# AC and DC definitions

Discussion in 'General Electronics Chat' started by gootee, Jan 22, 2013.

1. ### gootee Thread Starter Senior Member

Apr 24, 2007
447
51
Actually, I wasn't confused. I just disagree. I could be wrong, of course.

I always thought that AC meant that the current waveform had to cross through zero, so it would change (i.e. alternate) directions, making it "Alternating Current".

In other words, DC is anything with only one polarity and AC requires both polarities.

If a voltage or current was constant, or fixed, we called it "constant DC", or "time-invariant", or "fixed".

I think it would be difficult to define and use AC and DC, your way. What kind of "fluctuations" would be counted, in order to call something "AC"?

When my battery voltage changes (fluctuates) from the fully-charged voltage to the discharged voltage, or when it merely dips momentarily under a heavy transient load, is it an AC battery voltage? What if I make the battery's voltage dip repeatedly, periodically, even sinusoidally? AC battery voltage. Hmmm.

What about bridge rectifier output voltage? I think it should be called "fluctuating DC voltage", or "time-varying DC voltage". I think that since it does not alternate polarities, it is "DC".

And then there would be cases like in an audio power amplifier, where the power rail voltages are almost constant DC (as constant as we want, at least), but the currents, which make the output signal, can be very dynamically time-varying, even sinusoidal, but still always only go in one direction. So then you would say that we have a DC voltage and its AC current in the same conductor?

Here it is on Wikipedia:

http://en.wikipedia.org/wiki/Direct_current

It's not entirely clear-cut. They do say:

"Direct current (DC) is the unidirectional flow of electric charge."

and

"Although DC stands for "direct current", DC often refers to "constant polarity". Under this definition, DC voltages can vary in time, as seen in the raw output of a rectifier or the fluctuating voice signal on a telephone line."

but they also have:

"The DC solution of an electric circuit is the solution where all voltages and currents are constant." and some even stricter statements about DC being "constant".

In the end, I don't think you can call something "AC" unless its current reverses direction.

I would probably prefer to see terms like "AC component" and "DC component" if there is a time-varying waveform that is of only one polarity, e.g. a sine with a large-enough DC offset.

I do tend to think of any time-varying waveform as AC, even if it is technically DC. But I almost never call anything "AC" or "DC", except for primary power sources.

I guess it's all a bit too academic, when time-varying signals are involved. "AC" is probably more of just a power-grid type of term. We eventually just looked at everything as general functions of time or frequency and didn't usually care what it was called.

But, of course, being just a wee bit overly-pedantic, I don't mind dessecting and debating almost anything.

Regards,

Tom

2. ### Georacer Moderator

Nov 25, 2009
5,173
1,284
I moved this post from here (http://forum.allaboutcircuits.com/showthread.php?p=566093#post566562), since the OP felt like discussing the matter a bit.
I 'll state the way I like to interpret things and how I use it in my routine and the rest of the AAC old timers are invited to give their view too.

Firstly, I want to detach the notion of current from the labels AC and DC. I believe the C is there purely for historical reasons. When talking about AC and DC I imagine quantities that are either time-stable (in our inspection time frame) or varying. These notions aren't exclusive to each other.

Fields where I usually use these labels are in voltage sources, speaker signals and image spectra.

Having the Fourier analysis heavily forced into my brain, thanks to 6 years in uni, I always expect to find a DC component and AC components of varying frequencies in any given signal.
The output of a rectifier bridge has Vdc mean value and Vac ripple imposed on it.
Speaker signals have all of their information on AC waveforms and 0 DC components, since they are usually AC coupled.
Image frames have a DC exposure component and AC harmonics which define the image details.

Since I mentioned Fourier analysis, the time frame over which you inspect the significant (and that depends on your application) changes of your signal matters for the interpretation.
The battery voltage during the time a circuit is powered, is constant in the eyes of the viewer. In that case I would say that the battery voltage is a signal with only a DC component.
If your application includes a switching load that loads the battery heavily and beyond neglect, instead of thinking that the battery is a varying voltage source, I would model it with a series internal resistance and keep the ideal DC voltage source.

3. ### t06afre AAC Fanatic!

May 11, 2009
5,939
1,227
From what I remember the wiki page is quite correct. If the current is not steady. Like say the output directly from a rectifier. You can use the term pulsating DC

Last edited: Jan 22, 2013
4. ### tshuck Well-Known Member

Oct 18, 2012
3,527
679
What if it reverses direction about a specific level? Does that level have to be with respect to 0(whether it is current or voltage)? What I mean to say is that is that if a current would change direction about 1A, moving sinusoidally from 0.5A to 1.5A, it would be alternating about 1A, right? It would be one polarity with respect to 0A, so, you could say it is not changing polarity(with respect to 0A), but the current is fluctuating periodically with time, so it must also be AC, just centered about a different point.
That is the correct terminology. I agree, I wish more people talked like this, It would clarify a lot of problems.

5. ### crutschow Expert

Mar 14, 2008
20,216
5,715
I agree that the "Current" in AC and DC is an historical reference and does not strictly refer to current. For example an AC source can be that without any current (or you can call it an AC voltage source).

My two cents: If it has any measurable Fourier component then it has an AC component. Thus a rectified signal has an AC component (or ripple) and an average DC component. There is no requirement that the "current" component go through zero to be call AC.

Last edited: Jan 22, 2013
tshuck likes this.
6. ### GopherT AAC Fanatic!

Nov 23, 2012
8,025
6,786
I like to use the definition my FLUKE 87 meter uses. If I have a 1 volt AC voltage overlaying a 10 v DC offset, the DC voltage selector will read 10 v and the AC volt selector will read 1.

7. ### MrChips Moderator

Oct 2, 2009
17,296
5,328
You are all correct. The terms AC and DC can mean everything you have all described... it all depends on the context that they are used. There is no hard and fast definition, so don't get too obsessed with a definition.

AC can mean a reversal of current direction as well as a zero crossing voltage signal.

From a frequency analysis perspective, it can also refer to any time-variant component of a signal whereas DC refers to the 0-Hz frequency component.

Both concepts of AC and DC are acceptable. Take your pick, depending on context.

8. ### WBahn Moderator

Mar 31, 2012
23,156
6,972
I second MrChips. The discussion comes down to semantics and semantics, more often then not, are context-dependent. In a detached conversation about what AC and DC mean, people want to use a single definition but what different people have in mind are different contexts and so they can't agree on a meaning. But in nearly all practical situations, if those same people were discussing AC and DC in the context of a specific situation they would have no problem communicating and would use the meaning that is appropriate to that context -- and then use a different meaning when using those exact same terms five minutes later when discussing a different situation.