caps in DC?

Thread Starter

maggie

Joined Aug 15, 2011
13
I'm teaching myself electronics. Numerous sources allude to the fact that capacitors block DC, after building up a charge equal to source. After that, i would presume no more DC current is flowing, so, as there would be no point in blocking DC in a DC circuit, I don't get what use capacitors are, in a dc circuit at all. Would someone be so kind as to explain?
 

SgtWookie

Joined Jul 17, 2007
22,230
In a nutshell: capacitors block DC, and pass the effects of AC.

Our E-book begins covering capacitors in Volume I, Chapter 13:
http://www.allaboutcircuits.com/vol_1/chpt_13/index.html

Capacitors are widely used in DC circuits; there is generally at least one "bypass" capacitor across the power/ground pins of each IC (integrated circuit) used to keep the supply voltage relatively constant. Without the bypass "caps", the inductance of the wiring between the circuit and the power supply would cause large fluctuations in the voltage as current flow in the circuit is changed.

They are also used in RC, LC, and RLC circuits for timing and filtering purposes.
 

Adjuster

Joined Dec 26, 2010
2,148
Capacitors are also used to store energy: this occurs in power supply bypass, and smoothing applications. In particular, in the most basic type of DC power supply obtained from a rectified AC input, reservoir capacitors are used to "fill in the gaps" of the pulsating output obtained from the rectifier.

In addition, larger quantities of energy are stored in capacitors for such purposes as photographic flash, laser power supplies, and other uses where it is useful to store and release electrical energy, possibly at higher short-term rates than batteries can easily deal with.

Special energy storage capacitors called "supercapacitors" have been developed as a sort of "half-way house" between ordinary capacitors and batteries. These are not so far a general substitute for conventional batteries, but do have specialised applications for short-term energy storage in power systems, while smaller versions can replace batteries in certain low-drain applications, such as memory backup.

Edit: But now, it seems, they speak of ultra capacitors. I am getting old...
 
Last edited:

tgotwalt1158

Joined Feb 28, 2011
110
I think a good example would be RC Oscillators, in which filter portion is a pure DC circuit built with resistors and capacitors! Am I right? Or confused, please comment.
 

Adjuster

Joined Dec 26, 2010
2,148
You could say that any to-and-fro movement of charge in an RC oscillator constitutes a sort of AC. The OP may not be aware of this sort of application though, so perhaps it's as well to mention it.
 

SgtWookie

Joined Jul 17, 2007
22,230
I think a good example would be RC Oscillators, in which filter portion is a pure DC circuit built with resistors and capacitors! Am I right? Or confused, please comment.
I made reference to those in my reply. ;)

You could say that any to-and-fro movement of charge in an RC oscillator constitutes a sort of AC. The OP may not be aware of this sort of application though, so perhaps it's as well to mention it.
Well, the "sort of" is actually rippled DC, which is different from AC. Rippled DC would be a DC level with some kind of signal riding on the DC level; where AC when averaged, would show ~0v.
 

Adjuster

Joined Dec 26, 2010
2,148
I made reference to those in my reply. ;)

Well, the "sort of" is actually rippled DC, which is different from AC. Rippled DC would be a DC level with some kind of signal riding on the DC level; where AC when averaged, would show ~0v.
Originally Posted by Adjuster
You could say that any to-and-fro movement of charge in an RC oscillator constitutes a sort of AC. The OP may not be aware of this sort of application though, so perhaps it's as well to mention it.
Note my original wording: I said you could say that the to-and-fro movement of charge in an RC oscillator constitutes a sort of AC. I made no reference to any standing DC charge, and in fact for some RC oscillators the average charge is negligible. (This is true for some op-amp triangle wave oscillators, as well as some phase-shift and Wien bridge oscillators - and in the latter cases the waveform can be very nearly sinusoidal.)

While such semantics may not be much help to the OP, note that in general a rippled DC voltage with a regularly repeating waveform could be analysed as consisting of a constant term (pure DC), plus a series of harmonically related sinewave (AC) terms.

A similar situation exists for a sinewave signal superimposed on a DC bias at the output of an amplifier. You could describe its waveform a rippled DC if you so chose, but there is certainly an AC component present, and in fact a blocking capacitor may be used to separate the AC fron the DC.
 

CraigHB

Joined Aug 12, 2011
127
Well, the "sort of" is actually rippled DC, which is different from AC. Rippled DC would be a DC level with some kind of signal riding on the DC level; where AC when averaged, would show ~0v.
That doesn't make much sense sense to me. You're saying, for example, that a sine wave of magnitude 2 with a an offset of 1 would not be AC since it does not average zero. Would you still call that "rippled DC"? There's no DC component other than an offset which is only a matter of ground reference and that can be arbitrary. I'm not trying to be contradictory, I'm just not understanding the idea.
 

colinb

Joined Jun 15, 2011
351
I believe that “ac” strictly speaking would mean the current would have to reverse direction -- the current is alternating its direction of flow. For current to flow in the reverse direction, the voltage would have to be negative with respect to the point of reference. If it's just a variation in current (or voltage?) without changing polarity, then it must be modulated or pulsed dc.

Remember that voltage is all relative anyway: it is the electric potential difference between two points. So when talking about whether voltage is negative or positive you need to define both reference points. “System ground” is commonly used as a reference and this is correct with it is the return path for current, but not all currents in a system will share the same ground.

I am likely wrong, but these are my reasonings.
 

CraigHB

Joined Aug 12, 2011
127
That would make more sense to me in terms of current direction. Like you said, voltage is an arbitrary value based on any two points. After all, AC is "alternating current" not "alternating voltage". Even so, we often consider AC response to DC systems so isn't that just an AC imposed on DC? The AC componet is still considered AC even when superimposed on the DC component.
 
Top