Easy Open Tesla coil on PCB

Very simple Tesla Coil - 3 components only = LED, BUL45D2G transistor and 82K resistor

Similar projects worth following
Starting from
bobricius has 2094 orders / 92reviews
Ships from Slovakia
Inspired with
It is nice project but very complicated.
My coil effectivity is far away from original but is very simple.

Hackaday article about this project: Heres-a-tesla-coil-you-can-wear

HABR article Трансформатор Теслы с печатными катушками, впаял три компонента

  • Input power 10V - 35V from 5V USB powered step up converter or notebook 19V power supply.
  • Frequency 2Mhz - 4Mhz
  • No special components, just LED, BUL45 transistor and 82K resistor

Development status:

  • v1 7/7mill not working
  • v2 on video working, 100turns, no visible sparks
  • v3 on pictures 160turns, works much better but no sparks, on sale some boards, used in last video
  • v4  150x150mm PCB, 240 turns, is now manufactured in PCBWay factoryboards cost 180$, PCBway cover all cost and give my this boards for FREE. THANKS
  • video V3b version 160 turns
  • video V3. version 160 turns
  • Video V2. version 100 turns


Hacked scrip to make bigger coils .... etc 350 turns :)

ulp - 23.17 kB - 04/13/2019 at 07:31


  • 1 × BUL45D2G Discrete Semiconductors / Power Transistors and MOSFETs
  • 1 × LED Fiber Optics / Emitters
  • 1 × 82K

View all 12 project logs

Enjoy this project?



Mechmallow wrote 07/18/2020 at 11:08 point

the transistor is PNP or NPN ?

  Are you sure? yes | no

bobricius wrote 07/19/2020 at 19:36 point


  Are you sure? yes | no

Mechmallow wrote 07/23/2020 at 03:03 point

Thx, another question, to get longer spark, which transistor should I chose on V4 board?

  Are you sure? yes | no

Albert Davis wrote 06/02/2019 at 04:52 point

when i place order does it have the picture of tesla on it and is it v3?

  Are you sure? yes | no

ClearloveG wrote 04/27/2019 at 08:11 point


  Are you sure? yes | no

K.C. Lee wrote 04/18/2019 at 22:49 point

You are supposed to have a "discharge" electrode connected to the other end - ideally with a pointy to produce a high electric field.  Also should hook up the grounding.

My 600kHz coil starts getting purple streamer at around 10V. My coil:

I ran mine to 60V (~200+W) as that's the limit of my bench supply and the MOSFET,

  Are you sure? yes | no

bobricius wrote 04/19/2019 at 18:59 point

do you think new pcb 150x150mm with 240 turns can help? I try next week another transistors FZT851TA have very limited life.

  Are you sure? yes | no

K.C. Lee wrote 04/19/2019 at 21:09 point

The new coil would have more inductance, larger enclosed area for the magnetic field, likely more stored energy and lower resonant frequency. This should work in your favor.

Do your damaged transistors all have shorted/opened Base-Emitter? Finding what type of damages might clue you into the issue.

  Are you sure? yes | no

bobricius wrote 04/19/2019 at 21:53 point

what is the effect of PCB thickness? now I use 1mm, can I go to the more standard 1.6mm? the distance between the primary coil and the secondary coil increases

  Are you sure? yes | no

K.C. Lee wrote 04/19/2019 at 23:08 point

The primary and secondary are supposed to be very loosely coupled, so I won't expect to see much of a difference there.  

BTW: you might be interested in the following sites (with Tesla Coil sub-forums)

  Are you sure? yes | no

Sophi Kravitz wrote 04/09/2019 at 12:56 point

Good comments @K.C. Lee I never looked for input capacitance on datasheet either

  Are you sure? yes | no

Anton Bolabolchenko wrote 04/09/2019 at 12:03 point


And how about using several layers of pcbs? (not multi layer pcb, but stack of 2-3 pcbs)

  Are you sure? yes | no

K.C. Lee wrote 04/09/2019 at 03:21 point

Gut feel tells me that your BD243 might be too slow for your '2MHz - 4MHz'

First clue:  they don't specifically say RF Power amplifier on the datasheet nor mention input capacitances.  It is listed as 'NPN SILICON POWER TRANSISTORS'

ton Turn-on time 0.3 µs (typ)
toff Turn-off time 1 µs (typ) 

i.e. it would switch at best at 1.3us period which is 770kHz.

  Are you sure? yes | no

bobricius wrote 04/09/2019 at 05:56 point

Thanks for advice, I measure this frequency with multimeter, It is close to original project.

Which transistor you suggesting ? My schematic is same like on Chinese kits. Or is better use MOSFET+ driver ?

  Are you sure? yes | no

K.C. Lee wrote 04/09/2019 at 11:48 point

It is not easy to switch MOSFET at MHz range because MOSFET have huge input capacitance.  Even a small device can have on order of 1nF. 
Homework: figure out the AC impedance of that at MHz range, what currents you'll need to drive that.  Use lTSpice

Rather than fighting the capacitance, people that make high frequency Tesla coils (HF-SSTC) use Class E drivers.

  Are you sure? yes | no

K.C. Lee wrote 04/09/2019 at 12:08 point

You are playing with high frequency stuff, always consider the AC parameters and parasitics.  :P 

As for BJT selections, use the parameter table at digikey or browse the BJT manufacturers device trees usually sorted by end applications.

 Higher current parts have lower hfe and are likely more difficult to drive.
If the datasheet don't talk about input capacitance, Gain bandwidth, go else where.

  Are you sure? yes | no

bobricius wrote 04/10/2019 at 20:10 point

Thanks for explanation, I try another transistor.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates