14

My teacher says that this signal is DC, but if the polarity varies from 0V to -8V , shouldn't it be AC?

He says that

DC + AC = DC

enter image description here

JRE
  • 67,678
  • 8
  • 104
  • 179
JEAN LEONARDO
  • 343
  • 2
  • 8
  • 37
    I doubt you'll get much agreement on this. If the waveform goes from +10 V to -8 V I'd call it AC with a +1 V DC offset. – Peter Bennett Mar 16 '21 at 03:49
  • 19
    Note that some people call any signal such a sine-wave that has been completely offset to always be of one polarity (positive for example), DC. It's wrong and it's annoying and confusing as all hell. Your teacher seems to start veering into this category. Unipolar and bipolar are other descriptive terms that can be used as well. – DKNguyen Mar 16 '21 at 04:28
  • 2
    ask your teacher what he calls the voltage at a power outlet ... AC or DC? .... then ask what he calls the same voltage at the same power outlet if the reference point is at -500 V ... yes, negative 500 volts – jsotola Mar 16 '21 at 05:29
  • 1
    Are you misreading this? I see the values on the vertical axis as "10V" and "8V". In which case you've got 9V DC with a +/-1V AC ripple. In which case it's valid to consider it mainly DC, especially if you then put it through a filter to kill the ripple. – Graham Mar 16 '21 at 13:56
  • 8
    @jsotola Your reasoning is valid, but move your reference point to -10 kV and I’d call it DC with a tiny AC ripple superimposed on it. – winny Mar 16 '21 at 15:28
  • 1
    @PeterBennett The question text says it goes from 0V to -8V, and I think the chart is just messy. So it's -4V with a 4V ripple, so the current only flows in one direction – coagmano Mar 16 '21 at 23:09
  • 1
    To my mind, AC means a current alternating in _direction_ relative to ground. If it crosses 0V, it's AC. – Ed Randall Mar 17 '21 at 07:51
  • 4
    @jean-leonardo please clarify the question (is it +10V or +0V to -8V) and label the peak, trough and 0V points on the axes more legibly. At what voltage level is the x-axis intended to be? – Ed Randall Mar 17 '21 at 07:56
  • @jsotola a -500V offset is largely irrelevant. A -500A offset, on the other hand, would easily change the signal's type from AC to DC. – Dmitry Grigoryev Mar 17 '21 at 09:57
  • 3
    "AC or DC" is a question of semantics - meanings of words. It's AC if you think that's what AC is, and it's DC if you think that's what DC is. We can describe the signal more accurately, "a 18V p-p signal superimposed on a 1V DC offset" and then the problem disappears entirely. – user253751 Mar 17 '21 at 13:14
  • 3
    It's a mistake to care about the answer to this question. Classifying this signal as AC or DC has no valid practical implications. – Matt Timmermans Mar 17 '21 at 17:28
  • It really depends on how much you have of each component and what is important. A 5V DC with 500mV ripple is still DC... A 500mV sine-wave (eg. audio-signal from a generator) with a +5V offset (ie. a 5V DC component) is still AC... – Baard Kopperud Mar 17 '21 at 19:50
  • 1
    If water is added to milk, it's neither milk nor water. It's either milk on water, or water on milk; depending on the quantity of milk/water on the mixture :-D – Meenie Leis Mar 18 '21 at 09:46

9 Answers9

47

DC + AC = DC

Plainly put: No.

It's AC with a DC offset, if you will.

This is less about whether the sign changes (that sign is just relative to some arbitrarily defined reference), but about whether the energy transport happens mainly in the DC or AC component.

That's all that makes the difference between "AC with DC offset" and "DC with AC ripple".

mmmm
  • 1,759
  • 4
  • 14
  • " (that sign is just relative to some arbitrarily defined reference)" I thought the sign was the whole point, as in it is 'DC' if the current only flows in one direction. – Jason Goemaat Mar 16 '21 at 15:32
  • 1
    @JasonGoemaat, how about e.g. a transistor amplifier where the input signal is coupled on top of a DC bias so that it stays positive all the time? E.g. the example amplifier from the Falstad simulator with a scope showing the transistor base voltage: https://tinyurl.com/yzmsghuh – ilkkachu Mar 16 '21 at 21:15
  • 1
    I would say the parts of the circuit that had current (the yellow dots) always flowing in one direction would be DC and the parts where they changed directions would be AC because the Current Alternates. – Jason Goemaat Mar 16 '21 at 21:25
  • 1
    Hmm, what about a 14.1VAC with 10VDC offset then? Isn't it more about the intent, which we can't tell without knowing what we're measuring? – Thomas Weller Mar 17 '21 at 08:01
  • 1
    Applied to a resistive load, if the sign of the potential difference remains the same, the current is (as commented elsewhere) not alternating. But applied to a reactive load, the key point identified by @mmmm is whether the energy transfer is predominantly AC or DC. – grahamj42 Mar 17 '21 at 11:23
  • 1
    I agree with this, but I would change "plainly put: no" to ["it depends"](https://i.stack.imgur.com/ZvRIQ.png): all three are AC, but `V(a)` is pure AC, `V(b)` is DC+AC, but overall it's still AC (current changes sign, as you say), and `V(c)` is DC (borderline, but anything from then on is DC), since the current has one sign, only. – a concerned citizen Mar 17 '21 at 15:06
25

It has both AC and DC components. Which of these you regard as "the signal" depends on its intent (i.e. on the information content).

If it is say analogue audio with some DC in there for some reason, then it is an AC signal with a DC offset or bias.

But if it is say a relay control line picking up interference from somewhere, then it is a DC signal with AC noise present.

You can see this difference of viewpoint among other answers offered here.

But be wary of hitting your teacher over the head in such cases; check the answer expected for the exam, as teachers are more focused on getting you through that than they are on the ultimate truth.

Guy Inchbald
  • 1,435
  • 5
  • 10
  • 2
    +1 for this. You've got the rest of your career to reflect on the wrong things you learnt in school! – CCTO Mar 16 '21 at 15:09
  • +1 for use of BIAS and not OFFSET, suggesting the sinusoidal pattern is a signal. Not as much magnetic audio tape around these days. – mckenzm Mar 17 '21 at 21:01
  • @mckenzm Probably not so many quadrupole mass spectrometers either. – Guy Inchbald Mar 18 '21 at 09:06
13

The signal is changing direction with time (alternating), as well as continuously changing amplitude with time.

I would call that an AC signal of 18 Vpp with a DC offset of +1 V.

Mitu Raj
  • 10,843
  • 6
  • 23
  • 46
  • 1
    It's not clear in the chart, but in the question text it goes from 0V to -8V, so only 8Vpp with a DC offset of -4. By my understandng this would be DC because the direction doesn't change since it spends ~100% of it's time in negative V territory? – coagmano Mar 16 '21 at 23:03
  • 2
    @FredStark maybe that example fits the definition of [Pulsed DC](https://en.wikipedia.org/wiki/Pulsed_DC)? – TylerW Mar 17 '21 at 02:29
  • Definitely looks like it. I'm guessing the teacher deliberately used an example that sits on the borderline of definitions – coagmano Mar 17 '21 at 03:04
  • Agree with Tyler on that. That's right on the borderline as well. Because DC amplitude is now the same as AC amplitude for that signal. – Mitu Raj Mar 17 '21 at 04:09
12

This is really a terminology question. For instance, Wikipedia has a classification under which your signal would be called either "Alternating" or "Pulsating":

enter image description here

Clarify with your teacher what is his definition of AC and DC, and then apply those definitions to the signals you need to classify.

Understand that such classification for arbitrary signals is descriptive rather than strict. There is no special Ohm's law for pulsating current. There are formulas/models which only work in special cases, such as constant voltage / current (which you would colloquially call DC) or for a periodic sinusoidal signal with zero offset (which you'd call AC). Obviously, the models which assume constant voltage won't apply to your signal, even if your teacher calls it DC.

Dmitry Grigoryev
  • 25,576
  • 5
  • 45
  • 106
4

EVERY signal contains components of both AC and DC. The issue is one of dominance. Which of these components dominates the signal content.

Examples:

A "DC" power source is likely to contain a small amount of noise or "ripple". The noise/ripple will be AC however the dominant component will be the DC offset.

An "Audio signal" might contain a small DC offset. The signal will be primarily AC, but will contain a small amount of DC.

Mark
  • 299
  • 3
  • 11
3

I wonder what the time scale is... If it's hours, days or years you could describe the voltage as 'slowly varying DC'.

Martin Rice
  • 31
  • 1
2

It is varying with time and also goes below zero, for all the definition is AC.

For some (few) people if a signal is always positive or always negative it could be defined DC (since the direction stays the same).

For all the practical uses DC is fixed, constant and only with some ripple or noise (if acceptable)

Lorenzo Marcantonio
  • 8,231
  • 7
  • 28
1

It's AC with a DC offset, or it's DC with an AC ripple. Your choice, depending on what you're going to use it for.

Pete Becker
  • 1,390
  • 1
  • 9
  • 15
1

Your question is a contradiction:

If the polarity varies from 0V to -8V , shouldn't it be AC?

If the VOLTAGE ranges between 0V and -8V then the polarity doesn't vary.

Furthermore your drawing says 10V not 0V, so your green graphline should be under the blue line.

VDC + VAC = VDC if only |VDC| >= VAC

Pieter
  • 11
  • 2