# Why use capacitors?

Why do you need to store the voltage for some time in a capacitor? I've always assumed circuits to work when you power it on and stop when you power it off.

Why can't the whole circuit be drawn capacitor free? If it's meant for storage why not just use a flip-flop?

• Two words: time travel. Commented Oct 4, 2010 at 4:52
• If you can lay your hands on a couple, let me know.
– user1307
Commented Oct 4, 2010 at 5:50
• Ya sure. I ordered an Arduino Board. they are delaying untill they get Uno Board in their Stock. I am waiting. Commented Oct 4, 2010 at 5:59
• Hope we were of help, I would suggest that you change your username to an actual name instead of an auto-generated one. Commented Oct 4, 2010 at 14:21
• Capacitors on power supply lines as you describe are called "bypass capacitors" or "decoupling capacitors". en.wikipedia.org/wiki/Decoupling_capacitor There are many other uses of capacitors in circuits. Commented Oct 4, 2010 at 14:28

If all you wanted to build was digital circuitry, and your voltage sources really held constant voltage no matter how much current was drawn from them, and nothing produced electrical noise, you wouldn't need capacitors.

But voltage sources sag when you draw current from them. Motor brushes (and lots of other components) produce horrendous voltage spikes that you want to filter out of your digital circuitry. Some people also deal with analog circuitry, where voltage and current signals vary continuously across a wide range. For that kind of time-varying circuitry, capacitors are needed.

• I was almost done when yours was posted, and since I took a lightly different direction I still posted it. We said the same thing in different ways, +1 for you. Commented Oct 3, 2010 at 19:56
• And for you sir! Commented Oct 3, 2010 at 20:09
• Figuring out parasitic components will explain much of the effects.
– XTL
Commented Oct 4, 2010 at 23:11

Digital circuits can be especially bad, but in general you are attempting to make the power rail stay a source of DC power. most circuits when they suddenly draw power from the power rail would not be too happy if the power rail reacted by dipping.

As you go to higher speed, inductance causes a bigger problem than resistance. The capacitor acts as a very close source of power. You pull your high speed power from the capacitor and the power source slowly charges the capacitor.

When done properly, everything works to spec. When making a commercial product and done improperly you get a product that has very odd bugs, normally tied to high load as the voltage really sags(sags= goes below what it needs to be). In the worst case high speed signals traverse your power lines and the FCC does not approve your product as it is radiating high frequency energy.

• In your second paragraph, do you mean you want the capacitor physically (geometrically) close to the high frequency component in order for it to get power? What happens if its not there? Noise or sag? (Context: I want to use some radio in next month's hobbyist project, and I'm trying to get up to speed on what is involved. I'm hoping to use a microcontroller with built in radio to do some low power transmission, maybe RFID style, basically just a few kilobytes every hour. Do I need to know about this use of capacitors?) Commented Oct 4, 2010 at 4:25
• i mean electrically close. This means that the ruler distance between parts may be short, but if the wire is a bent underneath 6 inch wire you have a problem. This will cause both noise and sag. the chip with a capacitor at a distance will have sag issues(possibly minor, look at the datasheet) but you will cause noise on the wire between them with it acting as an antenna. Commented Oct 4, 2010 at 5:22
• Cool thanks! I'm still gathering enough information to ask a reasonable question. Looking through DigiKey and Mouser's catalogs and reading \$5 chip's datasheets is like reading science-fiction. I'm trying to get up to speed by reading the textbooks from our undergrad EE courses, but I haven't figured out how to get the lab experience yet. Commented Oct 4, 2010 at 15:25

Capacitors are also widely used in oscillator, filter and timing circuits, because their charging rate and discharging rate can be accurately calculated.

In an RC circuit, the value of the time constant (in seconds) is equal to the product of the circuit resistance (in ohms) and the circuit capacitance (in farads), i.e. R × C. It is the time required to charge the capacitor, through the resistor, to 63.2% of full charge; or to discharge it to 36.8% of its initial voltage. These odd looking percentages are derived from the mathematical constant e (2.71828, the base for natural logarithms), specifically 1 − 1/e and 1/e respectively.

Oscillator and timing circuits are commonly used in digital systems to provide frequency generators and timing. Oscillators and filters are typically found in analog circuits, i.e. audio or radio-frequency (RF).

• This is a major major reason to use caps, but from a new guy to circuits whom does digital I was betting they were all decoupling caps. Commented Oct 3, 2010 at 23:52
• Thst's why I started with "Capacitors are also widely used..." If he is just starting out in electronics, he is going to run into them again. Commented Oct 4, 2010 at 0:46
• I did not mean to imply you were wrong, I was just giving a comment, something for others to read, and to note that I more than agree with your answer. Commented Oct 4, 2010 at 5:23

One of the most popular applications of capacitors in industrial electrical engineering is to provide power factor correction. The capacitors store energy and release it every cycle on an AC power distribution network to compensate for the fact that highly inductive loads such as electric motors draw a current which 'lags' behind the applied voltage. This results in poor power factor on the electrical distribution network, which typically means that network assets can not be utilized to their apparent power rating.

By using power factor correction, which for inductive loads means switching capacitors into the supply network, the power factor can be increased close to unity which means network assets such as large transformers don't need to be unnecessarily over sized.

Also, most electrical supply authorities will penalize users who have very poor power factor, since they usually bear the additional cost of over sized & under-utilized distribution assets. There is therefore a financial incentive for large industrial users to install power factor correction equipment.

Capacitors are also used to filter out the ripple when rectifying AC power to DC (eg: in the input stage of a variable-speed drive or inverter circuit).

Also, capacitors are used to 'amplify' DC power supplies (eg: to convert a 5VDC power supply to output 9VDC). These are called 'chopper' circuits.

• Chopper circuits? I know them as pump charge circuits. Commented Oct 21, 2012 at 14:08
• Pump charge sounds reasonable but I know them as voltage multiplier circuits. Commented Dec 28, 2013 at 14:09

Why can't the whole circuit be drawn capacitor free?

Circuits are occasionally drawn without capacitors, as it is implicit that they will be included on every logic power pin. Obviously if using an EDA tool, they must be on the schematic somewhere (usually disembodied in some corner), but it is implied that there will be at least one on each pin (multiple caps can cover a wider range of frequencies), and as close as possible.

For prototypes--especially for prototypes--bypass capacitors are even more important. There will often be much more inductance in the ball of wires than normal. Even if your switch frequency is low, the spectral content of edges can be extremely high.

Hello user1424 You seem to be asking a lot of questions about many things electronic. May I recommend you find a good book such as "The Art of Electronics" by Horowitz and Hill, and have a good read through that.

• the art of electronics is not the easiest read. Commented Oct 4, 2010 at 14:20
• But it's oh so full of stuff <3
– XTL
Commented Oct 4, 2010 at 23:07
• I think it is amazing, just warning. Commented Oct 5, 2010 at 0:50

See this question:
What is a decoupling capacitor and how do I know if I need one?

Decoupling capacitors serve several purposes. First they're a safeguard against variations in the power supply. It the capacitor wasn't there a dip could reset the whole circuit. Likewise, some power-hungry parts of the circuit may switch on and off during operation. Switching on also creates a dip; much current suddenly needed in one place means that it's no longer available elsewhere. The capacitor is a buffer storage that makes sure there's enough current for all components at these switching moments.

A good example is capacitive touch screens (e.g. touch screen in iPhone).

Capacitive touch screens use a layer of capacitive material to hold an electrical charge. Touching the surface of the screen results in a distortion of the screen's electrostatic field creating a voltage drop, which is measurable as a change in capacitance. This exact location of the voltage drop is picked up by a controller and transmitted to the processor.