Clarification on some terms regarding AC, DC and Waveforms.

[mart1n]

Hello folks,

Is any circuit containing a waveform classified as AC? What if the waveform doesn't have negative voltage? (Entire waveform above 0V).
I've always thought about the circuits I was learning with as DC...but if I build an oscillator, does it become AC? I mean just about every useful circuit has to perform some type of signaling...weather analog or digital, are most circuits "AC"?

Thanks!

[Ammar]

Good question and yes, it is confusing.

DC is used to refer to a constant voltage/current. As in a straight horizontal line. Strictly speaking AC refers to a "pure" (never actually perfect though) sinusoid.

Your question is, what about square waves, saw tooth, etc. What are they?

The beautiful thing about sine waves is that any periodic signal can be described as a sum of sine waves. This is thanks to the Fourier Transform. Maybe this link will give you a sense: https://en.wikipedia.org/wiki/Square_wave In the case of a square wave, this sum is made up of an infinite series of sine waves.

Saw tooth waves, square waves are therefore not DC or AC, but they exhibit more AC characteristics than DC. So a capacitive divider driven by a square wave will give you a square wave output with lower voltage magnitude, much like a sine wave. However, a capacitive divider driven by a DC signal will not.

I hope that gives you some sense.

[mart1n]

What about a sine wave without a negative voltage? Like 5vpp that is from 1v to 6v? I'm guessing that since the current would no longer alternate that it wouldn't be AC?


Sent from my iPhone using Tapatalk

[tautech]

To expand some on  Ammar's explanation, think of AC and DC as power or current sources only.

Waveforms can simply then be defined as periodic or nonperiodic or as some might prefer repetitive or nonrepetitive.

BUT to confuse the issue further a current probe might be referred to as a DC (does AC too) or AC current probe, the DC type self explanatory but the AC probe is only suitable for fluctuating current measurement that could be repetitive or not BUT above a minimum frequency, commonly in the range of 10-50 Hz.

Yep, to some it would seem confusing. 

[Benta]

What about a sine wave without a negative voltage? Like 5vpp that is from 1v to 6v? I'm guessing that since the current would no longer alternate that it wouldn't be AC?

You'd normally analyze this as an AC voltage superimposed on a DC voltage (or current, of course).
This can sometimes be tricky if the AC signal is not a pure sinewave, as the "imaginary" DC voltage can depend on the shape of the AC voltage.

[Ötti]

An AC waveform is per definition produced by an alternating current, so that the mean current over time is 0A. This definition implies that shifting the waveform to a higher level by adding a DC part, the waveform is no longer an AC type. Because doing that the mean current value will no longer be 0A. Even if you shift the waveform only a little bit upwards so that there are still big negative parts, the definition of AC is violated.

Surely this depends on the point of your circuit where you reference your signals to. Btw, forget the notation "0V". What you really mean is the reference point of the signals and supply voltages of your circuit.

[rs20]

The meaning is often clear from the context.

-- If a power supply is labelled as having "DC output", then that's a dead steady voltage (nominally).
-- If a power supply is labelled as having "AC output", then that's going to be a square wave or sine wave, centered on zero.
-- The 5Vpp sine wave you describe, superimposed on 3V DC, i.e. ranging from 1V to 6V, would be a bizarre power supply that would be considered faulty regardless of whether it was labelled "AC" or "DC".

So in general, your question "What is my signal ranging from 1V to 6V? AC or DC?" is a fool's errand; a false dichotomy that no professional engineer is troubled by. See how the terms AC and DC are used in reality, without obsessing over trying to find a precise definition that cleanly slices every possible waveform into one camp or the other.

It's much more useful to think about considering the DC and AC components of any given signal. The DC component is just the average voltage over time. Subtract that away, and you're left with the AC component. When your multimeter is set to read Volts DC, it (nominally) is telling you the voltage of the DC component, the average voltage. When your multimeter is set to read Volts AC, it (nominally*) gives you the RMS voltage of the AC component of the signal. This ties in with what Benta is saying, if you superimpose (add) the AC and DC components, you get your original signal back.

So I reiterate, your signal ranging from 1V to 6V is neither AC (as typically defined as mean = 0) nor DC (as typically defined as perfectly steady; variance = 0). It's a combination of both AC and DC.

* So many ways this can go wrong in reality in the case of multimeters reading Volts AC, but that's a topic for another day.

Waveforms can simply then be defined as periodic or nonperiodic or as some might prefer repetitive or nonrepetitive.

Tautech, you used this definition to confuse yourself, because it's utterly unrelated to AC vs DC. Sine waves and DC are both periodic, i.e., f(x) = f(x+p) for some period p. But sine waves are clearly AC and DC is clearly DC.

[tautech]

Waveforms can simply then be defined as periodic or nonperiodic or as some might prefer repetitive or nonrepetitive.

Tautech, you used this definition to confuse yourself, because it's utterly unrelated to AC vs DC. Sine waves and DC are both periodic, i.e., f(x) = f(x+p) for some period p. But sine waves are clearly AC and DC is clearly DC.

Thread topic: AC, DC and Waveforms.

I'm not confused, AC and DC refer to power or current sources, AC frequency internationally is 50, 60 or 400 Hz and is accepted as sinusoidal. Period.

Anything else is a waveform, beit periodic, repetitive or not and it might be riding on AC or DC.
And yes, AC power has a waveform while DC does not.

[rs20]

Anything else is a waveform, beit periodic, repetitive or not and it might be riding on AC or DC.
And yes, AC power has a waveform while DC does not.

You claim that the terms "AC" and "DC" refer to sinusoidal and DC power sources, but we see the terms AC/DC coupling on our oscilloscopes, AC coupling via capacitors, and speaking about the AC and DC components of signals. All of these terms hinge on the definition of DC being perfectly steady (or implying "time average"), and AC having a mean of zero*  -- including square waves, AC-coupled pseudorandom binary sequences, or white noise. Let's at least meet in the middle and say that the term "AC power source" is assumed to be sine without further qualification, but that this doesn't imply that AC coupling nor the AC component of a signal is pure-tone sinusoidal.

I have never heard of this idea the terms AC, DC, and Waveform are a "trichotomy" (3-way dichotomy) as you seem to imply in your message. I fail to find a single source to back this notion up after some brief google searching, either. Waveform seems to basically mean "mathematical function" as far as I can tell, and so by that definitions waveforms can be AC, DC, or a combination.

I mean, you mentioned AC and DC, then said "everything else is a waveform", and then asserted that AC has a waveform. That's a straight-up contradiction.

And I think it's also a bad idea to feed the pointless "black is not a colour"-style idea that "DC is not a waveform" to a beginner's poster asking for clarity. Whether or not that statement is true, it is useless at best.

* Or, 50% of the time, defined more loosely as going both positive and negative.

[tautech]

 
You're nearly there.
Correct, this is the Beginners board and we need to KISS so hence my attempt to keep things simple and easy to understand. These simple but very important principles set the Newbie with the core understanding to move into the electronic hobbyist world with a sound footing and then relate to others , maybe us, of what they are observing or trying to measure.

DC, a power or current source with an steady amplitude*.
AC, a power or current source with a set sinusoidal frequency and amplitude*.

Waveform, every other fluctuating voltage or signal that isn't one of the above.

Then we could go all analytical and define the types of waveforms but we'll wait for the OP to ask. 


* not mentioned in the OP but needed to clarify context.

[rs20]

Just briefly tautech,

1. You insist on defining AC to mean purely sinusoidal, but if one defines it as a waveform with a mean of zero, that definition is just as simple to understand, and is actually compatible with "AC coupling" and "AC component" and all sort of other uses of the term AC as well. So why do you choose the restrictive, purely sinusoidal definition?

2. You insist on defining waveform as anything that is neither DC nor sinusoidal, but that is at odds with every definition I can find anywhere. Why write that sentence down; how can it be of any value to the OP when it's utterly objectively wrong?

I deeply hate it when people provide incorrect definitions in the name of "KISS" when better, equally simple explanations exist.

[Zero999]

An AC waveform is per definition produced by an alternating current, so that the mean current over time is 0A. This definition implies that shifting the waveform to a higher level by adding a DC part, the waveform is no longer an AC type. Because doing that the mean current value will no longer be 0A. Even if you shift the waveform only a little bit upwards so that there are still big negative parts, the definition of AC is violated.

That's not true. Even if the voltage never goes below zero, the current still reverses. The conductors have parasitic inductance and capacitance which need to be charged and discharged.

[mart1n]

Btw, forget the notation "0V". What you really mean is the reference point of the signals and supply voltages of your circuit.

OK, I know voltage is relative, but doesn't voltage need to go negative before the actual electron flow will alternate (Maybe change polarity is a better term?)?

Even if the voltage never goes below zero, the current still reverses. The conductors have parasitic inductance and capacitance which need to be charged and discharged.

Really? Can someone explain and/or point me at some reading material?

Thanks for taking the time to reply, it's been educational.

[Vtile]

Btw, forget the notation "0V". What you really mean is the reference point of the signals and supply voltages of your circuit.

OK, I know voltage is relative, but doesn't voltage need to go negative before the actual electron flow will alternate (Maybe change polarity is a better term?)?

Even if the voltage never goes below zero, the current still reverses. The conductors have parasitic inductance and capacitance which need to be charged and discharged.

Really? Can someone explain and/or point me at some reading material?

Thanks for taking the time to reply, it's been educational.

[DC voltage source plus]----[CAPACITOR]----[DMM in voltage range]----[DC voltage source minus]
Where capacitor and DMM are in series. DC voltage source should allow current sinking.

Now turn slowly the voltage up to few volts, your DMM will show voltage (actually current because of measuring configuration)
Now stop the voltage and the voltage reading goes soon to zero (time depends the size of the capacitor)
Now turn down the voltage (DMM show again voltage, but in reverse polarity/direction)

AC plus DC combination at times is called as pulsating DC. This kind of pulsating DC is example the voltage after rectifier bridge.

[tautech]

1. You insist on defining AC to mean purely sinusoidal, but if one defines it as a waveform with a mean of zero, that definition is just as simple to understand, and is actually compatible with "AC coupling" and "AC component" and all sort of other uses of the term AC as well. So why do you choose the restrictive, purely sinusoidal definition?

Because for the sake of simplicity AC need be thought of as MAINS DERIVED POWER and other sinusoidal waveforms may not be, take a uP clock for example, call it AC to the confusion of everybody. Another, would you call the primary side waveform from a free running(fixed frequency) SMPS AC ? To do so would confuse many and while most of us might know what you're referring to, for the novice it again is confusing.
Another, Ripple on a DC supply is often referred to as AC on DC but what sinusoidal component does the ripple contain ? NONE, it is a ripple waveform. Period.

2. You insist on defining waveform as anything that is neither DC nor sinusoidal, but that is at odds with every definition I can find anywhere. Why write that sentence down; how can it be of any value to the OP when it's utterly objectively wrong?

Why does this thread exist......because the definitions themselves confuse.
We both have a picture in our heads of what these definitions mean but when we put these to print is when things get interesting and the whole point of this discussion. To use such general terms without them overlapping is where problems of meaning and understanding arise so we should define them for the novice in a way that make it easy to fall into a class: AC, DC and everything else.

Part of this discussion and my firm views stem from my early learnings of CRO scope usage when I was advised to best guess the waveform I would see before connecting to a DUT and the names of the waveforms to expect without blindly trusting the scope to display what seems to be there. Furthermore much of my learning was over the phone to a very patient EE Prof that insisted that waveform definitions were correct so zero misunderstandings arose. The definition of a waveform is just as important as where in a DUT it is seen and along with the fact it might not be a perfect ramp, pulse, square or sine.

I deeply hate it when people provide incorrect definitions in the name of "KISS" when better, equally simple explanations exist.

You might, and I deeply hate stuff I find online that is plainly wrong, whether it be in fact or meaning.
I don't mind at all that you challenge me, the discussion is of benefit to those trying to get a grasp on these simple principles that will provide guidance for them going forward.
That just you and I are having some small difficulty finding common ground is most interesting as some would seem to think like you and others like me as we neither have been challenged. 

[Ötti]

An AC waveform is per definition produced by an alternating current, so that the mean current over time is 0A. This definition implies that shifting the waveform to a higher level by adding a DC part, the waveform is no longer an AC type. Because doing that the mean current value will no longer be 0A. Even if you shift the waveform only a little bit upwards so that there are still big negative parts, the definition of AC is violated.

That's not true. Even if the voltage never goes below zero, the current still reverses. The conductors have parasitic inductance and capacitance which need to be charged and discharged.

What I wanted to say is: you have an AC waveform (sine wave centered at GND, mean current over time is 0A). When you add a DC part (shift the wave up or down), it is per definition no longer an AC waveform (mean current over time not equal 0A). That was one of the questions.

[tautech]

An AC waveform is per definition produced by an alternating current, so that the mean current over time is 0A. This definition implies that shifting the waveform to a higher level by adding a DC part, the waveform is no longer an AC type. Because doing that the mean current value will no longer be 0A. Even if you shift the waveform only a little bit upwards so that there are still big negative parts, the definition of AC is violated.

That's not true. Even if the voltage never goes below zero, the current still reverses. The conductors have parasitic inductance and capacitance which need to be charged and discharged.

What I wanted to say is: you have an AC waveform (sine wave centered at GND, mean current over time is 0A). When you add a DC part (shift the wave up or down), it is per definition no longer an AC waveform (mean current over time not equal 0A). That was one of the questions.

Sinewave with a DC offset is universally understood.
With terminology the devil is in the detail. 

[Brumby]

Btw, forget the notation "0V". What you really mean is the reference point of the signals and supply voltages of your circuit.

OK, I know voltage is relative, but doesn't voltage need to go negative before the actual electron flow will alternate (Maybe change polarity is a better term?)?

Everyone seems to have missed this part of your confusion.

The trouble comes from the terminology being used - and the fact that people who have used it for years don't realise what they are saying is - strictly speaking - incorrect.

Often people say AC - which means Alternating Current, when - to be correct - they should say Alternating Voltage.  (Let's not talk about current flow for the moment as we look at the voltage side of things....)

A voltage which varies in magnitude (especially in a periodic fashion) is said to be alternating.  We might have a voltage that can vary from 1v to 7v or from -3v to +3v, but it is still an alternating voltage with a magnitude of 6v peak to peak.  But, you might say, if the alternating voltage is 6v p-p for both situations, then what is the difference - because the voltage measurements are completely different?

This is where the DC Offset (Again, this should be Direct Voltage) comes in.  By adding in a steady (direct) voltage, the absolute voltages of the Alternating Voltage will shift accordingly.

Let's start off with a nice sine wave which swings between +3v and -3v.  Since the positive half of the sine wave is identical to the negative half, the net difference of voltage over one cycle is zero.  This means it has an offset of zero.

If we now add a steady 4v into the circuit, the sine wave will swing from +1v to +7v.  We still have a 6v p-p alternating voltage, but with a steady +4v offset.


Now we can talk about currents.

When the (6v p-p) alternating voltage with zero offset is connected to a circuit, so that current can now flow, the polarity does indeed reverse every half cycle and the electrons do indeed change direction.

When the (6v p-p) alternating voltage with 4v offset is connected to a circuit so that a current can now flow, the polarity never reverses, however, the flow of electrons changes in magnitude.


The trick (which every EE has learned to do automatically) is when you see AC and DC used in situations where voltages are the subject, is to think, speak and act in voltage terms.  It's just one of those things you'll need get used to.

...ff

[Brumby]

Now, why do we use the AC and DC components approach?

Simply - because it works ... and it makes life so much easier.  Especially the mathematics.

If you want to work out what happens to a signal in a circuit, it is so much easier to look at the AC component alone without having to worry about any DC.  Likewise in working out the DC side of things without having to worry about the AC.

Just do them separately, then add them together.

[tautech]

The trick (which every EE has learned to do automatically) is when you see AC and DC used in situations where voltages are the subject, is to think, speak and act in voltage terms.

Maybe so, but when it comes to an electrical circuit that actually does something it gets its power from a current source of an appropriate voltage. Voltage alone does jack shit, current is the driving force, beit uA or thousands of Amps.

Then it must be understood there are ideal and real current sources, real sources are subject to sag or droop and pull the voltage down too. This is where the engineer need allow for this in his/her design to permit the circuit to behave as it should.

[Brumby]

I don't disagree - but the nature of a beginner's question in the Beginner section of the forum needs understanding of where the confusion lies with the inquirer.

Applying the information in real world circuits is impossible unless the confusion is identified and cleared up.  Adding additional information that doesn't directly address the key elements of the confusion just adds to it - no matter how correct they may be.

My answer is very basic for this very reason.

Until the automatic conversion of AC and DC terms (which are, literally, referring to current) are seamlessly interpreted correctly when referring to voltages, then confusion will continue.


The vast majority of us have had this under our belt for so long, it might be hard to realise how anyone could be confused.  Reading the question properly is key to giving helpful answers.  I believe I have done that - but only the OP can say whether it has been helpful.

[Brumby]

Voltage alone does jack shit, current is the driving force, beit uA or thousands of Amps.

Agreed - but that understanding comes later for the OP.

[tautech]

Voltage alone does jack shit, current is the driving force, beit uA or thousands of Amps.

Agreed - but that understanding comes later for the OP.

It shouldn't, it's a fundamental fact, and one that's best emphasised early in teachings.

[Brumby]

There are lots of fundamental facts.

When I say 'later' it could be Chapter 1, paragraph 2 - but they still have to understand paragraph 1 first.

I'm not playing down the importance of what you are saying - just that the points I am making are directed at helping the OP join the dots correctly.

My words are simply focusing on the specific details in what I saw as the OPs cause for confusion.  Yes, there are several other important elements that come into play when looking at the subject as a whole, but when there is something specific that appears to be the problem, I would think addressing that specific thing is warranted, without clouding it with other details.