Author Options:

I need to make some capacitor? Answered

I need a bunch of high power capacitors and I don't have the money to buy them.  I have a couple ways I have seen and I have done one that kind of works.
2 L bottle - I don't drink enough pop to make enough of these
?pvc? - could I substitute a thin pvc pipe for the 2 litter bottle
Leyden jar capacitors - I saw these cool ones that have a tube like pringles uses, al tape around the outside, and then they just push the can into it as the storage.
?pvc? - Could I also substitute pvc for this.
Styrofoam cups - I have read about, but not seen this one.  Could I just wrap a cup in foil and then stack it.  Apply positive to top cup and negative to bottom


Most folks who inquire about constructing their own capacitors want them for a Tesla coil primary circuit. If that is true in your case, you will want to be careful in the materials and construcion you use, as the cap must withstand the multi kilovolt potential that it will be charged to, and the large bursts of current that it will handle.

One homemade type you did not mention above is the design using glass beer bottles. This seems to be one of the more popular homemade types.
Typically several bottles are needed to acheive the capacitance needed for a moderate sized coil. I don't think that any of the alternate designs you listed above would be better or more rugged than the beer bottle approach.

My first Teslas coil primary cap was made from 1/16" polyethelene sheets for a dielectric, with aluminum roof flashing for the plates, with each plate being a 10" square. If I recall, I needed 12 or more of them, to get the ~0.01uF needed in my particular design. It worked ok, but it would arc over around the edges at times. As I would dial up the transformer voltage with my variac, the edges of the plates would give off a faint blue corona. I eventually bought commerial caps on ebay.

Well, I have seen beer and wine bottle ones, but I don't have access to that many of them.

You can use soda or tea bottles. I made a Geek Group bucket cap using Trader Joe's root beer bottles. Any 12 to 16 ounce long neck, clear glass bottle will work fine. Try asking at a restaurant for empties if you're in a hurry. I once got a dozen identical champagne bottles from a restaurant just by asking.


Do they have to be clear?

Clear is best. Colored glass has metals in it that can affect their insulating properties--green glass, for example has iron in it. Clear glass is pretty much just silicon. I have used amber/brown glass without noticeable problems, but avoid green glass.


so green bottles wont work very well?
(and if this information helps my power supply is 19200 volts)

Conventional wisdom says no, don't use green glass. It has iron in it, which lessens the dielectric properties of the glass and makes the capacitors far less efficient. Use clear glass. Coiler tradition holds that Mexican Corona beer bottles are best--I suspect this is an underground legend. Any smooth, tall, clear glas bottles should work fine.


I need a bunch of high power capacitors and I don't have the money to buy them
Yes, that's because they are expensive to make. You'd be wasting either your time or your money (hopefully not both) trying to do better yourself.


Not really, I have seen good ones with beer bottles and wine bottles, they are byproducts which is definitely cheeper.

OK what are these "high power" capacitors that are too expensive?
Powerful suggests high capacity to me, which you'll not get from used bottles.


I just mean high voltage, the capacitance doesn't need to be that high, I have this solved already though.

Oh good, glad about that.


(It helps to say what you mean in these questions)

You know, you didn't say at what voltage you were planning on using these homemade capacitors.  That's important.

But first, consider for a moment the text book picture of an ideal capacitor. It has two rectangle-shaped conductive plates of area A, separated by an insulator (also called dielectric) with thickness d, and it has a capacitance of:

C= ε*A/d

Also by-the-way, the energy stored in a capacitor is U=(1/2)*C*V2, and the actual physical place where the energy is stored is in electric stresses in the dielectric, which has a volume of A*d. So your energy density in this material is U/(A*d) = (1/2)*ε*(V/d)2 =(1/2)*ε*E2, where E=V/d is the magnitude of the electric field inside the dielectic.

The reason the first equation is important, is because it gives you a clue as to how to make your capacitors bigger.  Basically there are three ways to make C bigger. These are: make ε bigger, make A bigger,  or make d smaller.

The usual trick capacitor manufacturers use is to make d smaller.  In the case of electrolytic capacitors, d is a passivization layer, just a few atoms thick.

But the drawback to making d very, very small, is that the electric field in that region, E=V/d, becomes very, very large.  There are limits to how much electric field any material can withstand. 

I don't think I'll be spoiling the surprise by saying that the most difficult parameter to adjust in your homemade capacitor is the distance between the plates, d. I mean unless it is a homemade electrolytic capacitor, but I can't seem to find a good recipe for one of those.

I mean you can roll up a couple of sheets of aluminum foil with a piece of plastic wrap between them, but it is hard to get them rolled really tight.  I think the last time I tried this I got a total capacitance that was really disappointing, less than 1 uF. 

You could try the same trick with a sheet of paper soaked in some kind of electrolyte.  That is two sheets of aluminum foil, but with wet paper between them.  Then you see if you can "grow" a passivation layer on one of them.

Final note: these capacitors people build for the resonator in a homemadeTesla coil, also Leyden jars.  Those capacitors have very small capacitance, since d is big and thick.  But the voltages placed across them are very large, e.g. 10s or 100s of KV.  Thus the energy stored in them, U=(1/2)*C*V2, can be significant. 

Also note that in a typical glass(or plastic) bottle Leyden jar, the only material storing electrical energy is the glass(or plastic) dielectric layer .  The rest of volume of the bottle is wasted space.

Anyway, what I'm saying is that for voltages less than 1KV, you are not going to be able to store much energy in homemade capacitors with a plastic dielectric layer simply  because it is hard to make that layer as thin, d as small,  as the guys in capacitor factory can.

Homemade electrolytic capacitors, rolled from sheets of aluminum foil, with homemade superthin passivization layer, that might be another story...  but I still have not found any good recipes for those.

What capacitance are you aiming for? What are you going to use the caps for?

My Leyden jar array is cheap and easy to build, but bulky and may take you a while to amass the materials. The design easily handles up to 100kV at 30 mA. I have run them as tank capacitors in a Tesla coil circuit running at 60 mA without a problem, too.

Quicker (but heavier, messier) is the Geek Group's bucket capacitor, found at http://www.youtube.com/watch?v=hjqKcRQB3RI

Paul aka Mr. Apol

How about double sided PCB material ?