Help understanding speaker harmonic distortion

[Volutrik]

I'm still deciding whether I should buy speakers or not. The options I have are the D3V's and the LP UNF's. But looking at their distortion measurements, it really throws me off.

I don't really know how to read distortion measurements, but with my limited knowledge in this topic, this looks bad. Near field monitors are supposed to be played quietly, and 86dB's isn't that loud, is it? Yet, at 86dB's, they seem to measure bad. Again, my understanding about speaker distortion is limited, so I don't really know if it's THAT bad. I'd like some help with this. Thanks.

 

[boxerfan88]

-30dB is about 3%
-40dB is about 1%

My personal criteria is <= -40dB from 100Hz to 10kHz if possible at the "normal" loudness.
For expensive speakers, I look for <= -40dB from 100Hz to 20kHz at high SPL (usually measured at ~95dBSPL).

Hope the above helps...

Edit ... added my current speakers 95dBSPL THD for your reference ...

 

[Thomas_A]

It is so much dependent on how loud you want to play. But I would say that there is a substancial number of speakers that have distortion peaks in crtitical frequencies. Bass frequencies naturally have higher distortion but it should be possible to keep 300 to 5000 Hz quite low and even, given good drivers (not nessecarily expensive drivers).

 

[mj30250]

I surely wouldn't describe 86dB hitting your ears as quiet, especially at higher frequencies. 86dB continuous would certainly be loud. Would you actually be using these speakers nearfield? If so, you may be fine with either speaker but I would be looking at pairing them with subs, especially the DV3s. Also keep in mind that a given pair of speakers playing at 86dB will distort less than a single speaker at the same output. If you have the ability to take SPL readings at your listening position at whatever volume is typically your loudest, that would be helpful.

 

[TurtlePaul]

86 dB is THX reference level, which you would achieve if you sit 1 meter from the speakers in those graphs. In my mind, that is very loud. Much louder than I ever listen. If you have an iPhone, download the NIOSH Sound Meter app. Point your phone mix at the speakers from your listening position and see what it reads. I rarely listen over 70 dB myself.

 

[a4eaudio]

Numerous studies show that distortion down 40dB is inaudible to MOST humans' ears. And 30dB may be more realistic, but 40dB seems to be the rule of thumb/standard that a lot of people adhere to.

Most people don't find 2nd order harmonic distortion nearly as objectionable as 3rd (and higher odd) order. So, start by looking at whether 3HD is down 40 (or 30) dB.

The standard measurements at 86 dB and 96 dB are not necessarily for continuous listening levels, but also peaks. It is common to allow for 20 dB peaks. So distortion at 86 dB could be audible at continuous listening levels of 76 dB or even 66 dB if the distortion is really too high.

The Kali speaker distortion measures a little higher than ideal but I doubt it matters much for real listening.

 

[dfuller]

86dB/1m free field for one speaker is pretty loud if you're sitting close up. A pair is above 90dB in room.

 

[Sokel]

85dB SPL (C) can very well be 95dB (C) for max and 105-110dB (Z) at peaks with certain types of music.

So I wouldn't go there with speakers/amps that can't handle it.

At the speakers shown above I wouldn't go over 75dB (C) (for a pair) with such works.

 

[kyuu]

I don't really know how to read distortion measurements, but with my limited knowledge in this topic, this looks bad. Near field monitors are supposed to be played quietly, and 86dB's isn't that loud, is it?

86dB is loud. Louder than a garbage disposal, about as load as the sound of a passing truck. If you listen to something at level continuously for 16 hours you can cause permanent hearing loss. Unless you like to really turn your nearfield monitors to ear-ringing levels, you're probably not going to be doing 86dB average at 1 meter. Those distortion charts are what you'll be getting on dynamic peaks of probably 90-95dB listening at 70-75dB average.

If you want speakers that can play very loud without heavy distortion, you're going to have to go bigger and/or more expensive than what you're looking at. But for realistic desktop usage, both speakers are more than fine.

 

[DanielT]

I think it's largely a matter of practice in detecting distortion.

Some tests:

D

'How Much Distortion Can you Hear?'

Jan 19, 2022

An interesting test probably posted before although it didn't come up in a search.

Also try difference sources ie ear phones vs speakers.

• DamianW
• Replies: 32
• Forum: Psychoacoustics: Science of How We Hear

• 

Listening Test

www.klippel.de

When you figure out what you're supposed to hear, listen for clues, in which frequencies it's easiest to detect and so on, you get better at detecting distortion. With practice comes skill, like so much else in life. With Hifi, it can also mean that you need to open a slightly larger wallet when buying speakers.

If you're really good at detecting distortion and don't want to hear it in combination with high SPL, then you really have to open a really big wallet when buying speakers.
(I am not in that category)

Another test:

How Well Can You Hear Audio Quality?

A hi-def test for your ears (and your audio equipment): Listen to these songs and see if you can tell the difference between an MP3 and an uncompressed audio file.

www.npr.org

Note I am not a master of discovering/hearing but I find it fun to try when such tests pop up. Which they do from time to time here at ASR.

Another test: Noise is usually easier to detect than distortion. Test it yourself:

'Audibility thresholds of S/N - signal to noise - test'

Jul 30, 2025

SOTA electronic devices like DACs, preamplifiers and amplifiers have now incredibly high S/N ratio. The limiting factors are in fact S/N of recordings and ambient noise in a listening room. Would you like to check how much S/N you can reliably hear in your listening conditions? Below is a link to test files. They contain a sine 200Hz signal and the same 200Hz sine with added white noise, with rms level of -20dB, -30dB, -40dB, -50dB, -60dB and -70dB relative to 200Hz sine rms level.

https://pmacura.cz/testsignaltonoise.zip

In case you claim to hear -70dB S/N, please add your ABX record.

• pma
• Replies: 88
• Forum: Psychoacoustics: Science of How We Hear

• 

 

[Thomas_A]

Speakers sets the limits and although distortion may not be first priority, there is good reason to find what is causing peaks of distortion and fix them. Evidently, some speakers seem to be fine while quite many show unexplained peaks of distortion. Cabinet resonances? Lack of passive notches in cone break-up? Leaking? Poor stuffing? Bad ports?

 

[boxerfan88]

Oh I forgot to mention… when choosing speakers… do inspect/compare the multi-tone IMD measurement chart if available. Not many speaker manufacturers publish it, generally have to rely on 3rd party measurement.

Again, if IMD can stay <= -40dB for 100-20k at high SPL; the speaker can enter the shortlist.

 

[DVDdoug]

For a "contrary" view on loudness How Loud Is Loud (YouTube video).

I don't know how much distortion is audible to me but usually small woofers can be easily driven into distortion. To a point, the bass harmonic distortion can somewhat make-up for weak deep-bass without sounding bad.

 

[Holmz]

86 dB is THX reference level,

Not really…
From the THX site it says…

• Experience Studio Clarity: THX Certified Receivers reproduce studio Reference Level, 85dB SPL with 20dB of headroom.

So testing and seeing that distortion at 86 dB (on average) means that everything that is coming through above average will have >= to the distortion seen in the plots.
You sort of need the plot at 106dB to show the data at the THX level… Which is 105dB peaks.

which you would achieve if you sit 1 meter from the speakers in those graphs. In my mind, that is very loud.

Not really…
It is certainly not quiet, but maybe you need to use tghe NIOSH app?

#2 who its 1M away?
If one is 3 meters (10’) then it is 76dB, and add some for 2 speakers…
If the room is big, then it gets quieter.

Much louder than I ever listen. If you have an iPhone, download the NIOSH Sound Meter app. Point your phone mix at the speakers from your listening position and see what it reads. I rarely listen over 70 dB myself.

That seems reasonable.
But if you are 10’ away then you only have 5-10dB of headroom before you hit the values in the plots.

If you are listening to modern compressed music, then it should be fine.
If it is classical or something else with a high dynamic range, then maybe it would don’t be quiet so fine.

 

[boxerfan88]

So testing and seeing that distortion at 86 dB (on average) means that everything that is coming through above average will have >= to the distortion seen in the plots.
You sort of need the plot at 106dB to show the data at the THX level… Which is 105dB peaks.

Exactly!!

That’s why I look out for THD chart @ ~95dBSPL. I rarely ever see THD charts at 105dBSPL.

 

[Thomas_A]

Note that music spectral content is not linear. If I would listen at a certain volume that I find ”good” for the music I have chosen, what test do I do to test for the distortion that mimics the spectral content of that? Meaning a more real life distortion test. Following energy of pink noise?

 

[boxerfan88]

Pink noise energy make sense.

From my perspective, if the HF area can pass the <= -40dB THD mark at 95dBSPL from a measured sweep, it should perform as well or better at lower SPL too.

/