Alternating Current vs. Direct Current

Thread Starter

helloeveryone

Joined Apr 8, 2011
64
Hi, I was wondering if someone could comment this:

https://qph.ec.quoracdn.net/main-qimg-0ab10d9f076e71356f8ba34ffdd189ed-c?convert_to_webp=true

which I found the web. I am trying to understand the difference between an AC signal and a DC signal. I understand that alternating current/voltage alternates between positive and negative. If the signal stays the whole time either positive or negative then it is a DC signal, correct? So, it can fluctuate(does not have to be linear) but as long is it stays positive or negative it is DC. If a signal keeps going positive, negative, positive, negative, then it is AC. The part I don't understand is the fifth picture where it says +ive A > -ive A and the very last picture where it says -ive > +ive. One can clearly see that the signal alternates between positive and negative in both cases. So I would think it is automatically AC? But in the picture the person has drawn a red line indicating that the average is positive in the first case and negative in the second case and designated both as DC signals. Is looking at the average some kind of special case or must we always look at the average amplitude to be certain whether or not a signal is indeed AC? Here's an example I drew of an exponential signal. The signal amplitude reaches positive 5 and goes down to negative 4. I would assume this an AC signal since it alternates, positive, negative. But the average is positive, so is it actually a DC signal?
exponential.JPG
 
Last edited by a moderator:

wayneh

Joined Sep 9, 2010
17,496
And conversely, a noisy DC source can be said to include an "AC component". I have a beef with that, since the current direction never changes and thus shouldn't be called "alternating".
 

Thread Starter

helloeveryone

Joined Apr 8, 2011
64
You have it right. Any time the long term average is not zero there is a DC component.
What would be the correct terminology for any signal that alternates between positive and negative but has an average other than zero, as in my example? "An AC signal with a DC component"?
 

wayneh

Joined Sep 9, 2010
17,496
What would be the correct terminology for any signal that alternates between positive and negative but has an average other than zero, as in my example? "An AC signal with a DC component"?
I think that's the best. Everyone knows what that means. It's correct to call it AC, but that omits a key feature. If the DC offset exceeds the AC component (so that the polarity never reverses), then you have DC with an AC component.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
Often the ripple output from a bridge rectifier is termed 'AC ripple' whereas it is fluctuating DC, (WRT the normal reference point) for AC, the current would have to reverse.
Max.
 

MrChips

Joined Oct 2, 2009
30,714
Every signal can be analyzed from two perspectives, in the time domain and in the frequency domain.
So far in this thread the signal has been described in the time domain.
In the frequency domain, every signal may have frequencies from 0Hz to infinite Hz.
DC is the component at 0Hz.
Everything else above 0Hz can be considered AC, a composite of pure sine waves at various frequencies and phase angles.
 

MrAl

Joined Jun 17, 2014
11,389
What would be the correct terminology for any signal that alternates between positive and negative but has an average other than zero, as in my example? "An AC signal with a DC component"?
Hello helloeveryone :)

Interesting screen name :)
All seriousness aside, if you were Bono you might choose "hellohellohelloeveryone" :)

Ok, on the more serious side we have the Fourier decomposition.
The most accurate representation of a mixed signal like something with AC and DC is the breakdown into the AC components and one DC component, and that is found by doing a Fourier analysis. This involves one or two integrals that are used to compute the AC components and the one DC component. We end up, using the series form, with several AC components and one DC component.
Each Ac component has an associated frequency like 1Hz, 10Hz, 23Hz, etc.

The interesting side line here though is that in theory any signal can be broken down into it's AC components and one DC component. That means even a square wave can be broken down into components. This is often the example given, so i'll repeat that here.
For a square wave of 100Hz that has peaks that go plus and minus 1 volt we have first the fundamental which is 100Hz, then we have the third harmonic which is at 300Hz but is reduced by a factor of 1/3, and then we have the fifth harmonic which is 500Hz but is reduced by a factor of 1/5 from the fundamental, and from there we have the same thing with all the odd harmonics, 7,9,11,13,15, etc.
The amplitude of each one is 1/n times the fundamental, so if the fundamental amplitude was 1 then the 99th harmonic is 1/99 times as high, so for a 1v fundamental the 99th harmonic would be 1/99 volts which is nearly 0.01 volts. So you see the amplitude is much less than the fundamental, and the frequency would be 9900Hz if the fundamental was 100Hz.

Also interesting is if all these harmonics are added back together with the proper amplitudes and phases, they add back up to the same square wave. It takes a lot of harmonics to see that though so dont expect to see a perfect square wave with only say 21 harmonics. In theory, that takes an infinite number of harmonics for the square wave example.

There are other ways of describing some of the more well known waveforms however, such as the sawtooth wave, and the output of a bridge rectifier without any filtering is often referred to as "pulsating DC" because the entire waveform is above zero but varies up and down from zero to peak. The phrase "pulsating DC" however should probably only be used to describe that wave or the half wave rectified sine, but it depends partly on the context. It seems simpler to call it that then to have to refer to all the Fourier components.

Sine waves can be offset by any amount. For a 1v peak sine wave it goes to +1v and down to -1v, but if we add an offset of +1v then it goes from a max of +2v down to 0v so we get the same wave but moved upward. If we add more offset we can make it vary from +20 to +18, never even getting close to zero. If we add negative offset, we can get it down to -20 to -18v, same idea just different DC offset. In case 1 the offset would be +19v and in case two it would be -19v, so it's easy to see what the DC offset is.

If you really want to know the most you can here then you should look at the Fourier Series which is all over the web so can be found easily. If you have trouble with it you can always ask that here too :)

Good luck with it.
 

WBahn

Joined Mar 31, 2012
29,979
What the terms DC and, particularly, AC mean depends on the context of the discussion. Note that we almost always talk about the voltage waveform and not the current waveforms, yet as soon as we get reactive components involved we can have positive voltages with negative currents. The terms themselves, as they originated, were quite limited in scope and were not meant to be applicable to any and every waveform; they were generalized later but the names were kept the same. In the most general sense, a DC signal is a signal that never changes and every thing else is an AC signal. As already stated, this is best viewed in the frequency domain in which the 0 Hz signal is the DC component and all of the other signals are the AC components. This is completely irrespective of whether the signal ever changes polarity or not.
 

wayneh

Joined Sep 9, 2010
17,496
This is completely irrespective of whether the signal ever changes polarity or not.
And there's my beef with an otherwise quite logical explanation. I don't like using the term "alternating current" when it isn't alternating direction.

In practical usage, the terms are imprecise and depend on context. A wall wart with a rectifier makes DC even though it may have an enormous ripple. An audio signal is AC even though it might be just a tiny ripple riding on a big DC offset. We call them DC and AC because of the application they are used in.
 

WBahn

Joined Mar 31, 2012
29,979
And there's my beef with an otherwise quite logical explanation. I don't like using the term "alternating current" when it isn't alternating direction.

In practical usage, the terms are imprecise and depend on context. A wall wart with a rectifier makes DC even though it may have an enormous ripple. An audio signal is AC even though it might be just a tiny ripple riding on a big DC offset. We call them DC and AC because of the application they are used in.
Yep. They are imprecise and context dependent. That simply isn't going to change. There's a few terms that fall into this category (though trying to think of an example off the top of my head is failing me) and we just have to deal with it, which sadly does open the door to miscommunications that can and have had consequences. It's part of the legacy of how our knowledge has evolved and how concepts have expanded to more general cases. Humans tend to cling to familiar terms and let their meanings evolve; sometimes this means that the terms become mere labels in which the words that make up the label are no longer strictly applicable, they are just words in a label.
 
Top