By Mark Frauenfelder at 10:27 am Wed, Mar 4, 2009
Collin Cunningham of MAKE made this terrific video that explains what a capacitor is and how to make your own.
In January, Collin produced a similar video about LEDs.
very educational. Now if you REALLY want to get killed:
Now I can make my own time machine… anyone got a spare DeLorean?
Another example of how school signally failed to teach me anything that stuck. Why couldnt they have done this with us? I struggled with electrical stuff all through A-levels and engineering degree. I would certainly have done better if taught like this.
now show us how to make a flux capacitor and we will have something to talk about.
I see someone has already linked to Bill Beatty’s site, but if you want to understand “electricity” (in quotes because the term confuses many different concepts) the articles on this page will sort you out:
Bill Beatty’s articles about “Electricity”
This article will make capacitors clear, avoiding the common errors perpetuated in the MAKE video:
“Capacitors store charge.” No! Flat out wrong! ”
“… one plate has less electrons and excess protons, and the other plate has more electrons than protons. Each plate does store charge.
However, if we consider the capacitor as a whole, no electrons have been put into the capacitor. None have been removed. The same number of electrons are in a ‘charged’ capacitor as in a capacitor which has been totally ‘discharged.'”
“My favorite capacitor analogy is a heavy hollow iron sphere which is completely full of water and is divided in half with a flexible rubber plate through its middle. Hoses are connected to the two halves of the sphere, where they act as connecting wires. The rubber plate is an analogy for the dielectric. The two regions of water symbolize the capacitor plates.
Imagine that the rubber plate is flat and undistorted at the start. If I connect a pump to the two hoses and turn it on for a moment, the pump will pull water from one half of the iron sphere and force it into the other. This will bend the rubber divider plate more and more. The more the plate bends, the higher the back-pressure the plate exerts, and finally the pressure will grow strong enough that the pump will stall. Next I seal off the hose connections and remove the pump. I now have created a “charged” hydraulic capacitor.
Now think: in this analogy, water corresponds to electric charge. How much water have I put into my iron sphere? None! The sphere started out full, and for every bit of water that I took out of one side, I put an equal amount into the other. When the pump pushed water into one side, this extra water also forced some water out of the other side. No water passed through the rubber. Even so, essentially I drove a water current THROUGH my hydraulic capacitor, and this current pushed on the rubber plate and bent it sideways. Where is the energy stored? Not in the water, but in the potential energy of the stretched rubber plate. The rubber plate is an analogy to the electrostatic field in the dielectric of a real capacitor.
It would be misleading to say “this iron sphere is a device for accumulating water”, or “this sphere can be charged with water, and the stored water can be retrieved during discharge.” Both statements are wrong. No water was injected into the sphere while it was being “charged.”
Imagine that I now connect a single length of pre-filled hose between the two halves of the capacitor. As soon as the last connection is complete, the forces created by the bent rubber plate will drive a sudden immense spurt of water through this hose. Water from one side will be pushed into the other side, and the rubber plate will relax. I’ve discharged my hydraulic capacitor. How much water has been discharged? None! A momentary current has flowed through the sphere device, and the rubber plate is back to the middle again, and the water has become a bit warm through friction against the surfaces of the hose. The stored energy has been “discharged,” but no water has escaped. The hydraulic capacitor has lost its energy, but still has the same amount of water. “
he’s a great explainer.
When we were really bored back in college, we’d get a big electrolytic capacitor, wire it to an extension cord, and plug the cord into a 110 AC line. kaboom.
Would it be more correct to say that potential energy is stored?
It’s much like carrying a weight up to the top of a water wheel – thereby increasing its potential energy. When you’re ready to use that energy, you drop it on the water wheel, making it turn and do whatever.
That’s the way capacitors were explained to me anyways. Along with “These things do stuff with power. They blow up when hot, don’t jab one with your screwdriver.” (my dad’s explanation).
I just can’t get away from this. I spent most of the day yesterday arguing about capacitors with strangers. AudioKarma is great for that. There is disagreement over whether a cap with a net DC charge is better at passing AC than one with an un-biased dielectric. Superficial analysis says ‘maybe’. People who paid money to make their stereo speakers sound better by installing this ‘upgrade’ say yes. Several people spent a large part of their day chewing this over.
and after that big one nearly nailed you? Migods, you’re being stalked by capacitors!
Charging an electrolytic capacitor can be very dangerous, if you try to charge it wrong (- to +, + to -) it explodes.
Gilbert Anonymous here:
That’s a cool bead. I can’t wait to see who pops up on screen when he finishes assembling the interocitor. I’m hoping it’s Exeter.
Sometimes they explode just for the hell of it. There’s a lot of energy in some charged up caps and if it comes out in a hurry, don’t be in the way.
I had the distinct displeasure of trying to prolong the life of audio cards in some Panasonic DVC-Pro decks. The electrolytic capacitors were surface mount. They were rated for so many hours at a certain temperature. In typical use that meant 3 to 4 years life. I worked with eye shade like head mounted magnifying lenses just to see them. They were tested with an AC ohm meter (impedance meter) that generated 5kHz. Capacitors could thus be tested in circuit. If their impedance was excessive they were replaced. In some cases all of a certain value were replaced just because they were most likely to be bad.
There were some dual polarized electrolytics but I never came across a description of how they functioned.
Back when everything was more macro than micro I had some very high quality 2 microfarad capacitors from Korean War Signal Corps. gear. In high school we’d charge them to 400 volts and leave them lying around our geek corner in the electronics lab. Soon word got out to the non geeks to avoid that corner.
Dang, I wish he would have told us what they are used for, practically, in electronics. He almost did, then said we would make one and never came back to tell us. :(
“Duct tape is a rip-off. I make my own duct tape.” — Joe on Newsradio
I think Ultan hit it on the nail.
When it comes to circuits, the best way to learn how components function is to think of the hydrodynamic analogy. It’s not perfect, but it helps beginners understand how things function.
Voltage sources can be modeled as a pump that pushes water through at a constant pressure (psi), current sources can be modeled as a pump that pushes a constant volume (gallons per min.). A resistor can be thought of as a bottleneck in a pipe. A capacitor is like having a rubber diaphragm blocking the water. Think of taking a condom and putting it over a pipe, then connecting the pipe into the rest of the plumbing. the condom stretches with the flow of the pipe until it can’t stretch anymore, thus stopping the flow of water. if you cut off the pressure feeding into the pipe, the condom will remain pressurized and would squirt water out if there was a valve to release the pressure. If the water was moving back and fourth in the pipe very quickly, the condom would have an entirely different effect on the water and in some cases wouldn’t appear to have an effect on flow or pressure at all.
What’s cooler is an inductor in this water model. It would be like having a long pipe where water flows freely. The water builds an inertial mass and likes to keep flowing – so if you try to stop it suddenly it creates a high pressure. Think of the noise your kitchen sink makes when you shut it off quickly, it’s like the water suddenly hits a wall and has nothing to absorb the kenetic energy which creates high pressure. An actual inductor is a little more complicated because of the magnetic field involved, but you get the idea.
Bill Beatty’s water analogy is more complicated than it really needs to be. The simple parallel plate capacitor, ignoring dielectrics, is a good place to start in wrapping your head around the capacitors properties.
There may be more or less net electrons in the capacitor when it is “charged,” but it still has a charge differential, and an energy potential.
In the hydraulic model, it looks like if no water flows, then nothing really happens. Electron drift is very slow and even if the net amount of electrons in a device doesn’t change, it doesn’t mean productive energy isn’t being stored and released.
Who, in this age of DSLRs, still has 35mm film containers?
p.s. This video is awesome. Why isn’t it Creative Commons licensed, or hosted on blip.tv where it can be easily downloaded?
p.p.s. Here are the URLs from the end of the video, hyperlinked for your convenience:
Spark, Bang, Buzz
Beavis Audio Research
Water analogies for electricity work until you get to tubes and semiconductors. Then you’re stuck with a mental model that fails utterly on active devices, which are the really important components. I recommend thinking of electricity in terms of electrons. This is made difficult by the convention of referring in many schematics to current flow as ‘hole flow’ which is opposite to the flow of electrons. My formal education in electronics started 34 years ago and still nobody has explained to me where that nonsense came from.
this is fun:
So as soon as I said there wasn’t one, I found a fluid analogy for transistors.
Great, but shouldn’t this have been part of high school physics? Hell, I still remember the relation between capacitance, distance and the diameter of the plates from my high school physics class.
Well, I like the rubber diaphragm analogy for the capacitor, so I guess I will attack the question about their use in circuits:
[Let me note that the physics description of a capacitor is: Localized concentration of the charge field. This def is important in RF]
There are two distinct schematic symbols for Cs. One is a typically horizontal device with two parallel lines. The other is a typically vertical device with one of the (typ. the lower) lines now converted into an arc. Callouts for the first type are sometimes ‘DC block,’ or ‘decoupler.’ Callouts for the second type are sometimes ‘PS bypass’ or ‘PS decoupler.’ Let’s say we are plugging our walkman into our home stereo. We only want to convey signal energy. So the front-end of the stereo will have a decoupler. In a properly designed ckt, the input side of this cap will have a pair of back-biased, low-drop, fast diodes across the supply, with the central tap connected to the input. There is no reason that the walkman and the stereo will be at the same rest potential when they are connected. This circuit will protect us. Of course, a connector that makes the gnd connection first will keep us from hearing the Dirac drop when we plug in. What value should this device be? Theoretically, very large. But then this device would be big and expensive. Capacitive reactance is 1/2 * pi * F. This gives the counter-intuitive result that larger capacitors feature lesser Z. This is not the case with Rs and Ls. If our walkman impedence is 2K and our stereo impedence is also 2K, a capacitive reactance (determined through algebra) on the order of <2K, at our lowest frequency of interest should serve us well. At frequencies above this, our signal will be in phase and not (significantly) attenuated by the decoupler. Interstage amps in the stereo will likely feature their own decouplers. Sometimes, interstage decoupling is achieved through transformers, instead. Now imagine that you are a fast logic chip. If your power supply isn’t as fast as you are, all of your swift efforts will be in vain. Power supplies aren’t, typically, all that fast. Just as bad, the traces between you and the PS are inductive, effectively slowing things further. What’s needed is a capacitor in shunt with the supply, located proximately to your supply pins. When the circuit powers up, we will have to wait a moment for this PS decoupler to charge, and we will have to deal with the stored energy when we power-down, but otherwise we are cool, as long as the cap is rated for a breakdown potential greater than the supply potential. Else it will blow up, and that would be less cool than a prohibitionist ranting on FARK! How large (in value) should this cap be? Again, theoretically, the bigger the better. Again, practical limits come into play. Actual caps have equivalent series and shunt resistances and inductances, which have to be considered if you are a really fast chip driving a low-Z bus. These are the two largest uses of caps, accounting for at least, say, 90% of applications. There are all kinds of caps. If we were an RF circuit, we might need expensive silver-mica caps for stability and speed. If we were a shmancy audio circuit, polypropylene caps might be the order of the day. Russia, with her huge material resource base, probably makes the best of both of these kinds of caps. If we were making an equilizer circuit, or other type of filter, ratios of values of our C with associated Rs would be quite significant. Then, we would expend much more computational effort to determine their value. Nowadays, digital filtering is often employed, if signal frequencies are tractable. Caps can set the operational fq of relaxation oscillators, here Teflon or polystyrene dielectrics might be used, for stability. If higher stability is required, we will be tempted to use crystals, instead. We can make a resonant tank to construct a sine-wave oscillator for an old-fashioned rf deck. This is where we find those macroscopic uber-cool variable caps with their delicate fins. Russians make the best of these, as well. These aren’t very stable, so we are going to shunt in a fixed cap, limiting our tuning range :(. Also in shunt with our tank ckt will be an L. Theoretically, Ls can be used in perpendicular arrangements from Cs in circuits, but Ls are more expensive than Cs, less stable and potentially more susceptible to interference. So we go to some lengths to avoid them. Caps can be used in very small switching power supplies, the big fellows use Ls for energy storage. Annular caps called ‘feed-thru caps’ are used to supply dc power to rf circuits through the chassis, which is called a ‘Faraday cage.’ Finally, if we can make our cap alter its geometry in accordance with some physical parameter, we can use it as a transducer if it is the storage element in an oscillator, probably we will use a counter to monitor the oscillator at a later stage. An air-dielectric cap might measure humidity, weather we want it to or not!
This isn’t a fully comprehensive list. If you want to know more about this, READ A BOOK! About electronics ;)
Xc = 1 (2 * pi * F * C)
I meant to correct #24 thusly:
Capacitive Reactance = 1 / (2 * pi * F * C)
not 1/ (2 * pi * F)
It’s too late at night for this.
Another mantra for the beginner is: Capacitors integrate potential by sinking/sourcing current. Inductors integrate current by sinking/sourcing potential. Equivalent statements could be made by interchanging ‘current’ w/’potential’ and ‘integrating’ w/differentiating.
//Detroit Ross left out a ‘/’ before the ‘(.’
//Also this analysis leaves out the important, but elusive, concept of phase. This can be represented by using complex numbers, rendering a term for ‘impedance’ the real term resistive, the imaginary term reactive. Capacitive reactance has an associated – signum, inductive reactance a + signum.
‘Phase’ refers to sinuate signals. For other signals we need a little Fourier (properly: Euler) analysis.
XL = 2 * pi * F * L
Fc = 1 / ( 2 * pi * sqrt( L * C) )
Q = XL / R
E / I = futile
(*Homer Simpson hoarse whisper”: “I think that last one was a joke”)
Our high school electronics teacher explained it thus, “The positive terminal is where the blue fairies fly in, the negative terminal is where the pink fairies fly out, and in the middle is the Magical Unicorn Breeding Grounds!” not long before he was taken away.
A compressible fluid is not a bad analog for electricity.
A high-order flaw in capacitors is ‘dielectric soakage.’ Oil-paper caps, once used as power-supply capacitors, would charge back up on their own after being discharged, yikes! Nowadays, this is a matter for concern with capacitors used in long-term integrations.
Energy stored in capacitor: V^2 * C /2.
Transistors get all the love, but caps are where it’s at, man. They kick coils ass! Woo! Go caps! And don’t even get me started about resistors.
#8 posted by Jerril
“Would it be more correct to say that potential energy is stored?”
Yes, or just: “energy is stored”.
#9 posted by ROSSINDETROIT
“There is disagreement over whether a cap with a net DC charge is better at passing AC than one with an un-biased dielectric. ”
Definitely, if we’re talking electrolytics, (otherwise the would be mispolarized half of the cycle) but they have no place in audiophile signal paths anyway. Noisy, nasty brutes.
But for film caps, I don’t see how it could make a positive difference – soakage would lead to distortion as half the wave would be opposed by the permanent charge on the now electret dielectric … I suppose you could make a circuit where that would be an advantage, but not normally. Maybe it’s a good-sounding distortion in some cases. It could be that some caps need to age into the right degree of soakage to get to the way the circuit was designed. More likely in most circuits it’s another audiophile superstition. Or maybe I don’t understand what the situation being discussed is.
#14 posted by Roy Trumbull
“There were some dual polarized electrolytics but I never came across a description of how they functioned.”
Two ordinary electrolytics with their negative (or +) leads tied together and not connected to anything else.
#15 posted by Jason Rizos
“Dang, I wish he would have told us what they are used for, practically, in electronics.”
Usually for AC stuff, sometimes for energy storage. For AC they are used together with one or more resistors to make filters that pass the high frequencies or the low frequencies of the signal. A capacitor by itself will prevent DC from passing through it. Larger capacitances can pass lower frequencies. This is a series, “high-pass” filter.
A capacitor in parallel with a battery or power supply will tend to prevent variations in the voltage as loads are applied. This is a “low-pass” filter. Large electrolytic capacitors are found in power supplies to smooth out the ripple that remains when converting AC to DC. When using digital chips there will be an electrolytic (often tantalum) across the power supply leads as close as possible to the chip to provide a squirt of current and prevent the power supply voltage from fluctuating under the rapidly varying power demands of the chip.
When used (usually with op-amps) in analog circuits, a capacitor can be used to accumulate a voltage across it as a current is fed through it. If the current represents some quantity, the voltage is the integral of that quantity. Worked the other way around, the capacitor calculates the derivative of the voltage signal. Capacitors used for such applications are usually high-quality plastic film capacitors. The key thing to remember about capacitors is that you have to run a current through them before a voltage accumulates across them. (With inductors it’s the other way around – voltage leads current.)
Capacitors are cheaper and generally behave better than inductors so they get used more. Ideal inductances in series are just like capacitances in parallel, and vice-versa, but real inductors and capacitors are quite a different thing.
#20 posted by ROSSINDETROIT
“I recommend thinking of electricity in terms of electrons. This is made difficult by the convention of referring in many schematics to current flow as ‘hole flow’ which is opposite to the flow of electrons. My formal education in electronics started 34 years ago and still nobody has explained to me where that nonsense came from.”
Benjamin Franklin decided that resinous (-) and vitreous (+) electricity (so named because of experimenters rubbing amber [gr. elektrum] and glass with silk and getting opposite charges) ought to have better names. He chose “positive” and “negative” and decided which was which by a coin toss.
I have trouble with conventional current direction, too, especially when combined with transistors, but it isn’t as illogical as it seems at first.
From another Bill Beatty article, “Which Way Does the Electricity Really Flow?”:
“Actually, in some situations electric currents can really be a flow of positive particles. In other situations the flows are negative particles. And sometimes they’re both positive and negative flowing at once, but in opposite directions. The true direction of the flowing particles depends on the type of conductor.”
and also from “Electricity Misconceptions” his “Ben Franklin Should Have Said Electrons Are Positive? Wrong” in which he refutes the following:
1. All electric currents are flows of electrons. Wrong.
2. “Electricity” is made of electrons, not protons. Nope.
3. Electrons are a kind of energy particle. Wrong.
4. “Electricity” carries zero mass because electrons have little mass. No.
5. Positive charge is really just a loss of electrons. Wrong.
6. Positive charge cannot flow. Totally wrong.
7. To create “static” charge, we move the electrons. Not always.
Regarding capacitors with a net charge. Signal coupling capacitors between DC biased amplifying stages work with large potential differentials. Example: The cap coupling signal from the plate of a driver tube (+200V) to the grid of an output tube (-50V) in a tube amp. 250VDC potential exists across the cap and a small AC signal flows through it. Capacitors are ideally suited to pass AC while blocking DC. It’s commonly called RC Coupling. They are not the only way to get this done but 99.9% of the time that’s how it’s handled in AC coupled amplifiers. Anyone interested in reading an extended argument on using charged capacitors in non-charged applications is invited to join this epic AudioKarma discussion thread at around post#4570 .
I’m reading Bill Beatty’s article
How Do Transistors Work?
NO, HOW DO THEY REALLY WORK?
It’s a contrarian explanation based on the physics of charge motion rather than the Beta model of current math but I suspect he’s on to something there.
If you want to know more about this, READ A BOOK! About electronics ;)
The Art of Electronics, perhaps?
Thanks, ZUZU. I need that one bad. My copy is on order.
And here’s a page of downloadable vacuum tube electronics books for the curious.
I’m an electrical engineer and I approve of this video.
Very good video. I made your capacitor style. If I feel like using direct battery power to power it up, which end should I use? Positive or minus? Thanks!
Mail (will not be published) (required)
Submit a tip
The rules you agree to by using this website.
Who will be eaten first?
Jason Weisberger, Publisher
Ken Snider, Sysadmin