• 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!

Questions about loudness in headphone

Samudra1825

Member
Joined
Jan 10, 2022
Messages
23
Likes
2
I kind of get the gist of how to interpret headphone sensitivity and impedance, and using the spec sheet from amp manufacturer to get the power level for the headphone because they have different resistance.

But anyway what i want to know is what that number means? Take for example, i have hifiman sundara and topping L30. From the review by amirm, the L30 gives 1040mW @33ohm, which will give me 5.86Vrms, which means 881mW power for 39 ohms for sundara (also i think it's a good time to say that i learn this from internet and have no bavkground at audio engineering so i appreciate if anyone corrects anything i say wrong) (also this is assuming the amp is limited by its voltage and not current, is an amp ever limited by its current?). With 92db/mW, now the sundara is being run at 121db, probably (at max power from L30 anyway). My question is what exactly is this 121db?

Also there is guideline for safe sound environment from OSHA and NIOSH, is this number represents that? And how does this 121db figure (again if I crank the amp all the way up, and max window volume maybe?) translates to volume in music session? I guess at this stage there is way too many factor because recording themselves i feel sometimes have different loudness.

Also the last thing i want to ask, kinda related kinda don't, what exactly is dynamic range? When i search it i find that it's "the difference between the loudest peak and the quietest part" but isn't the quietest part is silence, so if you listen to 80db song, the dynamic range us 80db?

Anyway, thanks for reading my rambling!
 
(also this is assuming the amp is limited by its voltage and not current, is an amp ever limited by its current?)
Most Amps in High gain are current/voltage limited. That's how you get the highest output power out of your product.

The original L30 is the exception. Above 14Ω, it is limited by gain, unless you can find a DAC with »2Vrms output voltage.

With a standard 2Vrms DAC, the L30 in high gain (+9.5dB) can output 2*10^(9.5/20)=5.97Vrms.

Since Amir measured the Sundara's sensitivity at 94dB SPL/0.318Vrms, we can calculate the the maximum achievable loudness as 94+20*log10(5.97/0.318)=119.5dB(Z) SPL Peak.

That also answers your second question:
My question is what exactly is this 121db?
It's the Z-weighted Peak SPL, assuming that your audio is mastered to 0dBFS Peak.

And how does this 121db figure (again if I crank the amp all the way up, and max window volume maybe?) translates to volume in music session?
Two important aspects of your music files is their dBFS Peak and Crest factor.

Let's say a track is mastered to -0.5dBFS Peak and 15dB Crest factor.

That means that Peak SPL will drop by 0.5dB to 119dB and AVG SPL will be 15dB below that (104dB).

Also the last thing i want to ask, kinda related kinda don't, what exactly is dynamic range? When i search it i find that it's "the difference between the loudest peak and the quietest part" but isn't the quietest part is silence, so if you listen to 80db song, the dynamic range us 80db?
The quietest part is dictated usually by your listening environment.

A 0dB(Z) SPL environment would be mighty impressive!
Most are around 20-40dB(A) SPL, which then reduces the effective playback dynamic range.

Passive noise isolation or active noise cancelling can increase the dynamic range.
 
I kind of get the gist of how to interpret headphone sensitivity and impedance, and using the spec sheet from amp manufacturer to get the power level for the headphone because they have different resistance.

But anyway what i want to know is what that number means? Take for example, i have hifiman sundara and topping L30. From the review by amirm, the L30 gives 1040mW @33ohm, which will give me 5.86Vrms, which means 881mW power for 39 ohms for sundara (also i think it's a good time to say that i learn this from internet and have no bavkground at audio engineering so i appreciate if anyone corrects anything i say wrong) (also this is assuming the amp is limited by its voltage and not current, is an amp ever limited by its current?). With 92db/mW, now the sundara is being run at 121db, probably (at max power from L30 anyway). My question is what exactly is this 121db?
When you want to compare 'loudness' of headphones you need to realize:
A: Manufacturers don't always provide the correct data.
B: Sometimes manufacturers mix up or do not bother to mention if the published number is efficiency (dB/mW) or sensitivity (dB/V) or even if it is max SPL at a specific power rating.
C: Most manufacturers don't specify if their numbers are measured at a certain frequency or noise band.
D: Most manufacturers don't specify measurement method/fixture specifics.

The best way to compare sensitivity (efficiency can not be compared easily as headphone amps provide a voltage and power is drawn) is to calculate dB/V values (when impedance and dB/mW are given.
Also find independent measurements and if needed calculate everything to dB/V.
Also there is guideline for safe sound environment from OSHA and NIOSH, is this number represents that?
These numbers are about exposure time to continuous average levels of 'noise like' sounds. Music is dynamic. The average levels are kind of comparable if they are A weighted.
The weighting is important and one must realize that SPL numbers in measurements are dBSPL measured on a specific fixture.

And how does this 121db figure (again if I crank the amp all the way up, and max window volume maybe?) translates to volume in music session? I guess at this stage there is way too many factor because recording themselves i feel sometimes have different loudness.
Yep... the 121dB number is valid at 1kHz or 400Hz and at max. output of said amp with an input level above 2V to that amp.
Note: You will have immediately thrown off your headphone if you played a tone at that level but may well absolutely love it when subbass is reaching 118dB peaks with some 3dB added to that in the mids and treble. You may not want to do that longer than the duration of one song but still...

Also the last thing i want to ask, kinda related kinda don't, what exactly is dynamic range? When i search it i find that it's "the difference between the loudest peak and the quietest part" but isn't the quietest part is silence, so if you listen to 80db song, the dynamic range us 80db?

There is a 'technical' dynamic range (loudest signal versus noise level) and a dynamic range 'in music' which is VERY different, yet both use the same name and specify it in dB.
Dynamic range in music is the peak level opposite the average level of a song, part of a song or an album.

The dynamic range of the hearing is not 130dB but is around 70-80dB when listening to music. So when you have peaks at 100dB then sounds (within a short time frame) below 30dB, that are not masked, are simply not audible.
It does not matter if the 'electrical dynamic range' is 100dB or higher.
 
Most Amps in High gain are current/voltage limited. That's how you get the highest output power out of your product.

The original L30 is the exception. Above 14Ω, it is limited by gain, unless you can find a DAC with »2Vrms output voltage.

With a standard 2Vrms DAC, the L30 in high gain (+9.5dB) can output 2*10^(9.5/20)=5.97Vrms.

Since Amir measured the Sundara's sensitivity at 94dB SPL/0.318Vrms, we can calculate the the maximum achievable loudness as 94+20*log10(5.97/0.318)=119.5dB(Z) SPL Peak.

That also answers your second question:

It's the Z-weighted Peak SPL, assuming that your audio is mastered to 0dBFS Peak.


Two important aspects of your music files is their dBFS Peak and Crest factor.

Let's say a track is mastered to -0.5dBFS Peak and 15dB Crest factor.

That means that Peak SPL will drop by 0.5dB to 119dB and AVG SPL will be 15dB below that (104dB).


The quietest part is dictated usually by your listening environment.

A 0dB(Z) SPL environment would be mighty impressive!
Most are around 20-40dB(A) SPL, which then reduces the effective playback dynamic range.

Passive noise isolation or active noise cancelling can increase the dynamic range.
When you want to compare 'loudness' of headphones you need to realize:
A: Manufacturers don't always provide the correct data.
B: Sometimes manufacturers mix up or do not bother to mention if the published number is efficiency (dB/mW) or sensitivity (dB/V) or even if it is max SPL at a specific power rating.
C: Most manufacturers don't specify if their numbers are measured at a certain frequency or noise band.
D: Most manufacturers don't specify measurement method/fixture specifics.

The best way to compare sensitivity (efficiency can not be compared easily as headphone amps provide a voltage and power is drawn) is to calculate dB/V values (when impedance and dB/mW are given.
Also find independent measurements and if needed calculate everything to dB/V.

These numbers are about exposure time to continuous average levels of 'noise like' sounds. Music is dynamic. The average levels are kind of comparable if they are A weighted.
The weighting is important and one must realize that SPL numbers in measurements are dBSPL measured on a specific fixture.


Yep... the 121dB number is valid at 1kHz or 400Hz and at max. output of said amp with an input level above 2V to that amp.
Note: You will have immediately thrown off your headphone if you played a tone at that level but may well absolutely love it when subbass is reaching 118dB peaks with some 3dB added to that in the mids and treble. You may not want to do that longer than the duration of one song but still...



There is a 'technical' dynamic range (loudest signal versus noise level) and a dynamic range 'in music' which is VERY different, yet both use the same name and specify it in dB.
Dynamic range in music is the peak level opposite the average level of a song, part of a song or an album.

The dynamic range of the hearing is not 130dB but is around 70-80dB when listening to music. So when you have peaks at 100dB then sounds (within a short time frame) below 30dB, that are not masked, are simply not audible.
It does not matter if the 'electrical dynamic range' is 100dB or higher.
Thank you so much for the answer, sorry I posted this and forgot about it lol. More niche knowledge obtained!
 
Back
Top Bottom