it full bridge too. your design seems to be quite efficient, because from what can i see from 30v minimum input, it already gives pretty amazing breakout.

  391. Yes, CT might need a manual tuning or start-up oscillator, as its harder to get a powerful enough feedback signal, compared to an very sensitive antenna.

    15 cm sparks from 250W is a good result from a SSTC. SSTCs are however not known for long sparks, best way to get that is with larger toploads, more capacitance to discharge from.

    SSTC sparks are nice and quiet compared to DRSSTC, that is so noisy you need ear defenders 🙂

  392. Hi!
    I have build an SSTC based of this project, and I have just started testing (so far with 30V DC as main Vsupply). There appears to be some high voltage (it is capable of turning on small neon bulbs and small CCFL from a few cm, and there are some tiny sparks), but I have a problem. Two of the MOSFETs in the H-bridgre are heating up quite a bit (the two “left” ones looking at the schematic, the two “right side” ones are totally cool). I am suspecting a bit of shoot-through, but how can that be the case in only one side of the h-bridge when all FETs are all driven from the same GDT? I’ll add that I have 5R resistors in parallel with a UF4007 diode (anode to gate) on all gates as well as 18V TVS diodes. Sadly I only have access to a multimeter and that DSOsomething toy oscilloscope, so my measuring capabilities are quite limited. Do any of you guys know what can be causing my problem?
    Best regards,

  393. Hi Maks

    Be sure to check the polarity of the GDT to be correct, you need to drive one MOSFET from each side of the load (primary coil), it sounds like you are turning on both left side at the same time, then you are just short-circuiting your DC bus through those two MOSFETs.

    You could sign up at https://highvoltageforum.net and post some pictures of your coil and video of it running, more users to help you there 🙂

    Kind regards

  394. Hi Mads,

    I’ve (almost) successfully built my sstc, but there are few things concerning me. When I run in in CW at~150 VAC on the input there was nice plasma forming on the tip of my resonator, but the D-S mosfet waveforms were really nasty – there was huge overshoots almost twice the supply voltage. I’ve reduced the tracks length as much as I can to reduce parasitic inductance but it didn’t seem to help much. I added snubber capacitors, even tried ferrite beads on power components but with no luck. I also made sure that there is no cross-conduction in the full bridge. And there’s also one thing the mosfet waveforms seem to be much worse on high-side mosfets and I don’t know why. Do you have any tips for me ? Because I’m running out of ideas :/

    Despite the overshoots coil worked nicely for a few minutes (full bridge was barely warm) I could not measure the current because all of my meters were going crazy even 2 meters away from the coil. The fun part happened when I tried interrupt mode. I could hear signal being interrupted but didn’t see any sparks. At around 50 VAC on the input one of the mosfets just exploded with no reason. ( I know that without any pictures / waveforms it is hard to help, but I was so focused on operating the sstc that I completely forgot to capture waveforms). Thank you in advance for help.
    Best regards

  395. Hi Mads.
    What is reason for bridge fault, when U run it at full input voltage ?
    “During a run of CW at full input voltage, the full bridge blew apart completely, with a loud bang.”
    Overvoltage spike, or just overcurrent load for IRFP460 ?

  396. Hi PowerTech

    I suspect it was something with the new topload, it was smoother, but also a bit smaller. The resonant frequency did maybe go up too much and the higher losses caused the MOSFETs to exceed their maximum die temperature.

    Kind regards

Leave a Reply

Your email address will not be published. Required fields are marked *

The maximum upload file size: 10 MB. You can upload: image, document, spreadsheet, text, archive, other. Drop file here