How can I improve my DIY high voltage power supply to produce 12,000V or more?

I followed many of the great suggestions in this POST, and made a DIY high voltage power supply.  Unfortunately, it's several thousand volts low for what I need it for (currently producing perhaps 8,000V to 10,000V I would guess).  My project, however, requires 12,000V to 15,000V.  The following points describe the current setup (keep in mind I will only be using this for a few seconds at a time):

1. 120 Volts AC input (through a common lamp cord) ...connects to:
2. 600 Watt dimmer switch ...connects to:
3. an automotive ignition coil (brand new) ...connects to:
4. a capacitor with rating of 0.75uF and 2100V ...connects to:
5. a simple 16 gauge wire, completing the circuit back to the lamp wire (not a hard connection...I just move the two wires close to each other just long enough to get a spark).
LVL 15
David L. HansenProgrammer AnalystAsked:
Who is Participating?

Improve company productivity with a Business Account.Sign Up

d-glitchConnect With a Mentor Commented:
This has taken a lot longer than I thought it would (the story of my career).
Mostly because I haven't been able to get into the lab for the past couple of weeks.

I hope to finish the power supply tomorrow and launch the lifter on the weekend.

But here's what I have so far in case I get run over by a bus before I finish:
You know why the typical ignition source for dynamite, C4, etc. is called a "blasting cap," right?
David L. HansenProgrammer AnalystAuthor Commented:
"Capacitor," I'm guessing (some are initiated using high voltage)...or are you just being cute?  Yes, I know this prototype isn't the safest design.  I do understand the dangers here (and I've taken several steps to prevent injury....insulating oil, grounding stick, and I'm using the one-hand in pocket rule).  The area is clean and dry and the components are kept about a foot away from each other during testing.  I'm even conducting the experiment (less than 2 seconds each) using a PVC rod to manipulate the arching wire.

Once I'm getting the voltage I need, I'll place the components in a case (and replace the lamp-wire with a thinker (lower-gauge) input main.
NEW Internet Security Report Now Available!

WatchGuard’s Threat Lab is a group of dedicated threat researchers committed to helping you stay ahead of the bad guys by providing in-depth analysis of the top security threats to your network.  Check out this quarters report on the threats that shook the industry in Q4 2017.

Sorry for the delay.  Several snowstorms, a funeral, and some vacation since I checked in last.
I am actually very interested in this circuit.  What is the part number for the ignition coil you are using?

Reiterating from the last post:

  1.  Automobile ignition coils are designed to run with 12 VDC, not 120 VAC.

  2.  Auto engines have between 4, 6, or 8 cylinders, and typically run between
      300 and 600 rpm.  A typical operating frequency for the coil in automobile
      would be between 20 and 80 sparks per second.  My calculation in the last
      post was incorrect.

          ( 300 rev per min) * (4 sparks per rev) / (60 sec/min)  =  20 Hz
          ( 600 rev per min) * (8 sparks per rev) / (60 sec/min)  =  80 Hz

  3.  At 80 Hz, you get 12.5 ms between pulses.  So you might what the pulse
      width to be on the order of 2 ms.

  4.  I think the best approach would be to make a flyback converter with two
      555 timers and a MOSFET transistor.  These parts cost less than $10.
      Then you can run it from a 12V battery.

There is an online calculator for the 555 timer chip.  I have attached some screen shots for the design.

You can make an astable that runs from 12 to 96 Hz with this circuit with a 50K potentiometer/variable resistor.  But you could also substitute resistor values in this range.

The output of the astable goes to the trigger of the pulse generator.  The output of pulse generator goes to the gate of the MOSFET.

Here is a similar circuit from the web.  There are lots of others too.

I am going to try to build one of these myself over the next week or so.  Let know how you are making out.
David L. HansenProgrammer AnalystAuthor Commented:
Super d-glitch!  I'll get that part number for you and over the next week or so we can watch each other's progress :)
You can also use a mechanical relay with a normally closed contact as the oscillator.
There is a pretty good write-up here:.
David L. HansenProgrammer AnalystAuthor Commented:
David L. HansenProgrammer AnalystAuthor Commented:
Sorry this took so long (I'm getting my shop put together so I can do more projects :) ).  Here's the part number for the coil (I just took a photo to be more thorough).  
Automotive CoilBox with part number
David L. HansenProgrammer AnalystAuthor Commented:
Still going to do your circuit too d-glitch?
hello sl8rz,

I am still moving forward slowly,  but free time has been a scarce commodity lately.

I got a coil for $20 from my local Napa Auto Parts Store.  Not exactly the same as yours, but similar in that it requires an external resistor.  I think that is the best option.

I have done some bench tests and measurements:

With a Fluke meter I get     Primary = 1.5 ohms     
                           Secondary = 8.17k

With a Wayne-Kerr I get      Primary =  3.4 ohms  +  7.04 mH   
                           Secondary = 24.7K      +  56.7  H   all at 500 Hz

The voltage step-up ratio is only 75.  That number matches
what I found on line.  So we need to boost the input voltage 
to 150 or 200 volts using the inductance of the primary.

I have enough info to do a PSpice simulation.  I probably 
won't get arround to building the actual circuit until next 
Monday evening.

Open in new window

Sorry to keep you waiting so long.
My Coil
David L. HansenProgrammer AnalystAuthor Commented:
So, are our coils pretty close as far as specs?  They look almost identical, but I'm sure they all look like that.
David L. HansenProgrammer AnalystAuthor Commented:
Sounds great.  I'll see if I can get my circuit as close to yours as possible before that.
I am interested in what you are doing guys, so I hang up in here.
I made also some projects with high-voltage pulses, but less lower, up to 1.5-2KV only, 8Khz, other application.
My oscillator was not 555, but small PIC microcontroller as PIC10F200 or PIC10F322.
You do not need external components to set the frequency and is more stable, high precision in case is needed. Of course you need the external programmer for PIC, but is not so expensive. In case you have electronics hobby, you even can make a programmer.
David L. HansenProgrammer AnalystAuthor Commented:

So that this thread doesn't get marked as abandoned, I wonder if I should just award the points and then we can continue to use it as we collaborate. <edited>  Thoughts?
viki2000Connect With a Mentor Commented:
Here is a suggestion with PIC microcontroller oscillator:

PIC Oscillator
It simplified the circuit, but requires a bit of programing and the programming tools.
Any small PIC with analog-digital input may be used.
The advantage of PIC is that you can have very precise oscillations which does not depend by external capacitors and resistors, therefore are not influence so bad by temperature. Additional you may have a wide range for oscillation frequency from Hz to MHz. PIC10F322 has the internal oscillator up to 16MHz, but are others, also small type, up to 32MHz. You may see it as precision not necessarily as used frequency. PIC10F322 has inside beside the microprocessor, additional circuits which helps you to obtain very high precision frequency.
By the way is very cheap, costs 50 cents.
If you want, you may replace the potentiometer with simple pushbuttons to adjust frequency or with a small encoder, but that would require a PIC with more inputs. The encoder or push-buttons can give very precisely steps in adjusting the frequency.
If the PIC has more input/output pins, then you can attach a small LCD to it. For instance PIC16F88 is a common type, internal oscillator up to 8Mhz or external up to 20Mhz with enough pins to attach an LCD and a small encoder, also called digital potentiometer.

Why I insist on PIC, digital programing frequency, precision in oscillation?
Because the highest voltage pulses you can get when you are close to resonance of the circuit. In fact, if you find that point and you are able to tune and maintain the frequency stable around that point, the pulses become so high that theoretically may destroy your circuit, physically.
Of course you do not have to do that, but that's the idea to get higher voltage: stability of frequency around the resonance point.

Don't you know the story of the solders which crossed the bridge in marching steps (cadence) and the bridge collapsed?
David L. HansenProgrammer AnalystAuthor Commented:
d-glitch and viki2000,

Hope you don't mind how I did the points (d-glitch you've helped me the most and viki2000 you seem to have some interesting solutions too).  Mostly, I just needed to do something to save this thread from continually being marked as abandoned.  Since this project is so fun (and unfinished) by all means let's PLEASE keep collaborating through this thread.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.