importance of THD with a sub

It depends on the type of distortion. Higher harmonic distortion is much more audible than lower order harmonic distortion. Supposedly odd order harmonic distortion is more offensive than even order harmonic distortion (kinda like the Star Trek movies), so third order, fifth order, etc ought to be something to look for, but, to be honest, I personally don't exactly know which distortion sounds like what. I do know that a sub with greater distortion is going to sound louder than a clean sub for the same output level on the same source material. Second order harmonic distortion is not easy to perceive, and is considered generally harmless, however I am sure that is only for the levels it is normally kept to in any decent subwoofer. I would guess something like 50 percent second harmonic distortion on a test tone would be pretty noticeable. By the way, I guess I should say at this point, in case you don't know, harmonic distortion are frequencies produced higher than the original fundamental frequency, so they should sound like a higher pitched addition to the original signal.

By the way, harmonic distortion at 20 hz ought to be very perceptible if you are playing a pure 20 hz tone. Since 20 hz is so difficult to hear as is, any other frequency produced by the sub should stick out like a sore thumb. However, in actual music, movies, and games, I am sure it would be far more difficult to discern distortion resultant from a 20 hz frequency playback. 50 hz should be a bit harder, but, of course, it depends on the distortion amount and distortion type.

If that 20% THD is composed of higher order harmonic distortion, that would be a cause for concern. If it is second order harmonic distortion, that isn't bad- in my opinion. There are those here who would find even that much second harmonic distortion unacceptable, but that type of distortion is so subtle I don't think they would realize it is there if it wasn't measured. If I were you, I would be looking at the composition of the THD, not the THD itself. I would be especially weary if fifth order and beyond were to show up in measurements. Another thing to look at is at what output level the distortion show up at. If you don't intend to push your sub to the bleeding edge of its performance, you don't need to worry about high levels of distortion that show up in its maximum output sweeps. Most subs will get a bit cranky when pushed to the limit, and some will even self-destruct.

Those Data-bass numbers are based on max spl's so I would not worry about them unless you are going to push the sub to max levels. I have played with sine waves and THD and 10% THD on 10-20hz sounds good to me but anything higher started to sound bad(not like real bass). Basically the bass was the same up until 10% THD for me and then at 11-13 % THD it started to change the sound of the bass. I say sound because I was at very high levels testing a sub system and not just one sub. It all depends on what you like and each sub is different. I am sure Shady was right about different types of THD because my CHT 18.2's sound great at 10% THD but at 11 % it was awful. My F-20's bass did not change sound until it reached 15% THD(20hz same levels). I did not check to see which had higher THD (2nd or 3rd) but I can look it up(saved graphs).

I have recently been experimenting with THD myself. Once it was beginning to get over 10% it starts to basically sound like the tone of whatever harmonic it is. So for example if you have 20% thd at 20hz, you would most likely also be hearing a 40hz tone or whatever other harmonics are being produced.
I guess it depends on how sensitive you are to it and also what sub is producing it. Definitely in action scenes it gets drowned out. at any reasonable amount. IMO

If I understand it correctly, the CEA max clean output tests use relatively practical harmonic distortion limits. So by looking at those test results for recently measured subs (like the PB12NSD) you can begin to get a bit of a handle on how much of each distortion product can be tolerated without degrading the experience. Before that, I'd often seen 10 percent as a reasonable limit for subs. Clearly the CEA test allows higher levels. That little SVS passes at 25 Hz with a bit over 20 percent distortion. Converting the dB difference in the measurement graphs to percentages is something a person could google. Did it myself once but if I saved what I found, I didn't save it on this computer

FWIW, I've seen it said (although without attribution) that in the musical instrument amplification field, guitar players will idetify anything below 10 percent distortion as clean, undistorted sound. And we're largely out of the typical sub range pretty quickly on a standard-tuning six string.

Here's an older thread on this topic with good replies:


http://www.avsforum.com/t/1299983/the-problem-of-distortion

would you buy an amplifier that had 20% THD at 20 Hz ?

people accept large distortion in woofers simply because woofer distortion has always been bad and comparing one subwoofer with another is like comparing one badly bruised apple with another

All you have to do is turn them down so the THD is lower and then buy more to make up the spl difference. This gives us a reference point. I don't run higher than 10%THD.

I think you meant the SB12-NSD. The PB12's limiter engages at all frequencies long before it hits 20% THD.

Nope, those are Ricci's measurements for the SB12NSD. http://www.data-bass.com/data?page=system&id=80&mset=86

You cannot purchase a speaker of any ilk that has amplifier-low distortion characteristics, AFAIK. I suspect that the distortion actually is masked or otherwise unobtrsive or CEA's limits would be different.

while that is generally true it doesn't mean that distortion in speakers is less of an issue than it is in an amplifier just because it can't be reduced to amplifier levels

less is better applies to speakers as well
Edited by benny blanko - 10/1/12 at 8:38pm

Read your original post and my quote of that post: we agree that the SB12 produces 20% THD at 25Hz and maximum output; however, the sub you named in your post was the PB12-NSD, which we agree does not produce 20% THD in any circumstance.