what does THD % MEAN?

[allanluberda143 10+ year member]

WHATS DOES THD % MEAN, I SEE AN AMP FOR SALE AND IT SAYS

THD at RMS output: 0.01%

AND IVE SEEN OTHERS SAY

0.04% IS IT BETTER OR WORSe?

AND WHAT DOES THE TERM MEAN LIKE T.H.D WHATS IT STAND FOR...

Click to read more...

 

[Side Show 10+ year member]

Total harmonic distortion.....

 

[whitemdx 10+ year member]

http://en.wikipedia.org/wiki/Total_harmonic_distortion

basic the less THD, the better, but it comes to a point where human ears can't tell the difference.

 

[Immacomputer 10+ year member]

Whenever you have an input signal that is slightly distorted, you have harmonics of that signal's frequency. The ratio of the total harmonic signal strength compared to the original signal strength is your THD. Anything below ~5% and you won't be able to hear the difference unless you have really good ears. In amps, the difference between 1% and .01% is not as huge as it appears. Your speakers will usually distort the signal more than 10% anyways.

Here are some pictures from a spectrum analyzer and o-scope that show you what I'm talking about.

First is the clean 50hz sine wave. This signal is undistorted.

Scope shot (amplitude vs time graph):

Spectrum shot (this shows the power vs frequency graph):

Now here is the clipped 50hz sine wave and its spectrum graph:

All those extra bars in the spectrum represent power from the harmonics. It's the ratio of the summation of the power from these signals compared to the power of the original signal that gives you the total harmonic distortion. This graph also shows why clipping a signal introduces more power than it's clean counterpart. That extra power is what can blow the speaker, not the waveform.

 

[AcidicDreams 10+ year member]

All those extra bars in the spectrum represent power from the harmonics. It's the ratio of the summation of the power from these signals compared to the power of the original signal that gives you the total harmonic distortion. This graph also shows why clipping a signal introduces more power than it's clean counterpart. That extra power is what can blow the speaker, not the waveform.

I can't quite wrap my brain around that... could you go into more detail or use more layman's terms?

 

[Immacomputer 10+ year member]

Well, it's actually much more complicated than what I typed and difficult to understand without knowing about Fourier Series and why it's used.

When a signal gets clipped, upper order harmonics begin to enter into the main signal. This harmonics are multiples of the original signal. That means if your original signal was a 50hz sine wave, you would have harmonics at 100hz, 150hz, 200hz, 250hz, etc.. Once you clip the signal, these harmonics become stronger and stronger. Their amplitude (the y-axis on the graph measured in dB) corresponds to the power that each harmonic adds to the original signal. So once you clip the 50hz sine wave, you're actually adding more power to the signal due to the harmonics at the other frequencies. You can hear these harmonics when you listen to the clipped signal. This will give the original signal a "crunchy" sound that we call audible distortion. This is very similar to the distortion knob on a guitar and how it adds crunch to the sound.

Now to understand THD more, think about these harmonics each adding a small amount of power to the original signal. If you add up all the power that the harmonics are producing (and exclude the original signal) and divide that by the power of the original signal, you will have a percentage. This is your THD. The lower the amplitude of the harmonics, the lower your THD will be.

This can be calculated using Fourier Series but that is pretty complicated and laborious.

Did that explain things any better?

 

[AcidicDreams 10+ year member]

Well, it's actually much more complicated than what I typed and difficult to understand without knowing about Fourier Series and why it's used.
When a signal gets clipped, upper order harmonics begin to enter into the main signal. This harmonics are multiples of the original signal. That means if your original signal was a 50hz sine wave, you would have harmonics at 100hz, 150hz, 200hz, 250hz, etc.. Once you clip the signal, these harmonics become stronger and stronger. Their amplitude (the y-axis on the graph measured in dB) corresponds to the power that each harmonic adds to the original signal. So once you clip the 50hz sine wave, you're actually adding more power to the signal due to the harmonics at the other frequencies. You can hear these harmonics when you listen to the clipped signal. This will give the original signal a "crunchy" sound that we call audible distortion. This is very similar to the distortion knob on a guitar and how it adds crunch to the sound.

Now to understand THD more, think about these harmonics each adding a small amount of power to the original signal. If you add up all the power that the harmonics are producing (and exclude the original signal) and divide that by the power of the original signal, you will have a percentage. This is your THD. The lower the amplitude of the harmonics, the lower your THD will be.

This can be calculated using Fourier Series but that is pretty complicated and laborious.

Did that explain things any better?

Yes, much better. That makes alot of sense... now what causes "Harmonics", is it just the nature of electronic amplification or can you explain why that happens... also when you clip the signal do the harmonics get sent to the speaker at their freqs? So you get the one main 50Hz clipped signal and then small signals at 100, 150,200 etc?

Also can you explain clipping as well? You are explaining this stuff pretty well, I might as well take advantage of that to learn something...

 

[DejaWiz 5,000+ posts]

THD isn't going to mean much if it's at or below about 1% at measured power. Most humans can't discern distortion coming from a speaker until it's in about the 3-5% range. Of course, that's only what I've read on the internet, so yrmv.