[megabigeye]

I have what I think is another simple question that I think I have the correct answer to already... But I see (what I believe to be) the incorrect answer so frequently repeated that I start to doubt my own understanding.

Do changes in power and voltage have any effect on a transducer/amplifier system other than changing loudness? I.E., more power = more louder; less power = less louder.

To be a bit more specific, there's one poster in particular that insists that the HD 600/650 requires more power and voltage than most amplifiers can produce and claims that those headphones can get loud enough with low power amplifiers (phones, laptops, etc.), but that they'll sound somehow lacking. My own (entirely anecdotal, entirely unscientific) experience does not align with this, but my own experience isn't what I'm really interested in. What I guess I'm after is whether there's an objective reason that a change in voltage and/or power could cause a change in anything other than sound pressure.

By my thinking, if you know the impedance and sensitivity/efficiency of the headphone and the power of the amplifier into a specific load, then the equation P = V²/R should pretty faithfully tell you whether or not an amp can produce sufficient power (assuming that it's otherwise properly designed and functioning), and beyond a certain limit more power is completely useless.
By the other poster's thinking, more power is almost always better.

There are two possible caveats to my thinking that I could come up with:
1) that the HD 600/650's impedance actually tops out at about 500Ω at 100Hz, and an amp that can produce sufficient power at 300Ω may not be able to do so at 500Ω. I'd imagine this would make the headphone sound a bit "thin" in the bass. A higher power amplifier wouldn't run into this problem.
2) that an amp with high THD at maximum power output could audibly distort as it struggles with a 500Ω load. A higher power amplifier wouldn't start distorting so soon.

Basically, I'd like to know if I'm right or if there's any validity to what the other poster is saying before I start trying to correct him.

Also, I'm sorry to see that @gregorio is banned from posting. I've always found his knowledge and explanations to be especially helpful. Hopefully he'll be back soon!

 

[KeithPhantom]

Do changes in power and voltage have any effect on a transducer/amplifier system other than changing loudness?

Mmm, not for what I know (unless you count excessive power and destroying your transducers as a change). Maybe THD can increase, but that's depending on the design and the tolerances for different power ratings.

 

[megabigeye]

Mmm, not for what I know (unless you count excessive power and destroying your transducers as a change). Maybe THD can increase, but that's depending on the design and the tolerances for different power ratings.

Thanks for confirming my thoughts.

What about my two caveats? Sounds like you agree with the second one, anyway. I used to have a FiiO E11 that was loud enough with the HD 650, but that sounded pretty awful in the bass regions. I assume it was because of a similar reason.

 

[KeithPhantom]

that the HD 600/650's impedance actually tops out at about 500Ω at 100Hz, and an amp that can produce sufficient power at 300Ω may not be able to do so at 500Ω. I'd imagine this would make the headphone sound a bit "thin" in the bass. A higher power amplifier wouldn't run into this problem.

This is only valid if the damping factor is too low. Actually, the increase of resistance will make the headphone draw less power in this frequency range. Output voltage is pretty much the same.

that an amp with high THD at maximum power output could audibly distort as it struggles with a 500Ω load. A higher power amplifier wouldn't start distorting so soon.

Because it has higher power doesn't mean that it measures better. Also, at maximum power, there are amps that do not clip or have a sudden rise of THD. Remember, a 500-ohm load with a high sensitivity is pretty easy to drive. I would be more worried if it was low sensitivity low impedance, that's the area most amplifiers show their maximum THD-to-power ratio.

 

[castleofargh]

Things can rapidly become complicated because getting a more powerful amp isn't always just a matter or turning the "power up" knob. If we consider everything else equal and that we already had nominal behavior, then extra power on headphone amps would just mean increasing the voltage gain and making the headphone louder. That may or may not lead to significant quality drop in the signal(noise level, distortion). What's important and was mentioned by Keith, is that a power specification might not inform us about fidelity.
In practice, you'll usually have 2 different amps with way more differences than just the max power value. That may takes us away from simple Power=Loudness and everything else is identical.
But at the same time, it kills the argument of people telling you that the headphone "scales up" with more power. Because they randomly picked a correlation between power and subjective sound quality. But they usually have no idea about the objective sound quality. They also cannot establish causality because of all the other differences between amps that they conveniently ignored to serve their rational.

Yes, if an amp was designed for higher max power, it's possible that some components will be better, that it will measure better, that it will have higher slew rate...
But it's also possible for it to measure worst from the get go, because an amp with a much higher gain is usually designed with low sensitivity in mind. And it can easily have a higher noise floor for example. Or just be a PITA when it comes to set the listening level. Maybe thank to the uber gain, you'll end up having to turn your volume down into an area with important channel imbalance. There are so many variables involved, it's easy to make up an argument going in whichever direction you like. But in fact, 2 different amps can't be fully characterized by the max power specification into a given load. We need that variable to make sure we'll get enough for our needs, but it's only one value about one issue.

To know what happens with a 500ohm load, you need measurements into that load. 300ohm specs might not always be enough to extrapolate the 500ohm behavior. It usually will, but not always. With a load(headphone,IEM) that varies in impedance over frequency, it would be nice to have specs for both ends of the impedance range. But we're not often that lucky.

 

[bigshot]

The easiest thing is to avoid buying equipment that is persnickety. Difficulty has no relation to fidelity. As Devo says, "Why believe in things that make it tough on you?"

 

[KeithPhantom]

A trick to finding the power you need for a specific transducer is to have a peak SPL in mind, using the sensitivity/efficiency and the impedance you can find the power needed for that SPL. If the maximum peak of your loudest song/movie that you listen at the maximum level you will listen is below what the amplifier can supply, your amplifier is powerful enough to reproduce any signal at those values without worrying about distortion or other factors. More power will not benefit you other than raising the volume to uncomfortable levels. Music isn't composed of only peaks, your "average" listening level is measured in RMS, not in peak levels, we're only use peak levels so you do not distort the amplitude of these, and also you cover the RMS level under this method.

Another trick I can give you is that for every 10 dB increase in SPL you need 10x the power. If the sensitivity is 93 dB/1mW, 103 dB comes at 10mW, 113 dB at 100mW, and 123 dB at 1 W.

 

[havarduf]

I have (maybe) a basic question.

What determines the amount of audible background noise on a recording? Does it depend solely on expensive recording equipment being used, or is it a skill of the producer/engineering team to eliminate noise post-production?

Disappointing to purchase a record (CD) only to find out it has a relatively loud background "hiss" on all of the tracks.

Example of a completely (at least in my ears!) silent track:

Anette Askvik - Liberty

 

[VNandor]

What determines the amount of audible background noise on a recording? Does it depend solely on expensive recording equipment being used, or is it a skill of the producer/engineering team to eliminate noise post-production?

It also depends on how noisy the recording studio is.
The more channel there is to mix the harder it becomes to manage the noise. You can have 50 individual tracks that doesn't sound too bad in themselves but once you mix them together the noisefloor of the mixed track will be higher. Common processing that increases the already existing noise is compression and limiting. "Harmonic exciters" and saturators are adding their own noise. These processings are typically used in the individual tracks as well (except the limitng) so they can add up to something noticable eventually. So it isn't as much about removing the noise that's already there, it's about not amplifying and adding to what is already there.

 

[megabigeye]

Hooray! Somebody else asked a basic question! I was kind of disappointed more people weren't using this as a kind of introductory understanding resource. So I'm glad someone is here asking questions!

I feel like I've had a ton of questions in the last couple months, but I can only think of one right now:
For calculating required power for a headphone, until recently I've only ever seen 2x power per +3dB volume. But the other day I saw 10x power per +10dB.
I'm wondering which is correct (if either is)?

 

[castleofargh]

For calculating required power for a headphone, until recently I've only ever seen 2x power per +3dB volume. But the other day I saw 10x power per +10dB.
I'm wondering which is correct (if either is)?

10*log2=3.01
10*log10=10
Both are good.

 

[megabigeye]

10*log2=3.01
10*log10=10
Both are good.

Yeah, I noticed that they're close and I figured that for all intents and purposes that it probably doesn't much matter, but my OCD wants to know which is technically correct. I figure that a physicist or mathematician probably wouldn't settle for eh, close enough.

 

[sander99]

10*log2=3.01
10*log10=10
Both are good.

Yeah, I noticed that they're close and I figured that for all intents and purposes that it probably doesn't much matter, but my OCD wants to know which is technically correct. I figure that a physicist or mathematician probably wouldn't settle for eh, close enough.

The general exact rule:
If you multiply the power by n, the sound pressure level increases with 10*log(n) dB.

So you could say:
10*log(10) equals 10 exactly, so 10 times power for 10 dB extra is exact.
10*log(2) does not equal 3 exactly, so 2 times power for 3 dB extra is not exact (but very close).

On the other hand (now I am going to nitpick): physical quantities often have limited precision, often expressed by the number of significant digits.
If a physicist says U = 2.0 V it actually means 1.95000... V <= U < 2.05000... V, and if you say [Edit: if you multiply the power by 2 then] there will be 3 dB increase in SPL it could be interpreted as: the increase in SPL will be at least 2.5 dB and less than 3.5 dB, and that would be correct!

 

[megabigeye]

Following... Following... Confused? Isn't "2.0V" shorthand for "2.00000...V"? I guess not, if you're being really precise (which I guess is what I was asking for).

Thanks!

(I also now see that this is basically what @castleofargh also said, but I didn't quite understand. Maths is hard.)

 

[havarduf]

More basic questions from me...

Some songs/albums are considered to be of 'audiophile' quality and are often used as reference. What makes these records special as opposed to other supposedly high quality produced material made by reputable artists?
Gear quality? $$$?
The experience and skill of the producer/engineering team?
Mastering?

To mention a few examples: Nils Lofgren, Dire Straits, Miles Davis.