• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Audibility of low frequency distortion in speakers

Status
Not open for further replies.

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
@ctrl recently showed some interesting data regarding the masking effect which limits the audibility of low frequency harmonic distortion in speakers

1607673876680.png


The effect of the masking does change with frequency and fundamental amplitude. Higher distortion is required before it becomes audible with higher fundamental level. Please note the remark about the 2nd 3rd harmonic value of 10%. This appears to be a test equipment error, it would be much lower.

Also note that these are a test stimulus of a pure tone. It is easier to hear the distortion with this than with a music signal.

To investigate further I decided to perform a similar test. The following is just a start, further tests will follow.

I replicated (as far as possible) the distortion signature of a typical small speaker playing at 96dB SPL 1m. Luckily the free REW software has functionality to add harmonic distortion to signals so actually an easy thing to do and something any of you can try for yourselves.

1607674300217.png


A tone at 78Hz and 85dB /1m was played and the distortion levels for harmonics 2 to 5 were measured and set using an earthworks M23 mic.

The distortion was switched on and off several times and the FFT and sound is captured in the video below.

Its best to set the playback to HD and full screen so you can see the numbers.



So can people hear the distortion?

Try playing it back at low through to high levels. If you go up to 96db please only do it for a short time.
.


.
 
Last edited:

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,741
Likes
16,174
I can of course hear easily the distortions you added there, so I replicated your settings at my REW and enabled each component one by one, the most intense to hear is the 3rd and the lowest intense is the 2nd but it also changes the tonality a bit.
The question for me though is how we define such as audible, just the tonality change like in this case or being perceived as "sound quality deteriorating distortions".
I am sure also that with usual music material the audibility as distortion will be quite lower.
 

mcdn

Addicted to Fun and Learning
Forum Donor
Joined
Mar 7, 2020
Messages
559
Likes
780
Wait, can you do this for any speaker that Amir's posted distortion measurements for?
 
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Wait, can you do this for any speaker that Amir's posted distortion measurements for?
For sure, but of course it should be borne in mind that single tone is not representative of what you may hear with music. We can move on to those sorts of tests later.

To start we can get an informal idea of what levels people find audible with tones.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague

So can people hear the distortion?

.

This is of course extremely audible on a sine tone as used (with music it is in question and depending on music sample).

It is also possible to make recordings at different SPL levels and then to remove the background noise in CoolEditPro or in Audacity, which is for free. Then to make a DBT with a pure sine matched in level. The noise removal procedure is surprisingly effective and the task is then not so trivial.
 
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
I can of course hear easily the distortions you added there, so I replicated your settings at my REW and enabled each component one by one, the most intense to hear is the 3rd and the lowest intense is the 2nd but it also changes the tonality a bit.
The question for me though is how we define such as audible, just the tonality change like in this case or being perceived as "sound quality deteriorating distortions".
I am sure also that with usual music material the audibility as distortion will be quite lower.
Thats what you would expect as tones nearer the masker (the fundamental) are masked more.
 
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
This is of course extremely audible on a sine tone as used (with music it is in question and depending on music sample).

It is also possible to make recordings at different SPL levels and then to remove the background noise in CoolEditPro or in Audacity, which is for free. Then to make a DBT with a pure sine matched in level. The noise removal procedure is surprisingly effective and the task is then not so trivial.
We can bypass the microphone recording altogether, this was just done so people could see it was a real speaker playing and the distortion levels without the added distortion
 

mcdn

Addicted to Fun and Learning
Forum Donor
Joined
Mar 7, 2020
Messages
559
Likes
780
I'm convinced low frequency distortion is much more audible than commonly assumed, but also that it doesn't sound like distortion of higher frequencies. Any way to investigate this in more detail would be very welcome.
 

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,638
Likes
3,599
Location
Sweden, Västerås
Masking only works if there is a mask ? With music is it less audible that's my question ? , some other member here pointed out that most bass instruments are in fact more overtones than fundamentals ? Maybe on some very clean electronic music you get high auditability for this .
Is this sine masking enough ?

And subwoofers ! I'm thinking that localizing sub's are more to their distortion and/or port noises than fundamental tone.
A bad sub is audible all the time ?
 
  • Like
Reactions: GDK

bennybbbx

Addicted to Fun and Learning
Joined
May 27, 2020
Messages
691
Likes
124
Location
germany
this is only 10% distortion. it sound as if it was the main tone and then a additional frequency with same level. maybe you do a test with several distortions and not tell which distortion it is and do a anonymous poll vote in the forum. so can vote which is the first distortion that people can hear
 

mcdn

Addicted to Fun and Learning
Forum Donor
Joined
Mar 7, 2020
Messages
559
Likes
780
@Mnyb that's what was confusing me, where is the masking tone? For dummies like me I'd love to have a sequence of test tracks like:
1) pure tone
2) pure tone with distortion profile X
3) pure tone with masking tone
4) pure tone with masking tone and distortion profile X
5) music programme (@mitchco likes a Rebecca Pidgeon track I think)
6) music programme with distortion profile X

Where we could try various distortion profiles to see how the literature on low frequency distortion audibility and masking effects translates to our setups. That wouldn't produce any new scientific results as @bobbooo is right to point out, but it would generate some interesting data points.
 
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Better to skip the flawed, unscientific polls and do proper blind ABX tests of distorted music files generated using Distort with their originals using Foobar's ABX Comparator which has verifiable logs.
Totally agree and I will be moving on to this. It was just a simple demo and not designed to draw any scientific conclusions from. Just to give an idea to the effect on the sound. Many may not have any idea what that distortion may sound like.

It was really just a cheeky question as of course with a simple tone it would be abundantly audible to all.
 
Last edited:
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
Masking only works if there is a mask ? With music is it less audible that's my question ? , some other member here pointed out that most bass instruments are in fact more overtones than fundamentals ? Maybe on some very clean electronic music you get high auditability for this .
Is this sine masking enough ?

And subwoofers ! I'm thinking that localizing sub's are more to their distortion and/or port noises than fundamental tone.
A bad sub is audible all the time ?
The mask is the fundamental in this example.

1607686772703.png


Yes it certainly less audible with music as I mentioned above.

I agree about sub localisation.
 
Last edited:
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
@Mnyb that's what was confusing me, where is the masking tone? For dummies like me I'd love to have a sequence of test tracks like:
1) pure tone
2) pure tone with distortion profile X
3) pure tone with masking tone
4) pure tone with masking tone and distortion profile X
5) music programme (@mitchco likes a Rebecca Pidgeon track I think)
6) music programme with distortion profile X

Where we could try various distortion profiles to see how the literature on low frequency distortion audibility and masking effects translates to our setups. That wouldn't produce any new scientific results as @bobbooo is right to point out, but it would generate some interesting data points.
Not Spanish Harlem ? NNOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
 
OP
March Audio

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
this is only 10% distortion. it sound as if it was the main tone and then a additional frequency with same level. maybe you do a test with several distortions and not tell which distortion it is and do a anonymous poll vote in the forum. so can vote which is the first distortion that people can hear
Hi Benny

It was actually intended to be about 16% THD. The profile was copied from a typical small speaker test.

1607687106590.png


-16.2dB = 15.5%
-22dB = 8%
-17.6dB = 13.2%
-36dB = 1.6%
-38.1 = 1.25%
 

mcdn

Addicted to Fun and Learning
Forum Donor
Joined
Mar 7, 2020
Messages
559
Likes
780
The mask is the fundamental in this example.

View attachment 98485

A naive reading of that graph would be "play these exact tones aparft from those at 400Hz and 450Hz and you won't be able to tell it apart from a pure 100Hz tone". Is that what the source claims?
 
Status
Not open for further replies.
Top Bottom