May be this video helps you a little bit. From 1:50 pure tones with added distortion and the corresponding graph.
The lowest couple of harmonics to any signal are usually not offensive to the ear. Higher and especially odd harmonics are more offensive. This has been purported to be because most sounds in nature have lots of 2nd and 3rd harmonics and we are used to hearing that, while the others are not "natural". Also the music scale is congruous with low order harmonics, so these additional (but still unwanted) tones sound consonant with the original signal to the point that some (like Nelson Pass) even try to add them into the amplifier's output.
I am more of a "let's keep the signal pure" kind of guy and prefer very low THD and IMD in an amplifier.
Last edited:
@djn:
You might want to read some of the pages found at links from this Google search:
Google search terms: harmonics+consonance+dissonance+distortion
You might want to read some of the pages found at links from this Google search:
Google search terms: harmonics+consonance+dissonance+distortion
I think you misunderstand the concept of a "resonance".... don't confuse the what with the how.
A harmonic is a resonance: all resonances are integral multiples of the fundamental, f. 2f, 3f, 4f.... nf.
Nevermind how it is caused, by this is what we measure when we introduced a pure, single frequency sine wave and do a spectrum analysis. We don't see stuff at 1.5f, 2.688f....
You can tapdance around this BS all you want - it's still BS.
Harmonic distortion is not caused by resonance.
Off to the ignore list! (You're in excellent company, some of the greatest diyaudio hucksters are there!).
Jan
Harmonic distortion is not caused by resonance.
Off to the ignore list! (You're in excellent company, some of the greatest diyaudio hucksters are there!).
Jan
Oh wow.... Jan. you sure are Noise on the Internet. Pfft...
And to think, all of those years in college studying and all of those decades stuck in a metrology/electronics/computer lab only to be
"Ignored By Jan"
That sure hurts....
Heck, I didn't even bring up the other types of distortion... time based, frequency based, etc.... I mean, a whole career of taking and analyzing electronics and computers out the window because Jan thinks it's all BS.
Do I have to return 40 years of paychecks then?
Last edited:
OKaaaay, so you are saying that there are little "resonances" going on in silicon or within an electronic component? Sorry I don't think so!
A resonance that generates harmonics has physical origins, not electronic ones. Are you a non-native English speaker perhaps?
Also, telling Jan that he "misunderstands" the concept of a resonance is pretty ballsy! Good one there, Tony. Whoo boy. Do you also know more than all the generals, etc.???
Jesus man.... did someone pass out the vinegar pills in this forum lately?
A resonance is a signal based on the harmonic.
We are taking a frequency measurement and looking at resonances -harmonics- that are integer multiples of the fundamental frequency associated with the test signal.
This has nothing to do with electronics, nothing at all. It's a TYPE of measurement.
It could be acoustic, for example, as a bong... or quantum, as the frequency of photons.....
Yes, you all are confusing the how with the what. And Jan is missing my point too, he is not missing the concept of resonance, he is missing the point I am trying to make, you too.
BTW, English is my 4th language and I dare say I speak it and write it with better precision that the vast majority of "native born" speakers. What kind of ethnocentric insult is that.?
BTW-II, in my time, my PSAT and SAT English scores were PERFECT.
Last edited:
Maybe, but a signal based on the harmonic is not necessarily a "resonance". Context is important here, if you want to use the term correctly.
For example an electronic resonance is a way to describe the storage of energy in a system, but that doesn't create new harmonics now does it?
I think you are confusing non-linearities in electronic circuits that create harmonics (and possibly other non-harmonic tones), with resonances in (for example) a string or object. These are completely different things, and to refer to the electronic effects as "resonances" is just plain misuse of the language.
Wow... I am not confusing anything. Let me make this extremely clear....
Again, you are describing the HOW this distortion arises in an electronic system. Of for that matter, in any system. I don't care about the mechanisms by which this type of distortion arises.
I am describing WHAT this distortion is.
I described harmonic distortion in term of what we look for, which is harmonics of the fundamental signal. It is also, defined as a resonance in the vernacular.
Definition of resonance
1a: ...
b(1): a vibration of large amplitude in a mechanical or electrical system caused by a relatively small periodic stimulus of the same or nearly the same period as the natural vibration period of the system
(2): the state of adjustment that produces resonance in a mechanical or electrical system
2a: the intensification and enriching of a musical tone by supplementary vibration
Not quite. Sure an electronic resonance is an "intensification". I called it "energy storage". Same thing. But that is not creating harmonics.
You are calling harmonics "resonances". They are not, when the system in question is an electronic one.
The "vernacular" that you refer to is some layman's idea of harmonics. As an EE I would expect more than this level of understanding of the concept!
The "spike" at 1kHz is fine. Thats what the input signal looks like. It widens a tiny little from a pure line but you can ignore that. If you would listen to that "spike" it would be super clean i.e. distortion less.
Harmonics, ie. the "H" in THD is the unwanted even and odd multiples of the input signal. Distorsion is when you get new tones that was not in the input test signal but created by the e.g. an amplifier. So from 1 Khz you get the
even harmonics:
2,4,6,8 etc kHz added and these are distortion
odd harmonics
3,5,7,9 etc Khz added and these are also distortion
Depending on the device that tries to do only 1kHz but fails, the output becomes 1Khz and some odd and some even harmonics.
Now, the Total Harmonic Distortion, THD is as someone already indicated - all the extra unwanted harmonics "added" as a number and expressed as a percentage relation to the test signal (1 kHz above)- e.g. 1% (the sum of all unwanted extra "tones" summed up had a level that was 1% of that of the test signal. Maybe the 3rd harmonics was 0,1% and the 2:nd 0,9% and no other harmonics was present)
//
I am a physicist. Not an EE, Thank God.
I attempted to use the vernacular to describe what you see in a THD graph. To describe what is being shown without adopting the world view of the engineer who predisposes a given paradigm (*).
I keep getting clobbered, told that I'm ignorant, "ignored", ESL writer, etc, etc... by people who, IMHO, conflate the "what" with the "how".
Forget it.
(*) Please go read Thomas Khun.
At first I thought he meant a resonance within the circuit, on reading it again what he actually means is a resonance in the output of the circuit, no tap dancing required
Good question.
A steady state, repetitive waveform only contains frequencies that are integral multiples of the fundamental frequency.
If the waveform is a pure sine, it only contains the fundamental frequency. If the waveform is a square, triangle, ramp, etc.
then it will contain some harmonics, odd, even, or both.
A nonlinear circuit will distort the shape of a pure sine waveform, but it will still be repetitive, and so it will contain
some harmonics which are still only integral multiples of the fundamental frequency, but no other frequencies.
As both a physicist and EE, I've never heard electrical harmonic distortions termed as resonances.
Last edited:
Jeez... I tried to describe it in non technical terms.
yes, I'm also a physicist, and EE by profession. We don't use the term resonance in our work, but I tried to describe it in non technical terms.
The vast majority of people have NO clue what harmonic means, but they understand resonance.
A huge missing component of the "standard" harmonic distortion measurement is that it takes the measurement with a single 1K sine wave, not taking into account frequency dependent non-linearities.
Nevermind, btw, the back EMF from the load which is indeed resonant.
There....
djn,
You said: "I've been reading a lot about THD, but it all seems to be about what it is not how it can affect music reproduction."
. . . Good question!
Consider the source, the musical instruments:
1. A skillful flautist plays a flute note at about 1kHz. It sounds very pure and smooth, the flautist played it that way.
Almost all the sound is 1kHz, there is very little sound if any of 2kHz, 3kHz, 4kHz and higher harmonics.
2. A skillful flautist plays a flute note at about 1kHz. It sounds very raspy, the flautist played it that way.
The 1kHz is there, but there are also fairly large sound levels from the 2kHz, 3kHz, 4kHz and higher harmonics.
3. A saxophone can be skillfully played to produce either a pure and smooth note, or to produce a raspy note.
Very low sound level harmonics; or quite loud sound level harmonics.
4. A good singer can produce either a smooth sound, or a very un-smooth sound.
Very low sound level harmonics; or quite loud sound level harmonics.
So what?
There are many characteristics of the sound of an instrument that helps us hear and identify what instrument is being played. One very important characteristic is the absence or presence of various harmonics and the amplitude (sound level related to the fundamental note amplitude).
A clarinet sound usually has the fundamental and 3rd harmonic (like 1kHz and 3kHz, and no significant others).
Well, amplifier harmonic distortion Adds harmonics that were not there in the flute, saxophone, and singers original sound and recording (or at least adds to the sound level of the original music harmonics).
THD, Total Harmonic Distortion, is the sum of all harmonics that are produced, when a pure single frequency sine wave is applied to an amplifier (or speaker). Those harmonics were not there in the original sine wave that was applied to the amplifier or speaker.
That applies to a test tone, but also applies to the musical instrument tone and the Timbre of the sound.
djn
Does that answer your original question?
You said: "I've been reading a lot about THD, but it all seems to be about what it is not how it can affect music reproduction."
. . . Good question!
Consider the source, the musical instruments:
1. A skillful flautist plays a flute note at about 1kHz. It sounds very pure and smooth, the flautist played it that way.
Almost all the sound is 1kHz, there is very little sound if any of 2kHz, 3kHz, 4kHz and higher harmonics.
2. A skillful flautist plays a flute note at about 1kHz. It sounds very raspy, the flautist played it that way.
The 1kHz is there, but there are also fairly large sound levels from the 2kHz, 3kHz, 4kHz and higher harmonics.
3. A saxophone can be skillfully played to produce either a pure and smooth note, or to produce a raspy note.
Very low sound level harmonics; or quite loud sound level harmonics.
4. A good singer can produce either a smooth sound, or a very un-smooth sound.
Very low sound level harmonics; or quite loud sound level harmonics.
So what?
There are many characteristics of the sound of an instrument that helps us hear and identify what instrument is being played. One very important characteristic is the absence or presence of various harmonics and the amplitude (sound level related to the fundamental note amplitude).
A clarinet sound usually has the fundamental and 3rd harmonic (like 1kHz and 3kHz, and no significant others).
Well, amplifier harmonic distortion Adds harmonics that were not there in the flute, saxophone, and singers original sound and recording (or at least adds to the sound level of the original music harmonics).
THD, Total Harmonic Distortion, is the sum of all harmonics that are produced, when a pure single frequency sine wave is applied to an amplifier (or speaker). Those harmonics were not there in the original sine wave that was applied to the amplifier or speaker.
That applies to a test tone, but also applies to the musical instrument tone and the Timbre of the sound.
djn
Does that answer your original question?
Last edited:
I suspect you are referring to a higher order resonance, when harmonically related signals combine. This is a linear effect... Do you think it could it be used to describe the production of harmonics from a single tone?