[theaudiologist1]

So today I made my research on amp power and I want to know:

1) Is it true that if you use your amp at max volume at high gain, even if it drives the headphones loud, it will cause distortion, higher THD% and clipping since the amp is using its full power? It is said that a stronger amp outputting the same power to a headphone while not being at 100% is better than an amp using all its power? And how can I tell if it's distorting or clipping? I notice the lower end isn't as loud but I don't know it that's the amp or the recording.

2)This website:
https://nwavguy.blogspot.com/2011/09/more-power.html
claims 115dB peak SPL is preferred for dynamic recordings (which probably means classical).

"The research indicates the average maximum level should be at least 85 dB, and with classical music, that puts the peak level up to 30 dB higher at a worst case 115 dB)."

How true is this and how much of a difference does it make? My amp drives my headphones to 108-112dB SPL peaks (maybe more on high gain).

3)Is it true that a weak amp will also skew the frequency curve of a headphone? Someone mentioned it in a thread.

 

[theaudiologist1]

Generally, symphony recordings have the largest dynamic ranges of all music genres, particularly the later romantic period symphonic composers such as Mahler, the post romantics such as R. Strauss and some of the impressionists and later composers, such as Ravel and Stravinsky.

G

Agreed. Mahler's music, specially his 8th, are some of the most dynamic recordings I ever heard.

 

[71 dB]

So today I made my research on amp power and I want to know:

1) Is it true that if you use your amp at max volume at high gain, even if it drives the headphones loud, it will cause distortion, higher THD% and clipping since the amp is using its full power? It is said that a stronger amp outputting the same power to a headphone while not being at 100% is better than an amp using all its power? And how can I tell if it's distorting or clipping? I notice the lower end isn't as loud but I don't know it that's the amp or the recording.

2)This website:
https://nwavguy.blogspot.com/2011/09/more-power.html
claims 115dB peak SPL is preferred for dynamic recordings (which probably means classical).

"The research indicates the average maximum level should be at least 85 dB, and with classical music, that puts the peak level up to 30 dB higher at a worst case 115 dB)."

How true is this and how much of a difference does it make? My amp drives my headphones to 108-112dB SPL peaks (maybe more on high gain).

3)Is it true that a weak amp will also skew the frequency curve of a headphone? Someone mentioned it in a thread.

1) Amps always generate some distortion, but at lower levels it is typically totally inaudible (unless we are talking about something like tube amps). When the power level approaches the maximum power level, distortion starts to rise rapidly. So, an amp may give 50 watts, but with large distortion (e.g. 2 % THD) and 40 watts at a low distortion rate such as 0.05 % TDH. As a rule to thumb 0.1 % THD or lower is safe/inaudible, but the audibility depends a lot on the frequency and distortion spectrum (odd/even harmonic etc.) If you can't tell if the amp is distorting or clipping you probably don't have a problem. What you can hear is a problem.

2) Peaks of 115 dB is very very loud (almost pain threshold!) and endangers hearing. Peaks of 100 dB allow very loud music listening experience. Dynamic music doesn't mean the peaks must make you deaf. It means the quiet parts are QUIET!

3) Too high output impedance together with uneven headphone impedance curve causes further curved frequency response, because the amp voltage is divided differently over the amp output impedance and headphone (+cable) impedance. Let's assume the output impedance of the amp is 33 Ω and the minimum and maximum impedances of the headphone are 25 Ω and 72 Ω. The level diffrence caused by the amp output impedance compared to a zero or very small output impedance is:

20 * Log10 ((72*(25+33)) / (25*(72+33))) = 4 dB.​

If the amp output impedance was just 1 Ω, we would have:

20 * Log10 ((72*(25+1)) / (25*(72+1))) = 0.2 dB.​

If the amp output impedance was just 33 Ω, but the headphone impedance was flatter (min 45 Ω, max 52 Ω), we'd have:

20 * Log10 ((52*(45+33)) / (45*(52+33))) = 0.5 dB.​

 

[VNandor]

That's the problem: when listening to classical, in order to hear the quiet parts I have to put my amp on hi gain, max volume to hear them, and then the loud parts hurts my ears and I have to turn down the volume. Don't know how many dB the peaks are.

If the loud parts are already getting too loud and you have to adjust the volume during playback, that's not really the the amp's fault. Either you listen in a too noisy environment (even PC fans or HDDs can be too loud in this context) so you can't turn down the volume as much as you should without losing detail in the quiet parts, or if you listen in a quiet room but the music is still too dynamic, that's really just bad mastering.

Is it true that if you use your amp at max volume at high gain, even if it drives the headphones loud, it will cause distortion, higher THD% and clipping since the amp is using its full power? It is said that a stronger amp outputting the same power to a headphone while not being at 100% is better than an amp using all its power?

This isn't always true. For example a tube amp would most likely create more distortion at any volume level compared to an almost maxed out solid state amp even if the solid state amp's max output power is lower compared to the tube amp.

 

[theaudiologist1]

1) Amps always generate some distortion, but at lower levels it is typically totally inaudible (unless we are talking about something like tube amps). When the power level approaches the maximum power level, distortion starts to rise rapidly. So, an amp may give 50 watts, but with large distortion (e.g. 2 % THD) and 40 watts at a low distortion rate such as 0.05 % TDH. As a rule to thumb 0.1 % THD or lower is safe/inaudible, but the audibility depends a lot on the frequency and distortion spectrum (odd/even harmonic etc.) If you can't tell if the amp is distorting or clipping you probably don't have a problem. What you can hear is a problem.

2) Peaks of 115 dB is very very loud (almost pain threshold!) and endangers hearing. Peaks of 100 dB allow very loud music listening experience. Dynamic music doesn't mean the peaks must make you deaf. It means the quiet parts are QUIET!

3) Too high output impedance together with uneven headphone impedance curve causes further curved frequency response, because the amp voltage is divided differently over the amp output impedance and headphone (+cable) impedance. Let's assume the output impedance of the amp is 33 Ω and the minimum and maximum impedances of the headphone are 25 Ω and 72 Ω. The level diffrence caused by the amp output impedance compared to a zero or very small output impedance is:

20 * Log10 ((72*(25+33)) / (25*(72+33))) = 4 dB.​

If the amp output impedance was just 1 Ω, we would have:

20 * Log10 ((72*(25+1)) / (25*(72+1))) = 0.2 dB.​

If the amp output impedance was just 33 Ω, but the headphone impedance was flatter (min 45 Ω, max 52 Ω), we'd have:

20 * Log10 ((52*(45+33)) / (45*(52+33))) = 0.5 dB.​

2) so you're saying 100dB is enough for peaks? what about for a classical recording with a dynamic range of 40dB? Wouldn't that make the very quiet parts hard to hear at 60dB? What difference will a headroom of 110/115dB bring compared to a headroom of 100dB?

3) My DAC/AMP outputs 2.2Ohms in single-ended and 4.4Ohm in balanced mode and my two headphones have impedances of 470Ohm and 35Ohms. Will this problem affect my headphones?

Another thing: today I learned my DAC's max input is 1.5V (from source I guess), will this affect the amp output power?

If the loud parts are already getting too loud and you have to adjust the volume during playback, that's not really the the amp's fault. Either you listen in a too noisy environment (even PC fans or HDDs can be too loud in this context) so you can't turn down the volume as much as you should without losing detail in the quiet parts, or if you listen in a quiet room but the music is still too dynamic, that's really just bad mastering.

Sometimes I have the room window open, and my headphones are open headphones. Maybe that causes interference.

This isn't always true. For example a tube amp would most likely create more distortion at any volume level compared to an almost maxed out solid state amp even if the solid state amp's max output power is lower compared to the tube amp.

I see. So it all depends on the amp and solid state amps have a lower distortion than tube amps?

 

[bigshot]

The average quiet living room has a noise floor of about 30dB. With a fairly dynamic track with a dynamic range of 40dB, That would allow you to boost the loudest peaks up to a comfortable level of 70dB and still hear the quietest parts just above 30dB. You'd have room to push it up to 80dB for a very loud listening level too.

40dB is a comfortable dynamic range for music. When you get into peak levels above that, you have to listen loud to hear everything. That's OK with Wagner, but maybe not so much for Haydn.

 

[gregorio]

1) Is it true that if you use your amp at max volume at high gain, even if it drives the headphones loud, it will cause distortion, higher THD% and clipping since the amp is using its full power? It is said that a stronger amp outputting the same power to a headphone while not being at 100% is better than an amp using all its power? And how can I tell if it's distorting or clipping? I notice the lower end isn't as loud but I don't know it that's the amp or the recording.
2)This website:
https://nwavguy.blogspot.com/2011/09/more-power.html
claims 115dB peak SPL is preferred for dynamic recordings (which probably means classical).
"The research indicates the average maximum level should be at least 85 dB, and with classical music, that puts the peak level up to 30 dB higher at a worst case 115 dB)."
How true is this and how much of a difference does it make?
3)Is it true that a weak amp will also skew the frequency curve of a headphone? Someone mentioned it in a thread.

1. Yes, driving an amp or headphone at max power will likely introduce audible distortion. As a decent rule of thumb, calculate what amp you need on the basis that you won't use it with a setting higher than about 75%. This will avoid the possibility of audible distortion from any half decent amp, although not necessarily from tube amps.

2. That's not true. The research indicates average max of 85dB for films in a cinema but when listening to audio in a home environment, the research indicates an average maximum level of 78dB, which is why 78dB is specified for calibrating TV loudness. Furthermore, these levels are referenced to -20dBFS, which means the maximum peak level in a cinema is 105dB (not 115dB) and in a consumer environment (TV) the max peak level would be 97dB. However, it might not be a bad ideal to aim for an amp that could output levels up to 115dB on the basis of the previous point, that you will always use it at settings lower than about 75% of that.

3. It can be true, @71 dB explained it.

Another thing: today I learned my DAC's max input is 1.5V (from source I guess), will this affect the amp output power?

No, what could affect your amp is the output (not the input) of your DAC, as obviously, the lower your DAC's output the more gain your amp will need to apply. This generally wouldn't be of any concern with a standalone DAC though.

G

 

[71 dB]

2) so you're saying 100dB is enough for peaks? what about for a classical recording with a dynamic range of 40dB? Wouldn't that make the very quiet parts hard to hear at 60dB? What difference will a headroom of 110/115dB bring compared to a headroom of 100dB?

60 dB SPL isn't generally "hard to hear" in reasonable listening oonditions. Spoken communication happens typically at 60-70 dB and people can mostly understand each other. If the music has 40 dB of dynamic range, it means most of the music really is rather quiet. Otherwise the peaks get uncomfortable, even painfully loud. The difference is like similar to the difference of cars with stop speeds of 100 mph and 115 mph. You can go faster/louder, but should you? You may have a system that can give you 120 dB peaks, but when you listen to music you set the volume to comfortable levels and the peaks are maybe 100 dB.

3) My DAC/AMP outputs 2.2Ohms in single-ended and 4.4Ohm in balanced mode and my two headphones have impedances of 470Ohm and 35Ohms. Will this problem affect my headphones?

The 470 Ω model should be fine, because the impedance is over 100 times bigger than the amp output impedance. The 35 Ω (earlier you said 32 Ω) was closed, right? Closed headphones tend to have very flat impedance curve meaning higher output impedance isn't an issue so you should be fine.

Another thing: today I learned my DAC's max input is 1.5V (from source I guess), will this affect the amp output power?

( brainfart pointed out by gregorio extracted with a brainfart extractor fan )

No.

Sometimes I have the room window open, and my headphones are open headphones. Maybe that causes interference.

Certainly! That would explain why 60 dB is hard to hear. You need much quieter listening conditions when listening to dynamic classical music! Listen to much less dynamic music when your window is open. Closed headphones can help too.

I see. So it all depends on the amp and solid state amps have a lower distortion than tube amps?

Yes. The distortions tube amps generate is the reason why people bother to use them anymore. Some people love the distortion because it makes the sound "warmer". On technical point of view solid state amps are far superior, but sometimes people don't want technically correct answers.

 

[bigshot]

glad to see some answers being listened to

 

[gregorio]

1.5 V sounds low, because line level signals are typically 2 V and can be even more than that.

Don't forget he's talking about a DAC, so the input signal is an eye pattern electrical signal, not a line level analogue signal. So in the case of say USB 2, the input signal would be about 400mV.

G

 

[71 dB]

Don't forget he's talking about a DAC, so the input signal is an eye pattern electrical signal, not a line level analogue signal. So in the case of say USB 2, the input signal would be about 400mV.

G

Thanks for this correction, my bad! Some sort to brainfart.
I totally forgot the DAC-part and only thought about an amp taking in analog line lever signal.

 

[theaudiologist1]

Don't forget he's talking about a DAC, so the input signal is an eye pattern electrical signal, not a line level analogue signal. So in the case of say USB 2, the input signal would be about 400mV.

G

Yes. My DAC/AMP has a microUSB which only has USB 2.0.

glad to see some answers being listened to

Glad to hear I'm not a burden at least

60 dB SPL isn't generally "hard to hear" in reasonable listening oonditions. Spoken communication happens typically at 60-70 dB and people can mostly understand each other. If the music has 40 dB of dynamic range, it means most of the music really is rather quiet. Otherwise the peaks get uncomfortable, even painfully loud. The difference is like similar to the difference of cars with stop speeds of 100 mph and 115 mph. You can go faster/louder, but should you? You may have a system that can give you 120 dB peaks, but when you listen to music you set the volume to comfortable levels and the peaks are maybe 100 dB.


The 470 Ω model should be fine, because the impedance is over 100 times bigger than the amp output impedance. The 35 Ω (earlier you said 32 Ω) was closed, right? Closed headphones tend to have very flat impedance curve meaning higher output impedance isn't an issue so you should be fine.

It's actually 35 Ohms. 32 Ohms was my guess before I checked the official specs (It's an ATH-MSR7). Using the 1:8 ratio on the 4.4 Ohm balanced out means the headphone needs to be 35.2 Ohms, almost exactly the same as my MSR7. Is that OK?

Certainly! That would explain why 60 dB is hard to hear. You need much quieter listening conditions when listening to dynamic classical music! Listen to much less dynamic music when your window is open. Closed headphones can help too.

Thanks. It's just that classical sounds way better with open headphones.

Yes. The distortions tube amps generate is the reason why people bother to use them anymore. Some people love the distortion because it makes the sound "warmer". On technical point of view solid state amps are far superior, but sometimes people don't want technically correct answers.

One reason people like vinyls.

1. Yes, driving an amp or headphone at max power will likely introduce audible distortion. As a decent rule of thumb, calculate what amp you need on the basis that you won't use it with a setting higher than about 75%. This will avoid the possibility of audible distortion from any half decent amp, although not necessarily from tube amps.

Thanks. I put my DAC/AMP on high gain max volume (only for dynamic classical) and the sound appears clean to me. I don't know how to detect distortion. There is some noise when I change the volume on my DAC but nothing noticable fromm the music.

 

[71 dB]

1. It's actually 35 Ohms. 32 Ohms was my guess before I checked the official specs (It's an ATH-MSR7). Using the 1:8 ratio on the 4.4 Ohm balanced out means the headphone needs to be 35.2 Ohms, almost exactly the same as my MSR7. Is that OK?

2. Thanks. It's just that classical sounds way better with open headphones.

3. One reason people like vinyls.

1. The 1:8 ratio is very general and outdated (headphones and amps driving them are different beasts than when that rule was created). To be on the safe side with any headphone model out there one needs something like a 1:20 ratio rule, but much better is to know the impedance curve of the headphone and calculate how small output impedance is needed. In case of ATH-MSR7 (see the link below) the impedance is "ruler flat" meaning even output impedance of hundreds of ohms doesn't cause frequency response issues. So, 4.4 Ω is fine. So is 44 Ω or even 440 Ω! Flat impedance curve is one strength of closed headphones.

https://diyaudioheaven.wordpress.com/headphones/measurements/audio-technica/ath-msr7/

2. Perhaps true.

3. Yes, at least one of the reasons.

 

[71 dB]

Glad to hear I'm not a burden at least

You are not a burden. People who reject information and insist on knowing better when they don't are a burden.

 

[gregorio]

Yes. My DAC/AMP has a microUSB which only has USB 2.0.

It's a bit weird they would give the spec of max input signal level. Input level is defined by the USB 2.0 specification, so either a USB 2.0 DAC can easily handle that spec or it won't work.

G