Large capacitor bank for a science project

[Pete J B]

Hi, I am building a Rail Gun for a science fair, i am having trouble though. i do not know what kind of resistors i should get for charging\discharging. my capacitor bank is made up of 2000 50V 10000Pf radial type ceramic capcitors. i though all i needed was 2000 50V resistors. please reply by email too, my email address is stupified01@cs.com thanks everyone

 

[chroot]

I would suggest using one of the 0.5 or 1 F capacitors that are readily available at car stereo shops.

- Warren

 

[Pete J B]

i would agree, except they do not have the kind of power (measured in joules) to be useful. they bank i have right now has 100Kj(!) that's over 4 times as much energy as a small caliber naval gun. this much Is needed to experiment with plasma armatures (if your not aware of what ther are look at Powerlabs.org ). i plan to compare the results and get an idea of the kinds of effects plasma has on rail life and velocities. so in my case a 1F capacitor is out of the question( mine also has 2 Farads at 100Kv. thanks for the cocern though.

 

[chroot]

Power is not meaured in Joules. Furthermore, 1 F is 1 F, no matter of what smaller pieces it is composed. The only concern you'll have is the capacitors' equivalent series resistance (ESR), which limits the amount of current that can be drawn from them.

- Warren

 

[Pete J B]

lol, you can measure electrical energy in joules (although it isn't done very much) the formula is E=1/2*C*(V*V) E is energy(in joules)
C is farads and V is voltage. so useing a 1F capacitor with 12 volts would only yeild about 72 joules. which is approximitly 50 Lbf of pressure. which is about enough to acellarate a penny to about 130 fps. so it would still not be wise to use one for a project like mine. not to take anything away from 1F capacitors. You see i already have enough capacitors and a powersupply for charging them. i just need to find the right kinds of resistors for them. but if you are interested in rail gun technology you can make one your self with a 1F cap. screw two pieces of copper bar exactly the diameter of a penny away from each otheron a piece of wood, screw another piece of wood on top and charge the cap. hook it up to the rails so that the panny acts like a switch. make sure the penny is moving when it hits the rails, if it isn't then it might weld to them. any way thanks for your input, but as you can see 72 joules just isn't enough to create the plasma i need (after all the gun will be shooting non-conductive materials, so the plasma makes the connection) plus the plasma will heat the air inside the gun and cause it to expand rapidly, causing a reaction like a gun firing. away way if you know the kinds of resistors i need it would be helpful, and i'll let you know how the rail gun goes. thanks again

-Pete-


P.S. anything is a weapon if you swing it right!

 

[chroot]

Why would you only charge a monolithic 1F capacitor to 12V? You seem to be charging your small capacitors to 50V. If you charge two 1F capacitors to 50V, regardless of how they are assembled, you'll have the same energy stored in them.

And if you're looking for resistors, you really only have two figures of merit: resistance, and power capability. The normal axial resistors you see are 1/4 W resistors. If you can tell me more about your needs, I can help you find a resistor. However, I have no idea why you'd want resistors in the first place.

- Warren

 

[Pete J B]

i see what your saying but becase of the arrangement the total voltage comes out to 100,000. so the joules would be much higher, like two 50V 1F caps would make about 1700joules. my arrangement has 100,000 joules. any way, i need resistors to ensure that all the capacitors get the equal amount of current, so that they don't explode (it has happened b4 :( )but i need one resisitor for each and every capcitor, which totals out to 2000. and they also are going to be used for bleed off, in case something goes worng in the circuit. thanks again.

-pete-

 

[chroot]

How are you obtaining 100,000 V?

- Warren

 

[Pete J B]

by wiring them in series ( think its series, ill check around but i think I am right), but i am having second though about the 100Kv i might rearggange them so that i have 5000 volts at a high Farads (i don't know how much higher yet) but as of right now they are being wired in series.

-pete-

 

[chroot]

Wiring them in parallel does not multiply their voltages. Each capacitor has 12V across it. If you have a total of 1F capacitance, then you have nothing other than a 1F capacitor charged to 12V.

- Warren

 

[Pete J B]

sorry i meant series, plus they are 50V each, and i got 2000 of them. any way i need resistor for each.

 

[chroot]

Originally posted by Pete J B
sorry i meant series, plus they are 50V each, and i got 2000 of them. any way i need resistor for each.

If you wire the capacitors in series, the charge on the positive side of each will cancel with the charge on the negative side of its neighbor. Wiring capacitors in series decreases the total capacitance. The "50V" means that your capacitors are capable of being charged to 50V without breaking down (and becoming shorts). In neither case do you have anything close to 100,000V.

I get the distinct feeling you don't really understand anything about what you're doing. Here's a place to start:

http://hyperphysics.phy-astr.gsu.edu/hbase/electric/capac.html

- Warren

 

[Pete J B]

oh really? than how is it possible for 32 450 volt caps to equal 3.6kv? by wiring sub banks, each subank is made of 4 450V caps wired in parrallel. than all 8 banks are wire in series to creat 3.6 KV at 3100 uF equaling out to 16Kj. so if i do wire mine is series they won't cancel each other out. and i didn't come here to be critized, i came here to get some info on resistors. now if all your are going to do is insult me i might as well not post again, but that would not solve my problem, so you can help or critize, either way i will find my answers and i will complete this project. if you want validation of my claim look at http://powerlabs.org/railgun.htm, he has the cap bank i just told you of.

 

[chroot]

A 450V capacitor is one that is capable of supporting 450V without breakdown. That's all it means. Wiring these capacitors together does not "create" any voltages.

I'm sorry if you don't like criticism. It just seems that without understanding how capacitors work, you're not going to have much luck making your railgun project a success.

And I've already asked you several times about the kind of resistors you need. There are only two figures: resistance, and power handling capability. If you can tell me what you need, I can help you locate parts.

BTW, I am an electrical engineer.

- Warren

 

[Pete J B]

A 450V capacitor is one that is capable of supporting 450V without breakdown. That's all it means. Wiring these capacitors together does not "create" any voltages.

what i mean by create is allow the ablity to charge all the caps, allowing the use of more voltages by spreading it over all caps, i have had 2 engineers tell me that this will work if you would like to see it for yourself just take a voltometer and put 2 AA in series, since they both have 1.5 volts you will get a reading of 3 volts.

yeah i figured that much, I am still working on the exact values ill need ( it has to do with rail length, velocity of projectile, length of pulse, etc. etc.) in any case ill get back to u about the requirements for my resistor, basically i want it to discharge my caps constantly,and to prevent any back EMF to charge them, and to keep discharging them so that if the gun doesn't fire, i can wait a few minutes and it will discharge completely, making it possible to get the projectile out without killing my self.


-pete-

 

[chroot]

Well, there's not much more I can say. I don't really care what you believe. If you think wiring two "50V capacitors" in series produces 100V, put that voltmeter of yours to the test. You're arguing with someone of much greater education than yourself. I'm done with you. Good luck.

- Warren

 

[wimms]

chroot, your fingers are too fast

What he talks about is that he already has high-voltage power supply that will be charging the pack, and that he needs to avoid individual capacitors to get more voltage on them than they can stand.
He needs series connection to get high voltage across terminals of the whole pack, and that's needed to get high currents for short time, overcoming wire resistence and all.

Caps in series do not distribute voltages evenly, but depending on individual leakage currents. Even worse, during very fast discharge, some caps can discharge faster than others, leaving excess voltage across the slower ones, potentially braking them down.

Resistors parallel to each subpack, forming a series resistors ladder is meant to equalise leak currents, thus making voltage distribution more even. Downside is that to be of use, total leak on resistors must be quite high, thus more load on the power supply and more time to charge the whole pack. Also heat dissipation on resistors might become of an issue. Resistors parallel also don't solve the issue completely, only makes it alittle better.

As the discharge time is very short, and currents quite high, I'd say that resistors should be of type with very very low inductance, or you'd be forming LC resonant circuits. Resistence value of few times lower than expected worst cap's leak resistence.

What I'd suggest to also consider is high-voltage zener diodes instead of resistors parallel to 450V subpacks. They'd have low leak currents below their cutoff voltage, and won't allow overvoltage for individual caps. They also work as normal diodes for backemf, thus won't allow much of inverse voltage across caps. Though, depending on currents of backemf, you'd need to make sure you don't blow them up, they won't stand the currents you are playing with.

 

[Pete J B]

thanks, and Chroot,if i did have the know how that you have i wouldn't be here, but even if i did it still won't change the fact that series adds voltages, you obviously must not remember, because if you did know about wiring than you'd know that when wiring in series voltages are added, and when wiring in parallel it adds the Farads. so when i wire mine in series they total voltage well be 100Kv and 2 farads. this equals out to 100Kj, anyway thanks for the help wimms i apreccate the help.

 

[Jimmy]

What kind of supply do you have? What are the specs?

 

[wimms]

When you wire caps in series, you LOOSE farads, alot. You get not voltage, but ability to withstand voltage. To charge this pack, you need voltage source with 100KV. Though you are unsafe at already half that.
Your idea of series voltage addition works only if you charge each cap individually, and then connect in series. You don't want to do that.

 

[chroot]

Originally posted by wimms
Your idea of series voltage addition works only if you charge each cap individually, and then connect in series. You don't want to do that.

Not even then. The charge on adjacent plates will cancel.

- Warren

 

[Jimmy]

Originally posted by chroot
Not even then. The charge on adjacent plates will cancel.

- Warren

If you charged two capictors of equal size to the same voltage separately and then connectected them in series, would the voltage remain the same?

Ex, two 50uF caps charged to 12 Volts. When you connect them in series, you have 25uF effectively. If E(voltage)=Q/C, and if you decrease capacitance by one half, charge will decrease by 1 half, so the ratio of charge to capacitance remains the same.

Is this correct?

 

[chroot]

Eek, Jimmy, you're right. The capacitance will be cut in half, but the voltages will in fact add. Sorry! Brain fart.

- Warren

 

[Jimmy]

Thanks. I was beginning to confuse myself but I believe I got it straight.

I'm in dire need of a refresher course in electronics.

 

[Ivan Seeking]

Another critical consideration here is the ESR - effective series resistance - of the capacitors. This typically becomes the limiting factor for the peak current...and the rise time if that is also significant; such as in plasma guns.

You need to be able to discharge your caps safely upon faults, and you need to treat your capacitor bank like a bomb. These applications can cause catostrophic failures of some capacitors...again usually depending on the ESR. But wiring errors can cause big problems as well.

Next, the hardest part of a high voltage, high current system is the switch. I had really good luck once with a sledge hammer and two large bare wire ends. I also had real good luck by using a high voltage, LOW current source [a spark coil] to generate a spark which then intiates an ionized path, that then completes the circuit to the high voltage, high current source -the capapcitors. This way you have no physical high power switch to contend with. This also acts about as quickly as most any switch - mechanical or electronic - which can also be a concern depending on the application.

 

[chroot]

Ivan,

I mentioned ESR in the my reply to this thread.

Futhermore, most people use a device called a "contactor" to switch high voltage. Depending upon the necessary speed, you might also want to look into IGBTs (insulated-gate bipolar transistors) which can switch several hundred amps in a couple of microseconds.

- Warren

 

[Ivan Seeking]

Originally posted by chroot
Ivan,

I mentioned ESR in the my reply to this thread.

whoops

Futhermore, most people use a device called a "contactor" to switch high voltage. Depending upon the necessary speed, you might also want to look into IGBTs (insulated-gate bipolar transistors) which can switch several hundred amps in a couple of microseconds.

- Warren

At 100 KV, not on his budget I would think. The only solid state devices that I have ever seen rated for switching 100KV were SCRs, but those would hardly be appropriate for this application. Also, in some systems I have seen one-time switches favored because of this problem. The high voltage contacts that we looked at were very slow and they had a lot of bounce. This had a big impact on the energy tranfer due to internal losses in the switching. Maybe someone has found an elegant solution for this issue by now, but I have played with this quite a bit, and we looked at a lot of systems, and we built quite a few, and at the time this was a critical part of the system.

 

[david90]

Originally posted by chroot
Eek, Jimmy, you're right. The capacitance will be cut in half, but the voltages will in fact add. Sorry! Brain fart.

- Warren

I'm an electrical engineer.[/B]

 

[chroot]

david90,

Do you have some point? Am I not permitted one or two mistakes in the thousands of posts I've made here?

- Warren

 

[david90]

take it easy bud. Just kidding around. Didn't u see the ?