Can a Transistor Effectively Regulate the Release of Electricity in a Capacitor?

[taylaron]

Hi, I am devising a way to store large amounts of electricity, but at the moment it is more like a capacitor in that it greatly tends to release the energy in a spark.

I'm trying to come up with methods to regulate the transmission of electricity from this capacitor without wasting the majority of the energy.

I've conceptualized various semi-conductive polymers that act as the medium where it is partially conductive and will only let x amount of current pass through them. but I am concerned about current only passing in short pulses like a lightning bolt through air. Also once the voltage drops low enough in the capacitor, I suspect it will not be able to travel through the medium which "wastes" the remaining energy. I've also heard of using a network of capacitors in parallel to regulate the voltage. but this doesn't seem adequate.

In a lightning bolt, the bolt finds its path to the ground and once it reaches the ground, the majority of the static electricity is transferred. you can find this in slow motion videos of lightning. Is there a device or substance that can regulate how much power is going through it? this reminds me of a transistor, but with my understanding of the technology, the transistor wouldn't be able to vary its capacitance in a wide enough range to gradually release the electricity with useful results and without vaporizing the transistor.

My thoughts are really scattered at the moment, but any suggestions or advice would be greatly appreciated.

Regards,
-Tay

 

[vk6kro]

If you had a capacitor of 10 uF charged to 100000 Volts and momentarily connected it in parallel with a 0.1 uF capacitor, the smaller capacitor would charge to about 99010 volts and the big capacitor would also drop to the same voltage.

If that capacitor was discharged, the majority of the power would still be in the main capacitor. And you could do it again many times but at a slightly lower voltage each time.

At these voltages, though, you should probably use motor driven wheels with shorting metal bars on them.

 

[taylaron]

Thanks vk6kro, I agree this could work but using several capacitors in parallel would not allow a steady release of energy. It would be in pulses if I'm correct.
I'm striving to output DC power at around 5V using as small and simple a circuit as possible.
any other ideas??

 

[TurtleMeister]

I've thought about doing the same thing and this is what I came up with. Please note however that I've never tried it and I don't know if it would work. Also, your storage capacitor voltage will be limited by what the switch can handle.

Use an http://en.wikipedia.org/wiki/Electric_double-layer_capacitor" to reduce loses and slow down the charging of the output capacitor. Finally, use a 5v regulator for the output.

 

[taylaron]

Interesting, so let me make sure I understand correctly.

To regulate the output of the super/ultra capacitor you would use a high speed switch to turn the connection on and off at extremely high frequencies acting like pulse-width modulation. The faster the switching rate, the greater percentage of the charge in the capacitor you get and the slower the switch frequency, the less power it outputs over time. The frequency would change according to the load required and the remaining charge in the capacitor.

My one concern: Wouldn't the extreme voltage on the incoming side of the switch be so great that it would arc across to the other terminal frying the switch??

 

[TurtleMeister]

Yeah, you've got the idea. However, the switching rate, or frequency, would not be as important as the duty cycle. You would only need a frequency high enough to permit adequate filtering for the regulator input. The duty cycle would determine the rate of energy transfer from the storage capacitor to the output capacitor.

My one concern: Wouldn't the extreme voltage on the incoming side of the switch be so great that it would arc across to the other terminal frying the switch??

Also, your storage capacitor voltage will be limited by what the switch can handle.

The buck converter is the most important part of this idea. The inductor will prevent the charging current from going out of sight when the switch turns on. The rectifier will prevent arcing when the switch turns off. The tricky part is determining the best value of inductance.

 

[taylaron]

Haha, neat. For your information, I am a college student. That said, is it reasonable to replicate this circuit and determine the appropriate inductance in my spare time or does it require significant funding and facilities? lol. Can I calculate the inductance?

-Tay

 

[TurtleMeister]

Well, like I said before, this is just an idea that I've never carried any further than what I've told you. You were looking for ideas, so I thought I would throw it out there. Also, I'm not an engineer, just a hobbyist. So take it for what it's worth. Good luck with your project.

 

[Bob S]

Hi taylorone-
For short, regulated voltage pulses, look into pulse-forming networks (PFNs). A typical example might be charging up a 1-microsecond long RG-8 cable to 10 KV, and discharging it into a load. You will get a 2-microsecond long, 5 kV pulse. There are many variants of PFNs, including capacitor banks. See
http://en.wikipedia.org/wiki/Pulse_forming_network
Bob S

 

[taylaron]

Interesting Bob, It it sounds like this PFN outputs a very high frequency high voltage square pulse. Could I use a buck converter or another device to eliminate the pulse component and convert it to a low voltage non-Hertzian output?
How big is this device? I would think this synthetic transmission line is quite large.

 

[Bob S]

Hi Tayaron-
I have seen a 1.8 microsecond , 7000 volt, 700 amp PFN made up of 5 paralleled RG-220 cables each about 600 feet long. The switchgear also takes up a lot of space. So a PFN is not small. It also takes a lot of time to recharge.

Bob S

 

[taylaron]

Ok, Thanks Bob. Can anyone think of any other solutions?

-Tay

 

[taylaron]

So expanding the duration of a capacitors discharge by cyclically changing its resistance (a switch) through a substance very quickly is the only way to reduce the discharge of current per unit time?

What about a semiconductor that increases its resistance with the increase of current without losing the majority of the energy through heat?

What about a series of metallic carbon nano tubes short enough to be ballistic conductors lined end-to-end to create a limited amount of surface area? The amount of surface area (having no resistance) would ideally dictate how much electricity can flow through through the perfect conductor at the speed of light.
-So, only x amount of electrons can flow through the series of ballistic conductors per unit area per unit time and not produce any measurable heat.

Thoughts anyone? Perhaps my theory is inaccurate? I admit the probability of growing such a CNT with no resistance is highly improbable so a minuscule amount of heat would be produced, correct?

 

[arithmetix]

May I suggest that if you can manufacture a capacitor for such a huge voltage you probably can manufacture 10 or 100 lower-voltage, higher capacitance capacitors, which, in series, could be charged to the same potential. Then all you'd need to do is move them from series to parallel, and you'd have your lower voltage and lots more current, all without dissipating power in the current with which you would otherwise have flowing to the smaller cap.

 

[taylaron]

Interesting arithmetix, so instead of simply having one big honkin' capacitor with a massive voltage you're suggesting I "split" the capacitor into smaller segments and put them in series, charge them and then put them in a parallel circuit and get a lower voltage but higher current? Am I understanding this correctly? The size of each supercap would then be proportional to the output voltage and the number of supercaps would dictate the available current when in parallel. Correct?

 

[Bob S]

A marx generator is the opposite of what you want, but here is a nice website:
http://www.penguinslab.com/marx.htm
What exactly do you want; Coulombs, duration, amps, voltage?
Bob S

 

[taylaron]

So here is my device and desired situations:
My device consists of two containers surrounded with a strong dielectric material. A *secret* material/ substance is then inserted into the containers after various property modifications. The theory behind this "hyper capacitor" is electron diffusion (yes it is a term..). A large amount of energy at a very high voltage is dumped into one of the containers via a conductor through the dielectric (the container connected to the negative terminal). The *secret* material is able to discharge all its load almost instantly so a dramatic discharge or spark is going to tend to be produced from the charged container unless it is somehow prevented. This is unlike batteries which the flow rate is limited by a relatively slow chemical reaction.
In my device, after I somehow prevent the spark and "reduce" the power to an order of voltage and current that portable devices like computers, handheld devices, etc... can handle without being fried the "usable" load is now passes through whatever circuit it powers and then the electrons flowing from the negative terminal traverse through the reducer, through the desired circuit and enters the other dielectric container via a similar penetrating conductor. The charge then accumulates on the uncharged *secret* material inside the other dielectric container.
To my understanding only half of the initial charge inside container 1 will be used to power the device because the two containers will eventually reach equilibrium. (Unless I can come up with a way to move all the charge from one container to another without consuming a lot of energy...)

Another use for my device is for storing the energy not used from the grid in the USA to be used later and not wasted through heat because at the moment there is not a practical or existing method of storing large amounts of power in the form of electricity. My device is very scalar so if I succeed it could power a city(s) or a MP3 player... It all depends on the size.

I checked out the marx generator on your link and I find it interesting in that it seems similar to the possibility of splitting the dielectric containers into much much smaller units at a larger quantity to achieve the desired voltage and current (like you said: just the opposite)
Can the marx generator be modified to run "backwards" as to solve my issue of reducing the output to prevent a spark?

 

[Bob S]

See my post #7 in the thread:
https://www.physicsforums.com/showthread.php?t=329842
A 1000 volt capacitor is pulsed into a coil, the coil current rises to about 8,000 amps, and suprisingly, the capacitor ends up with -800 volts (note minus sign) on it. So reverse the capacitor polarity and do it again.

For really big energy storage I suggest:
http://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant
This unit can store enough energy to supply about 8,000 MW for up to 8 hours.
Bob S

 

[arithmetix]

Re post #14:
That's what I had in mind, Taylaron, and I think that theory is correct.