in layman's language, what is Total Harmonic Distortion

Some things can not be made simpler than:
. . . "You should make things as simple as possible, but no simpler" - Albert Einstien

He knew that breaking somethings down to less than their component parts simply (pun) makes them fall apart.

I tried to make my discussion as simple as I could.
 
  • Like
Reactions: JMFahey
Hello Tony,

I think all one has to do is consider how the instantaneous voltage of the sine wave changes and how the non-linear transfer function effects it.

Cheers

So long as that change of the wave is harmonic, then you would see those results...

I'm trying to see if I can find similar "harmonic distortion" in areas besides electronics.
 
That spring example seems the best yet for loudspeakers.

The deeper you scratch the deeper the rabbit hole you go down. Maths, physics, EE and reality are all all attempting scratch the same rabbit hole.. they just decide to stop and have lunch at different points along the way.

You could have a harmonic activate resonance although technically that’s covered by the harmonic being an input too.

Mathematicians want a lunch of caviar, pate de fois gras and champagne.

Engineers make do with Diet Coke and a cheeseburger. They don't stop, they wolf it down while they do something.

Physicists... Aaaaah.... for lunch we pause and enjoy our Cold Beer and Cheez Whiz ( the one with aged cheddar and bacon ) on whole wheat crackers and some sliced cucumbers. And then we resonate in the direction of the mathematicians and engineers while we pull each other's fingers. 😛
 
Last edited:
Physics . . . Hmm.

I seem to remember one of our many college engineering/physicist lab experiments (all the other students took the easier non-engineering/non-physicist classes).

The experiment was about Hook's constant.
It used a spring and a mass.
The mass was pulled down (not very far, so that it would not permanently deform the spring).
Then the mass was released.
The mass and spring began to bounce (Viola! . . . Resonance, Resonance at one and only one frequency).

My goodness, a woofer in free air seems to quite closely duplicate that.
Of course, there are several more other problems with the woofer that are separate from the very well controlled Hooks constant rule.
But in both cases we have Resonance.

Consider a poorly designed single ended no negative feedback pentode amplifier has a very large capacitance wired across the output transformer primary (putting that large capacitance across the primary was what made the amplifier 'poorly designed').
The secondary is loaded by a loudspeaker that has many times higher impedance than the secondary's rated tap impedance.
What do we have? . . . We have Resonance. Resonant frequency = 1/(2 x pi x (root of L x C)).
The Resonance is at One Frequency. That resonance is not harmonic distortion (even though the non feedback single ended pentode output stage does have Lots of 2nd harmonic distortion).

Suppose the resonance is at 200Hz.
Suppose a 70Hz tone is applied to the amplifier.
At the amplifier output we get 70Hz fundamental, and 140Hz 2nd harmonic distortion.
There is No 200Hz output from the amplifier, because there is no 200Hz signal in, so the Resonance is Not Activated.

Does that make the principle of resonance and principle of distortion clear, and completely separate;
or are all the answers distorted (pun intended)?

I regret I used such a word.... yikes.... I tried to make it simple.. I never expected the.....

.
1661618770390.png
 
There definitely seems to be anti-entropy out there.

On the too ordered it made me think of the wonderful world of the Fibonachi sequence and the equivalency to the golden mean.

dave

Fibonaci makes for a great password.... ooops! It is an elegant sequence.

In the case of the Golden Mean... or the sea shell, it sort of makes sense why it would follow such a pattern. The process of making the shell is based on history: that is, the mechanism used to build the next "rung" on the shell is based on the size of the previous two rungs.... hence something like Fibonacci might just appear. I think processes that have history like that would follow a general solution something like: N(i) = F ( N(i-1), N(i-2)... N(i-n) ) where ( n>0 and n<i )

I remember this from a physics journal back in the late 70s:

1661619372091.png
 
Speaking of numbers . . .

Pi is a number that makes the Eveready Bunny look like he is taking a short walk.
Pi has an infinite series of numbers describing it.

But, there are simple approximations:

3 1/7 is close to pi, error is less than 0.05%. That is really close.

Root (10) = 3.162277660. . . another infinite number series, is close to pi, error is less than 0.7%.
That is pretty close too.

The above errors are swallowed up by most slide rule in-accuracies.
 
Speaking of numbers . . .

Pi is a number that makes the Eveready Bunny look like he is taking a short walk.
Pi has an infinite series of numbers describing it.

But, there are simple approximations:

3 1/7 is close to pi, error is less than 0.05%. That is really close.

Root (10) = 3.162277660. . . another infinite number series, is close to pi, error is less than 0.7%.
That is pretty close too.

The above errors are swallowed up by most slide rule in-accuracies.

One of the most common "things to do" in Physics is to do coordinate transformations so problems can be solved in their natural coordinate system.

Note: lots of matrices to do this...

Anyhow, I wonder if we chose the wrong coordinate system to measure our environment. Maybe we should have Pi fingers on each hand.
 
Just stop spewing nonsense and it won´t be needed.

Wel,,, QED...

Cardinal Fang, you could be nice.... I explained WHY I used that word because it resonates with the vernacular and it does explains a frequency plot full of harmonics ( harmonic resonances if you will ).... but it's people like you who took it so literally that make life miserable for the rest of us.

Sometimes I wish I had a hook like they had in the Gong Show... when someone goes like the Energizer Bunny, incongruously, and irrationally, resonating on the same fundamental argument over and over and over, you just hit the gong.... ( filling the hall with a sonorous, authoritative, soul filling rich resonant tone that tolerates no further inharmonious mischief ) and pull them off the stage. And then move on to the next act, hopefully a lady less clad and a bit more entertaining.

BTW, speaking of the vernacular:

resonate​

rĕz′ə-nāt″​

intransitive verb​

  1. To exhibit or produce resonance or resonant effects.
  2. To evoke a feeling of shared emotion or belief.
  3. To correspond closely or harmoniously.
 
Last edited:
The
So in this picture I snagged from the thread link, does that huge spike at around 1khz going to sound horrible?

View attachment 1084393
The 1kHz spike is the signal into the amp input. It's supposed to be high. If the amp was perfect, nothing else would be there but the 1kHz. All the other stuff are harmonics produced by electronic circuitry in the amp, not present in the input signal. If the input signal was directly graphed, without going through the amp, only the 1kHz would be visible, unless you really dropped the floor on the graph, or if you had a noisy signal generator.
 
Ok, back to the OP's question of what is THD? (Total Harmonic Distortion).
It is the sum of all the harmonics, as that sum relates to the strength of the fundamental.
It is most often expressed in percent.

On the graph in question, the 1 kHz "spike" is the strength of the fundamental.
If you see other spikes at 2kHz, or 3kHz, etc., those are the harmonics.
Sum up the harmonic strengths, and compare to the fundamental strengths.

Forget about the math that it takes to do the proper sum, and then compare to the fundamental, etc.
That is not a simple addition of dB that the graph units are in (dB).
You wanted simple, so do not do the math, leave it to others to do that for you.

Different strengths of the harmonics versus the fundamental, whether they are from a musical instrument, or from an amplifier, or a loudspeakers changes the sound that you hear.

Simple?
Yes?
No?
 
Last edited:
  • Like
Reactions: JMFahey