How important is THD in an amp...

mwaarna

I am starting to do research on amplifiers...

One amp that gets referenced alot is the sunfire grand cinema amp.. The power is impresssive at 5x200 but .5% thd.

Another amp i am looking at is the outlaw 7500 which is 200watts but thd of .03%

How much does THD come into play with amplifiers?

Also how much overhead should one have for their speakers... as in the Polk LSI series are rated at 250watts.. how many more watts should you have in your amplifier so its not working at its max load? 50, 100watts, ??

heiney9

In today's world THD rating's mean nothing at all. In fact many early amp designs tried to achieve very low THD on the spec sheet and ended up sacrificing sound quality to get that measurement so low. As stated anything under 1% is virtually inaudible.

To be honest there isn't much on a spec sheet that tells about how the amp will sound or how well it will drive odd resistive loads.

H9

dkg999

Don't all amps sound the same?


















(Just kidding, don't get your panties in a bunch, all stunts performed by professionals, don't try this at home )

avguytx

Hey, I can pick ANY amp out in a quadruple blind ABCDX testing with everything wired out of phase at a construction site. How's about dat?














(Of course, I'm BS'ing too. )

heiney9

dkg999 wrote: »

Don't all amps sound the same?

Yes, if I cup my hands over my ears

venomclan

As long as the amp is 200 watts or better, that is all that matters. Brands, cost, ohms, amps, THD, rating - all hyberbole.

Monster Jam

Amps are over-rated. My panasonic receiver drives my LSi25s just fine.























....in the spirit of being over-the-top....

cmy330go

venomclan wrote: »

As long as the amp is 200 watts or better, that is all that matters. Brands, cost, ohms, amps, THD, rating - all hyberbole.

:rolleyes:
My 55w/ch tube integrated sounds great.

George Grand

Anything at or below .5% might be inaudible, but between .5 and 1.0% is why most of those HT receivers sound like a bag of **** cats okay?

I love it. Here we are at Tube Central (typically well above .5%), and the threshold of human audibility goes up to 1.0%!!!

You guys kill me sometimes.

It's pretty important mwaarna, but not TOO IMPORTANT if you have a shitload of money invested in your system (I think you can break the code on that one).

RuSsMaN

As high as 8-10% percent for under 100~cycles, right? I thought that MOST human ears (98%) couldn't discern less than one half of one percent .5% from 200 cycles and up?

A lot of designs introduce a ton of negative feedback to 'get the numbers down', which isn't always a good thing. SOME negative feedback is fine, you don't have to run in triode class A - neg feedback guys don't get your panties in a wad - I said TOO MUCH is bad.

.5 or less, you're good to go. Like George said, I've had tube amps which look HORRIBLE on paper, but man if you HEARD them....

Slew rate and damping factor are big players today, and almost as unimportant. Get a good amp, with a couple db headroom, plenty of capacitance in the power supply, and discrete outputs - you'll be good to go for most appilications.

Cheers,
Russ

George Grand

Mwaarna,

Be advised, the amp you're looking at, the Sunfire, is rated at .5% OR LESS, not AT .5%

Almost every Carver amp made was rated at no more than .5% THD. The M-500 and M-500t were the cleanest, at .15% or less. Nice amp.

venomclan

cmy330go wrote: »

:rolleyes:
My 55w/ch tube integrated sounds great.

That is not possible, it is not 200 watts!....:mad:

hypertone

lol

jakelm

Total Harmonic Distortion.

I have read many articles on THD, but if someone could put THD in English.

Input clean and output trying to amplify too much in which it come out distorted?

Is this literal distortion, like overdriving a speaker, with not enough wattage?

Or does THD mean something totally different?

heiney9

jakelm wrote: »

Total Harmonic Distortion.

I have read many articles on THD, but if someone could put THD in English.

Input clean and output trying to amplify too much in which it come out distorted?

Is this literal distortion, like overdriving a speaker, with not enough wattage?

Or does THD mean something totally different?

FWIW,

THD (Total Harmonic Distortion)
The ratio of the power of the fundamental frequency at the output of a device versus the total power of all the harmonics in the frequency band at the output of the device. Basically, all electronic audio devices introduce some distortion to audio passed through them. The simplest form of this distortion is the addition of harmonics to the outputted signal. THD represents the sum of all the harmonics added by a device as a percentage of the level of the signal being measured. The closer THD is to zero, the more "transparent" a device should sound (all other things being equal, which they never are...).
Various devices contribute differing types of harmonic content to a signal, this is part of what can give them their distinctive sounds. For example, tubes add different harmonics than transistors, different circuit designs emphasize different harmonics, etc. When engineers talk about the "sound" of a piece of equipment, this is a part of what they are referring to.

The other distortion.

Intermodulation Distortion (IMD)
The interaction of two or more frequencies in a signal that results in the generation of new frequency components not present in the original signal. These new components have frequencies equal to the sum and difference of the frequencies of the original signals, and integral multiples thereof. IMD is often a major issue in loudspeaker design due to the varying permutations of issues that arise as a speaker cone moves back and forth.

Harmonic Distortion
Since no electronic device is perfectly linear (meaning the output exactly equals the input) harmonic distortion is a fact of life in all audio components. Most audio signals have harmonics associated with them (a perfect sine wave is one notable exception), and that is what gives them their characteristic sound. An oboe sounds different from a violin mostly because of the harmonic series produced as part of their distinct sounds. The corresponding difference in the shape of their respective waveforms is easily distinguished when viewed on an oscilloscope or a computer audio editing program. Harmonic distortion is the result of a device subtly, or not so subtly, changing the shape of the waveform which alters the relative levels of various harmonics associated with that sound. The more harmonic distortion there is the more the sound will begin to take on the quality we all know and love that we call "distorted".

In spec land you will often see the specification for THD which stands for Total Harmonic Distortion. This is a rating given to most gear for the overall percentage of harmonic distortion added to the signal passing through the device while operating at (presumably) nominal levels. There are dozens of ways to measure this spec that can skew the results so keep that in mind when comparing product literature.

Harmonic
In audio a harmonic is sort of the opposite of a fundamental, though technically the fundamental is also considered a harmonic. Pretty confused? Harmonics of a particular waveform are multiples of its fundamental frequency. The first multiple is obtained by multiplying the fundamental frequency by one (1). Therefore in a strict sense the first harmonic is the same value (frequency) as the fundamental. The rest of the "harmonic series" (2x, 3x, 4x, etc.) of a sound make up the basic character, or timbre, of the sound based upon all of their relative amplitudes (levels).

In the discourse of guitar playing (though this concept applies to all stringed instruments) a harmonic is a technique where a string is made to sound at some multiple of its fundamental frequency. This is achieved by applying light pressure at some point along the length of the string and exciting it into vibration (usually with a pick). With this technique the fundamental frequency of the string is (nearly) muted by the pressure, but depending upon where along the length of the string pressure is applied the harmonics are excited differently. This has the effect of changing the apparent pitch of the note played, but the notes always have some relationship to the fundamental frequency of that string at its given tension and length. This action is fundamentally (no pun intended) different than fretting a string, which actually changes the length of the string and creates a new fundamental frequency.


Not sure if there a really cut and dried explanation of THD and an even less explanation of how much affects what, that makes it desireable or undesirable. There are odd and even harmonics which can either sound good or awful.

Hope this helps Jake

mhardy6647

THD below 1% is not audible.

Well, there have been studies that yielded this result, but it's not really quite that simple. The harmonic content of the distortion will influence whether it sounds "good" (like a tube guitar amp) or "bad" (like a transistor radio with weak batteries). 1% of even harmonics (which will mostly be the second harmonic, one octave above the fundamental) will not sound unpleasant. In the real world, if you hear it, it will be pleasant (euphonic). 1% of third harmonic won't be so pleasant.

Does this help or hinder the discussion?

You might want to look at some of the "Amplification" articles at:
http://sound.westhost.com/articles.htm

Maybe this, too:
http://www.emusician-digital.com/emusician/sample/?pg=68

Or even this, by tube hi-fi good guy Pete Millett!
http://www.pmillett.com/file_downloads/ThesoundofDistortion.ppt
(it's a PPT)

mhardy6647

I think the reason THD became an accepted (indeed, FTC-mandated) specification to indicate something about the quality of output power is this: There is a sharp increase in THD as an amplifier approaches clipping/nonlinearity. As a sine wave is clipped, it starts to resemble (and sound like!) a square wave. A square wave is by definition the (infinite) sum of a fundamental frequency and its (EDIT: odd!) harmonics in definite proportion; thus it has a rich harmonic content. This harmonic content, which was not present in the input waveform, is harmonic distortion by definition. If you sum the contribution of all of the harmonics (which will differ in amplitude, probably decreasing exponentially as the order of the harmonic increases) and compare it to the amplitude of the input signal, you can measure the total harmonic distortion.


Here's a nice depiction of the Fourier approach towards building a square wave from a sine wave of a given (fundamental) frequency and its harmonics.



http://www.allaboutcircuits.com/vol_2/chpt_7/2.html

mhardy6647

Sorry to post ho' -- deal with it :-)

Bear in mind that, for practical reasons, most of the quantitative specifications applied to amplifiers is measured with a discrete frequency sine wave (or finite combinations of sine waves of discrete frequencies, as for IM distortion). As Russ implied, there are tricks that can be used to drive harmonic distortion to undetectable levels, but they must be applied judiciously or they can cause more troubles than they fix. Excessive negative feedback can result in frequency-dependent phase shifts, which are quite likely to be audible and detrimental. HD can be far less unpleasant to listen to (thus the popularity of single-ended amplifiers!)

Technical evaluations of amps (see, e.g., good ol' Stereophile
s test reports) will show square wave response of an amp. Sometimes it ain't pretty! Good reproduction of a square wave by an amplifier requires very wide, flat frequency response and excellent phase linearity.

Russ also said:

Like George said, I've had tube amps which look HORRIBLE on paper, but man if you HEARD them....

The EICO HF-81 (shall I link to my favorite photo, Russ?) is an excellent example of this. Take a look at the Stereophile review that drove the price of this modest but exceptionally good sounding little amp through the roof. The square wave performance of the (actually rather tweaked-out) HF-81 they tested is not too impressive.
http://www.stereophile.com/integratedamps/606eico/




Bottom line: There are millions of years of evolution invested in your ears; trust them!