Generating and storing electricity for lightning in a lab

[taylaron]

Hello, I'm designing an experiment that requires a short pulse of electricity at voltages above and beyond 10,000,000 volts at about 300A. I'm having a hard time finding devices that can produce that high of a voltage and current without breaking down, much less finding a way to store that much power. I need the discharge time to be in the milliseconds range.

I thought about a one-time-use device which is essentially a giant capacitor with a strong dielectric between the two plates. the dielectric literally breaks down and an arc connects the two plates momentarily. The difficulty is getting a dielectric pure enough to be reliable. The electric field would be immense... Thats about all I've got.

I need to reproduce a massive lightning bolt that requires the high voltage AND current.

Any thoughts?
Regards,

-Taylaron

 

[Okefenokee]

You're not going to find a 10MV capacitor.

Why not use a coil and a circuit breaker to create a pulse? With some experimentation, you could get the right voltage and current (for a short time).

The problem I see here is that, no matter what you do, you are going to need a 10MV circuit breaker. In other words, you need a breaker that can block 10MV without arcing itself. This simply doesn't exist. Even the high end breakers that use a relay submerged in insulating oil are only rated for about 100kV.

Look at hobby projects for Tesla coils. You'll get a sense of what is achievable.

 

[Okefenokee]

I was thinking about this again and I wondered what some of the records are for producing high voltage arcs. I'll get to that, but first, let's put your figures into perspective. At 10MV and 300A, your device would be burning up 3GW of juice. That's 3 billion Watts, roughly 1/3 of New York City's power consumption. Even if it only lasted for a tiniest fraction of a second, I surely wouldn't want to pay your power bill. I also wouldn't want to have to explain to the power company engineers why dangerous voltage transients are originating from my house.

But, we could do this, if we had a multi-million dollar budget. The end result would, no doubt, be the world's largest coil or the world's largest multi-stage capacitor.

Now, on to the record. I did some Googling to find who has the biggest pulse generator and it appears to be the Russians. http://205.243.100.155/frames/longarc.htm#Longspark" can generate a 5-7MV pulse. An observer saw it dump 678kJ in 10ms. That's an average power output of ~68MW. That's a mere fraction of what you're trying to achieve.

Still, they did accomplish a high-voltage. I wonder what they use for a circuit breaker.

 

[taylaron]

I recognize the power consumption problem, but fortunately the pulse duration will only last a few milliseconds, dramatically decreasing the power consumed. However, the power consumption will still be large in comparison to every day electronics.

It sounds like I need to invent a new way to store high voltage, high current electricity.

 

[f95toli]

Years ago I visited a small fusion reactor where they were basically using a very large capacitor to heat the plasma (for about a 1ms or so); I have no idea about how power they used but I do remember that the "capacitor" was quite a substantial piece of equipment which was housed in a separate room (the capacitor room was bigger than the room where they had the actual reactor).

Anyway, you might want to look at what they use at fusion reactors (JET etc). But, I hope you do realize that even IF this turns out to be possible it will be very, very expensive. My guess is that something like this would cost at least $1M or so (probably mush more).

 

[taylaron]

Thank you for your input. I'm beginning to realize the impracticality of what I wanted to achieve. I have a new question along the same line of work though.
I want to create an electric discharge with current that increases with the square of the time.
My understanding of electric storage devices says this discharge function from a capacitor(s) is not possible and the closest function curve would be very jagged as the discharge amperage decreases over the discharge duration. Chemical batteries seem impractical as the amount of energy released is too great for such short time duration. Correct?
Does anyone have any suggestions on how to reproduce such a discharge function? The device should be able to vary the amount of power consumption from zero to several hundred jewels.
Thank you for your input.

-Tay