hot, bright noisy sparks needed

For an art project, I need to generate a bright, noisy arc about 1/2" long. In the past I've successfully done this with with this: https://www.sparkf un.com/products/11218 but that isn't working in the new installation; the d evices get hot and fail after a few minutes of use. I'm looking for someone with expertise in this area to suggest a design I can build.

The requirements are:

Able to be run off a small 12v lead acid battery. Extra points if it can ru n off 5v and 3A, which is what the sparkfun device uses.

Portable - I'll be carrying the circuit in a backpack. Large Marx generator s need not apply. Neither do designs that require twenty pounds of copper t ubing.

Stable and Safe - I'll be carrying the circuit in a backpack and long term stability is a concern. It will be heavily insulated but I am still very un interested in sudden failures, magic smoke, fires, internal arcing, and oth er surprise effects. I need long-term reliability.

A bright, fat, noisy, snarling, visually continuous arcing spark - which pr obably means I need an output frequency somewhere between 40Hz and 200Hz. I don't have specs on the sparkfun device, but it does "loud" and "bright" v ery well.

Capable of more or less any duty cycle; the design calls for about 10 secon ds on and a few seconds off, repeating, but I want it to be able to run con tinuously.

Able to survive any sort of spark gap from "too big to actually arc" to "br ief short circuit", but the typical gap is 3/8ths inch to 3/4ths inch.

Repeatability - I need to be able to build several and I want repeatability , so "scavenge a flyback from a CRT" is much less interesting than a mouser or digikey part number.

And - crucially - able to handle a capacitive load. The high voltage travel s through two longish (several feet) parallel wires, though a small brass t ube and into a chunk of aluminum that showcases the spark, but it all adds to the capacitance, and I think that's why the sparkfun devices are dying.

I'll say it now - yes, I know a spark that can bridge 3/4" and be "fat", br ight and noisy requires tens of thousands of volts and several mA. I know t hat misuse of that kind of power leads to pain and possible death. Since I' ll be carrying the device and the high voltage wires will be running down m y arm I will be paying a lot of attention to insulation and safety concerns . Extra points if the voltage and current can be minimized to limit risk, b ut no matter what I will consider any design as potentially fatal and treat it accordingly. I know about the one hand rule, and "heavily insulated" is my favorite phrase.

I just built a Mazilli ZVS circuit

formatting link
and it works as advertised but it's not sui table here. Running the output through a twisted pair of insulated bell wir e caused a very pretty purple corona glow along the length of the wires, an d no spark at the end. When it does spark. it's much too high frequency - i t hisses instead of snarls, and glows instead of being a fat bright spark. And it's too powerful - the voltage cracks the glass tubing I use as insula tion. If it's possible to lower the frequency (greatly) without destroying the MOSFETs, it could be a candidate, but I don't know how to do this safel y.

I'll accept any help, but I'm less interested in "well, you could try this" than "here is what is known to work." I've been researching things I could try for days, but mostly I've fried components and lost time. If you have expertise and don't mind drawing a schematic I will be very grateful. TIA!

Reply to
ScottM
Loading thread data ...

Oil burner ignition transformer is one way. A 12V to 120V inverter is needed. Then again, here are several hits on google

formatting link

Reply to
Oppie

Arrgh!

formatting link

Reply to
Oppie

g. In the past I've successfully done this with with this: https://www.spar kfun.com/products/11218 but that isn't working in the new installation; the devices get hot and fail after a few minutes of use. I'm looking for someo ne with expertise in this area to suggest a design I can build.

run off 5v and 3A, which is what the sparkfun device uses.

ors need not apply. Neither do designs that require twenty pounds of copper tubing.

m stability is a concern. It will be heavily insulated but I am still very uninterested in sudden failures, magic smoke, fires, internal arcing, and o ther surprise effects. I need long-term reliability.

probably means I need an output frequency somewhere between 40Hz and 200Hz. I don't have specs on the sparkfun device, but it does "loud" and "bright" very well.

onds on and a few seconds off, repeating, but I want it to be able to run c ontinuously.

brief short circuit", but the typical gap is 3/8ths inch to 3/4ths inch.

ty, so "scavenge a flyback from a CRT" is much less interesting than a mous er or digikey part number.

els through two longish (several feet) parallel wires, though a small brass tube and into a chunk of aluminum that showcases the spark, but it all add s to the capacitance, and I think that's why the sparkfun devices are dying .

Excellent! I was going to ask why the spark fun thingie wouldn't work. You may not like this answer, but it might be easier to try and reduce the cab le and stray capacitance rather than design a beefier HV supply. So perhap s you can separate the wires that carry the HV... and maybe move the genera tor closer to where you want the arc.

bright and noisy requires tens of thousands of volts and several mA. I know that misuse of that kind of power leads to pain and possible death. Since I'll be carrying the device and the high voltage wires will be running down my arm I will be paying a lot of attention to insulation and safety concer ns.

Oh! (don't kill yourself) OK some thin HV wire with big *thick* insulation . (Personally I wouldn't touch this project with the proverbial ten foot p ole.)

Extra points if the voltage and current can be minimized to limit risk, but no matter what I will consider any design as potentially fatal and treat i t accordingly. I know about the one hand rule, and "heavily insulated" is m y favorite phrase.

uitable here. Running the output through a twisted pair of insulated bell w ire caused a very pretty purple corona glow along the length of the wires, and no spark at the end. When it does spark. it's much too high frequency - it hisses instead of snarls, and glows instead of being a fat bright spark . And it's too powerful - the voltage cracks the glass tubing I use as insu lation. If it's possible to lower the frequency (greatly) without destroyin g the MOSFETs, it could be a candidate, but I don't know how to do this saf ely.

OK again I think you need wires with better insulation.. You can adjust the frequency of the flyback. (I think)

George H.

s" than "here is what is known to work." I've been researching things I cou ld try for days, but mostly I've fried components and lost time. If you hav e expertise and don't mind drawing a schematic I will be very grateful. TIA !

Reply to
George Herold

Basics:

Add a common points-type ignition coil and get a juicy-enough spark that you can saw Plexiglas with it >:-}

This dates to my early '70's ignition system designs. Now-a-days, replace NPN with a power MOS FET, capable of around 450V VDS. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

For insulated wiring, you should be able to use spark plug cabling. Don't use the resistive kind, use the racing kind with wires, not carbon. From memory it should be good out to around 10-20kV

I used to work in 150kV to 200kV power supplies. I cannot stress enough, Just keep THINKING: at high voltage EVERYTHING CONDUCTS! And probably not like you expect. There is 'through' conductivity and 'surface' conductivity. Even AIR conducts, not much, but enough to quickly build up charge somewhere else that may wish to arc over as the voltage gets too big.

Watch out for sharp edges. of anything. make stuff rounded with clean radii. Even inside an insulating blob a sharp point can ionize and set up disastrous failure mechanisms later.

Reply to
RobertMacy

formatting link

Reply to
Rheilly Phoull

You mean, like one of these:

formatting link

--
Regards, 

Adrian Jansen           adrianjansen at internode dot on dot net 
 Click to see the full signature
Reply to
Adrian Jansen

Could a metal jacket help. I don't know what they are called, or how do you get them. But I have seen cloths woven of/with? metal wires. At least larg er sparks do not penetrate them. But I am not going to try my self, of cour se. So it is your skin not mine.

Reply to
LM

The oil burner transformer looks like a possibility, but I can't tell from the description how to drive it. Does it spark continuously when power is a pplied? Can I pump 12v into it as a 40Hz square wave at, say, a 50% duty cy cle and get a 40Hz spark out the other end?

On changing the wire length - I probably cannot. The spark gap is a piece o f machined aluminum held (via appropriate insulators) in my hand. The circu it is in a backpack. And since I'd prefer the high voltage wires not be tig ht against my arm, I need at least 3-4' of wire between the driver and the spark gap. For the last 8" or so the wire goes through a narrow brass tube; if that's a significant source of capacitance I could perhaps replace that with plastic. The aluminum part itself probably adds some as well, but for what I spent getting them machined I'm not going to discard them. I design ed this whole thing around the sparkfun devices, because they worked ok in previous and vaguely similar designs, but the rash of rapid failures has le ft me with a large investment in a design that I can't modify very much, gi ven the costs. If there are ways to reduce the capacitance between two wire s (other than "make them shorter and further apart") I'm interested.

The 3v spark circuit looks like a more fragile cousin to the sparkfun devic e. For the price I could try it, but I'm skeptical. The stun gun circuit wo rries me because it's overtly designed to hurt people and that's something I need to avoid the suggestion of. I want the spark to look scary, but if p ossible not BE scary. I'll look at the ignition circuit.

Right now I'm wondering if the ZVS circuit I built could be modified. It pu ts out high frequency (~70kHz?) AC, which I assume is why everything I conn ect to it acts like a capacitor and drains the power away in a corona befor e it reaches the spark gap, but what if I rectified and smoothed it into DC ? And then turned the ZVS circuit on and off at 40Hz or so?

Reply to
Scott M

Den fredag den 1. november 2013 00.07.18 UTC+1 skrev Scott M:

m the description how to drive it. Does it spark continuously when power is applied? Can I pump 12v into it as a 40Hz square wave at, say, a 50% duty cycle and get a 40Hz spark out the other end?

of machined aluminum held (via appropriate insulators) in my hand. The cir cuit is in a backpack. And since I'd prefer the high voltage wires not be t ight against my arm, I need at least 3-4' of wire between the driver and th e spark gap. For the last 8" or so the wire goes through a narrow brass tub e; if that's a significant source of capacitance I could perhaps replace th at with plastic. The aluminum part itself probably adds some as well, but f or what I spent getting them machined I'm not going to discard them. I desi gned this whole thing around the sparkfun devices, because they worked ok i n previous and vaguely similar designs, but the rash of rapid failures has left me with a large investment in a design that I can't modify very much, given the costs. If there are ways to reduce the capacitance between two wi res (other than "make them shorter and further apart") I'm interested.

ice. For the price I could try it, but I'm skeptical. The stun gun circuit worries me because it's overtly designed to hurt people and that's somethin g I need to avoid the suggestion of. I want the spark to look scary, but if possible not BE scary. I'll look at the ignition circuit.

it doesn't matter if you write stun gun or spark generator on the schematic it is basically the same thing, and anything that makes a 1/2 spark is scar y

-Lasse

Reply to
Lasse Langwadt Christensen

I'm intrigued that what you had "broke glass". Got a photo of that?

Reply to
Greegor

There's now a hairline crack in a 6mm bit of glass tubing (the sort used in chemistry). Inside the tube is one wire of the output; just outside the tu be and all around it is the aluminum that forms the other leg of the output . In one test the arc decided that cracking through the 1mm or so of glass tube wall was a better option than the 3/4" of air I'd planned for it.

From videos I've seen on the web this kind of behaviour isn't too uncommon for this oscillator. Someone running one at 36v decided to put the output a cross (not through) a glass lightbulb and had something similar happen: htt p://

formatting link

I don't know if my phone camera can do a good job of photographing a hairli ne crack in a small glass tube but if I can get it to show up I'll send it along.

Reply to
Scott M

Can you explain the workings? If I'm reading that right - no guarantees the re - it looks like all the spark comes from the discharge of the 1uf cap, a nd I'm surprised there's enough power available from one. It also looks lik e the transistor gets to sink not just the power from the cap, but all the amps from the battery, when it goes on. Does it need a massive heatsink? I' m using IRFP260 for this kind of thing.

I'd expect to drive it with a 555 - what kind of timing would it want? From the notes it looks like the "on" time is critical...

Reply to
Scott M

[...]

Don't do it. Find another project.

--
~ Adrian Tuddenham ~ 
(Remove the ".invalid"s and add ".co.uk" to reply) 
 Click to see the full signature
Reply to
Adrian Tuddenham

More to the point anything producing hot bright electrical sparks 1/2" long is going to be dangerous or even deadly if you make a mistake.

Your best bet is something along the lines of a Jacobs ladder device where the rising plasma allows a lower voltage that strikes across the narrow lower arc gap to rise and carry a noisy spark with it.

It probably contravenes lots of RF emissions regulations too.

I expect to read about this exploit in the Darwin Awards.

--
Regards, 
Martin Brown
Reply to
Martin Brown

I appreciate the concern, but I've built devices like this several times an d used them successfully for days at a time, based around the sparkfun devi ce. I understand the risks and I understand how to manage the risks.

That said, I'm always interested in lower voltage approaches. Jacob ladder drivers would generate more voltage than I need, since they are trying to s ustain an arc larger than 1/2" at many points. There are designs on the web for circuits that generate several inches of arc, but they aren't anything I want to carry in a backpack. The ZVS oscillator is not a favored candida te for the same reason - there is more power there than I need.

Reply to
Scott M

This doesn't strike me as very creative.

Reply to
Greegor

You appear to miss a fundamental point of what you are playing with. (I spent about a decade working in a plasma lab...)

The voltage required to START an arc is much, much higher than the voltage required to SUSTAIN an arc. Once started, an arc (plasma) is a pretty darn good conductor.

Martin Brown (quoted below) explained it, but you skipped right past that. By striking an arc on a small gap, a much larger gap can be _spanned_ without needing a voltage that will _start_ an arc on the large gap.

Sharp, pointy electrodes also lower the voltage needed for arc initiation. You could probably follow the lead of most sustained arc circuits and have a pulse generator of a high voltage and short duration to start the arc, and a lower voltage sustained supply to maintain it. Every florescent light does this, for instance.

I'm a bit concerned about your proclivities from this snippet from your OP, as well:

HV --> Bell wire? Seriously? Shee-zeus! Does the concept of a voltage rating for insulation pass right over your head? HV wire with nice fat silicone insulation is really easy to get. HV-Heat-shrink can add some insurance (even regular heat-shrink adds quite a bit.) The more cautious approach is to run the HV inside a grounded sleeve - either coax or sleeved in hollow braid. Keep capacitance down by using small-center conductor large insulator coax.

But...given the size that "taser" type circuits can be shrunk to (and the truth that they are pretty much exactly what you are building, no matter how uncomfortable that makes you), no real reason to run HV along your arm - run 12V DC along your arm and limit the HV generation to right where it's used. If you manage to zap yourself then, you'll probably live (your hand will hurt like hell, but there's no reason for the current to travel through your heart, as there is with a back-pack generator and an arm terminal, should anything go wrong.)

Have fun.

--
Cats, coffee, chocolate...vices to live by 
Please don't feed the trolls. Killfile and ignore them so they will go away.
Reply to
Ecnerwal

Tesla coil? Jacob's ladder? Whimhurst wheel?

Bet on that.

Reply to
krw

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.