Head-Fi.org › Forums › Equipment Forums › Sound Science › Headphones NEEDING Power
New Posts  All Forums:Forum Nav:

Headphones NEEDING Power - Page 3

post #31 of 89
Quote:
Originally Posted by Brooko View Post

Thanks Mike

It doesn't need to be perfect for her.  She's all about the music - especially when you see her with eyes closed, K701 on, totally swept away.  If the X3 will get her close to that without major distortion or clipping, it should be the ideal solution for her smily_headphones1.gif

Really appreciate the help beerchug.gif
That's what it's all about.
post #32 of 89
Thread Starter 
Quote:
Originally Posted by mikeaj View Post

I checked FiiO's website, and that says only 220 mW into 32 ohms for the E17.

 

Even at that level, it's around 2.65 V rms into 32 ohms, so it should at least be that much for ~62 ohms and not really have performance issues for those. So more than 115 mW or so max. Depends.

 

Wait, I should have just checked the X3 page. 540 mW into 16 ohms, so 3 V no problem. So say 145 mW into 62 ohms, very similar to ~115 mW though anyway. If it can handle 16 ohms like that, it has plenty of current and probably won't mind 62 ohms.

 

On the other hand, InnerFidelity measurements show lower sensitivity than AKG's specs, more like 100 dB SPL / 1 V (88 dB SPL / 1 mW). Crazily enough, that means that 145 mW is not quite enough for 110 dB SPL peaks, which I think are realistic if you want to listen at a high volume to certain classical recordings with very low average volume. It may realistically be enough. Depends on tastes.

 

Or actually, their second sample, was slightly more sensitive and would be driven to just over 110 dB SPL when maxed out there. So let's just say that it's in that range.

I don't quite understand how to calculate the required power/voltage/current for a headphone.

 

I can't post a link to his blog but NwAvGuy posted in his power article:

Quote:
120 dB SPL is also the level at which even short term exposure can cause permanent hearing impairment. Studies have shown even sustained average levels above 85 dB SPL can cause hearing damage. For more on these thresholds see Sound Pressure Levels. 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). For more typical music peak levels of 110 dB SPL are more reasonable.

That makes sense since he also stated earlier:

Quote:
Wide Dynamic Range Classical: –18 dB to –30 dB

(85 dB SPL + 30 dB SPL = 115 dB SPL)

 

So for someone who has hearing damage, 85 dB SPL might not be enough. So maybe 90 dB SPL is adequate, and that's what Tyll measures at Innerfidelity, the K 701 (sample B from the downloads) requires 0.238 Vrms to reach that SPL with a measured impedance of 67 Ω at 1 kHz, or 1.20 mW of power.

 

With 90 dB SPL, or a theoretical maximum of 120 dB SPL with classical for the quietest parts of a song, 1.20 mW * 2^10 = 1228.8 mW or 1.2288 W of power is needed.

 

If Power = (Vrms^2) / impedance

Then sqrt(1.2288 * 67) = 9.07 Vrms at 67 Ω is needed

 

First of all, are these calculations and assumptions correct? XD

 

 

 

On the X3's specification page it says:

Quote:
MAX output voltage     > 8 Vp-p

From my understanding, 1 Vrms = 1 * 2 * sqrt(2) = 2.8284 Vp-p

So 8 Vp-p / 2.8284 Vp-p = 2.8284 Vrms

 

2.828 Vrms < 9.07 Vrms

 

Second of all, are these calculations correct?

 

 

Now this is where I'm having a hard time interpreting things. Based on this, the X3 shouldn't provide enough voltage to reach 120 dB SPL. Does this mean the X3 won't supply enough volume to reach the desired level?

 

 

 

A JDS Labs C5 portable headphone amplifier has a specification of:

From JDS Labs's C5 blog post:

Quote:
Power Supply 14.0 Vpp

Is power supply the same as the maximum output voltage? If so:

So 14 Vp-p / 2.8284 Vp-p = 4.9498 Vrms

 

4.9498 Vrms < 9.07 Vrms

 

So this headphone amplifier still doesn't provide enough voltage to reach 120 dB SPL either.

 

Here is the setup I have:

 

iPhone playing some V0 LAME MP3 classical music --> FiiO L9 line out dock -> JDS Labs C5 maximum volume -> AKG K 701 (newer 8-bump headband) -> my ears

 

Of course I would never listen to music this loud even with classical, but this setup was pretty dang loud. I don't completely know what clipping sounds like, apart from clipping due to poor mastering, but the music sounded just fine. I have no idea what 120 dB SPL sounds like but it was loud, for me.

post #33 of 89
Quote:
Originally Posted by miceblue View Post

 

With 90 dB SPL, or a theoretical maximum of 120 dB SPL with classical for the quietest parts of a song, 1.20 mW * 2^10 = 1228.8 mW or 1.2288 W of power is needed.

 

Not like it makes a big difference, but for 30 dB more (1000 times more), you mean 10^3 more, not 2^10.

 

Quote:
Originally Posted by miceblue View Post

On the X3's specification page it says:

From my understanding, 1 Vrms = 1 * 2 * sqrt(2) = 2.8284 Vp-p

So 8 Vp-p / 2.8284 Vp-p = 2.8284 Vrms

 

2.828 Vrms < 9.07 Vrms

 

Second of all, are these calculations correct?

 

> 8 V peak-to-peak. As in, more into some loads, or the specs are inconsistent with itself, or 8 is a number that's been rounded down.

 

Quote:
Originally Posted by miceblue View Post

Now this is where I'm having a hard time interpreting things. Based on this, the X3 shouldn't provide enough voltage to reach 120 dB SPL. Does this mean the X3 won't supply enough volume to reach the desired level?

 

It won't supply 120 dB SPL, but 120 dB is nuts (if continuous, and a whole lot even for a peak).

 

By the way, the headphones are listed as handling 200 mW (probably continuous) max input power. You'd get that from a 3.52 V rms sine wave and exceed it with more.

 

 

Quote:
Originally Posted by miceblue View Post

Is power supply the same as the maximum output voltage? 

 

It is not. It's an upper limit for the max output voltage peak-to-peak (barring some shenanigans like charge pumps internal to chips that are providing different, higher rails). The circuitry is limited by what the supply rails can give it. Depending on the design, it could be a volt or so, or much more than that, lower than what the difference of the power supply rails is. e.g. Schiit Magni supposedly does 130 mW into 600 ohms (25 V ptp) given +/- 15 V rails. O2 does 88 mW into 600 ohms (20.5 V ptp) given +/- 12 V rails. But some designs and chips would be closer than that.

 

And some designs may be current limited into 62 ohms. Most solid-state stuff probably isn't, if it's intended to be a headphone amp.


Edited by mikeaj - 6/30/13 at 8:43pm
post #34 of 89
Quote:
Originally Posted by Greenleaf7 View Post

With regards to the LCD2 and amp recommendation, I observed that most if not all the reviews on amp pairing were purely subjective. I realized that in the LCD2 amp recommendation thread itself, the perceived sonic quality has a strong correlation with the price of the amps used in the reviews. In almost any amp recommendation thread on head fi, you'll find reviewers that claim only certain high end amps are able to "bring out the best" in the headphone. Whether or not those sonic differences are able to be discerned in a proper volume matched DBT/ blind test remains to be seen.

 

Moving forward, is there a particular specification of amps that affects sound stage?

 

 

It might not be as direct as that. More expensive amps tend to have a more complex power supply and make more output, and that's what drives up the price aside from the mark-up (purely an example - an $800 amp without a large mark-up to make up for low volume production would still be, say, a $300 amp in parts). This is not to say that the 02 will be inadequate, it really depends on which headphone we're talking about. With my I've tried a number of amps below, at, and well above the price range of my current amp, and amps at all levels have exhibited making the system's midrange bloom worse (ie more excessive). Heck I would have dismissed the Soloist if I hadn't heard it on the Q701, given I was able to tell it was making the sound warmer,* but I'm still not throwing $1000 at it unless I can come by that amount easier (like not having to eat instant noodles for weeks).

 

Same thing with the LCD-2 - I've heard some cheaper amps (or were cheap used) not screwing up the sound. Like the O2 and one of those older Meier portable amps - they might not have the bass slam on the more expensive amps, but some of those amps not only added bass impact volume, sometimes they were too warm, adding midrange bloom** or the bass wasn't as quick. I'd forgo the perceived improvement in dynamic range if it costs that much more, if not make the sound worse in other areas.

 

 

*by putting it as a preamp/hpamp between my DAC and active speakers, and it was warmer than just the DAC and using the active speakers' volume pot
**note that people who like the more expensive amps tend to argue they are more more transparent to the source, based on their being more expensive, which also means that based on their logic a high end source and a high end amp with high end speakers/headphones have to sound borderline nasal

post #35 of 89
Quote:
Originally Posted by Meremoth View Post


So if someone had the intent of building an amp with a specific sonic impression, there's no guide or published research or data suggesting how to go about doing that?

 

Thanks for the info.  

 

There doesn't seem to be much literature because there's probably really not too much you can change unless you're intentionally going in with the desire to tweak things in various known ways like adding in an EQ or otherwise intentionally altering the frequency response (with many headphones, increasing amp output impedance will do just that too), ramping up noise, ramping up nonlinear distortion, etc.

 

I mean, increase the treble and most people would call it bright. But I'm not sure if that's the kind of change you're talking about or interested in.

 

 

Quote:
Originally Posted by Meremoth View Post

MOD EDIT:  Removed.  Let's keep it civil guys

 

Countering aside, the distinction is important to make.

 

A great deal of sonic differences that are attributed by users to amps seem not to be a result of the actual differences in electrical signals produced. It's just perceptions that change because people aren't careful with rigorous, unbiased A/B testing. (in fact, why should they bother?) When people are comparing amps, after level matching and a reasonable blinding protocol, obvious differences that people had "noticed" just minutes earlier during non-blinded listening frequently seem to vanish. Nobody can tell anymore. Now, that doesn't happen if you pick intentionally different enough amps, but for mid-to-high fidelity designs, even relatively cheap stuff...

 

So are you asking about electrical differences, which translate into differences in sound waves produced, or just perceived differences? There are any number of reasons for perceived differences.

 

There are a lot of descriptors like "aggressive" thrown around without any real evidence that people are actually hearing (as opposed to perceiving) such a thing. I mean, to figure out how to make an amp actually sound like that, first we'd need to figure out if people are legitimately hearing something like that and not just imagining it.

 

Just an FYI, but this subforum gets a lot of people passing through with hidden agendas and axes to grind hidden behind similar questions as you've had, so maybe words get misinterpreted.

 

 

Some people have the idea that amps are supposed to just multiply input signals, so any deviation from that is considered bad. i.e. something that's high fidelity. Take a look a the output/input relationship. So if that's the goal, then there's not much motivation to study sonic effects from amps, since there shouldn't be any (and it seems very possible to get amps close enough that there aren't differences discernible by the human ear). How do you add aggression or anything else without deviating from what the input was?

post #36 of 89
Thread Starter 
Quote:
Originally Posted by mikeaj View Post

Quote:
Originally Posted by miceblue View Post

 

With 90 dB SPL, or a theoretical maximum of 120 dB SPL with classical for the quietest parts of a song, 1.20 mW * 2^10 = 1228.8 mW or 1.2288 W of power is needed.

 

Not like it makes a big difference, but for 30 dB more (1000 times more), you mean 10^3 more, not 2^10.

 

Quote:
Originally Posted by miceblue View Post

On the X3's specification page it says:

From my understanding, 1 Vrms = 1 * 2 * sqrt(2) = 2.8284 Vp-p

So 8 Vp-p / 2.8284 Vp-p = 2.8284 Vrms

 

2.828 Vrms < 9.07 Vrms

 

Second of all, are these calculations correct?

 

> 8 V peak-to-peak. As in, more into some loads, or the specs are inconsistent with itself, or 8 is a number that's been rounded down.

 

Quote:
Originally Posted by miceblue View Post

Now this is where I'm having a hard time interpreting things. Based on this, the X3 shouldn't provide enough voltage to reach 120 dB SPL. Does this mean the X3 won't supply enough volume to reach the desired level?

 

It won't supply 120 dB SPL, but 120 dB is nuts (if continuous, and a whole lot even for a peak).

 

By the way, the headphones are listed as handling 200 mW (probably continuous) max input power. You'd get that from a 3.52 V rms sine wave and exceed it with more.

 

 

Quote:
Originally Posted by miceblue View Post

Is power supply the same as the maximum output voltage? 

 

It is not. It's an upper limit for the max output voltage peak-to-peak (barring some shenanigans like charge pumps internal to chips that are providing different, higher rails). The circuitry is limited by what the supply rails can give it. Depending on the design, it could be a volt or so, or much more than that, lower than what the difference of the power supply rails is. e.g. Schiit Magni supposedly does 130 mW into 600 ohms (25 V ptp) given +/- 15 V rails. O2 does 88 mW into 600 ohms (20.5 V ptp) given +/- 12 V rails. But some designs and chips would be closer than that.

 

And some designs may be current limited into 62 ohms. Most solid-state stuff probably isn't, if it's intended to be a headphone amp.

I did 2^10 because I thought you double the power for every +3 dB SPL increment. But yup the calculations ended up being pretty close, in this case at least.

 

Woah, I didn't realise how loud 120 dB SPL is. That is quite loud. >_<
http://www.etymotic-media.com/sliderule/

 

It makes me wonder how loud I listen to my music while riding on a bus...

 

Anyway, so 200 mW of power is the upper limit to what the K 701 can handle. Doing a reverse calculation from the same equation that I said before where you said I should use 10^3 instead, that means the K 701 can get to 112 dB SPL.

log[(200 mW / 1.20 mW to reach 90 dB SPL)] / log(2) = 7.38

7.38 * 3 + 90 dB SPL = 112.14 dB SPL

 

So if this is the case, what will happen if you're listening to dynamic music at a high volume level and a part of the song gets really loud and requires more power than what the headphone can provide? Will the headphone's driver get damaged?

NwAvGuy said that 110 dB SPL is a good ideal level for most music at a reasonably loud volume level, so the 112.14 dB SPL shouldn't be a problem under normal circumstances from what I understand.

 

 

 

And thank you for the explanation for the power supply limits.

post #37 of 89
Quote:
Originally Posted by miceblue View Post

NwAvGuy said that 110 dB SPL is a good ideal level for most music at a reasonably loud volume level, so the 112.14 dB SPL shouldn't be a problem under normal circumstances from what I understand.

110 dB is just over the red line that can cause hearing damage with extended listening. It's as loud as being on the tarmac at an airport. It's not a comfortable listening level if peaks are being hit often.

For comparison 130 dB is the threshold of pain.
post #38 of 89
Quote:
Originally Posted by bigshot View Post


110 dB is just over the red line that can cause hearing damage with extended listening. It's as loud as being on the tarmac at an airport. It's not a comfortable listening level if peaks are being hit often.

For comparison 130 dB is the threshold of pain.

We need to keep in mind there's a difference between peak level and average level that's called crest factor.  If you can hit 110dB max before clipping, then that's a maximum peak level.  The crest factor in music is about 15-20dB, so that put's your average at 90dB, which is loud, but not horribly.  Short duration peaks like snare hits can then hit 110dB SPL without hearing damage, and without sounding painfully loud.  Hearing damage and hearing pain results from a combination of SPL and exposure time.  

post #39 of 89
Quote:
Originally Posted by Meremoth View Post

What kind of  question is that?  Why would anyone want to play basketball, why would anyone ever eat spare ribs, why would you choose chocolate over vanilla?  

 

Most people in the sound science forums want high fidelity, not some weirdly distorted audiophile sonic nirvana or whatever they call it. People refer to the best amp as a wire with gain.

post #40 of 89
...same thing for DACs
post #41 of 89
Quote:
Originally Posted by uchihaitachi View Post

Most people in the sound science forums want high fidelity, not some weirdly distorted audiophile sonic nirvana or whatever they call it. People refer to the best amp as a wire with gain.

 

Ummm, okay?

 

Assuming you actually have one, what's your point?

 

MOD EDIT:  Portion removed.  Let's keep it civil guys

post #42 of 89
Quote:
Originally Posted by jaddie View Post

We need to keep in mind there's a difference between peak level and average level that's called crest factor.  If you can hit 110dB max before clipping, then that's a maximum peak level.  The crest factor in music is about 15-20dB, so that put's your average at 90dB, which is loud, but not horribly.  Short duration peaks like snare hits can then hit 110dB SPL without hearing damage, and without sounding painfully loud.  Hearing damage and hearing pain results from a combination of SPL and exposure time.  

I wouldn't recommend going much over 110, even for short peaks. The point is 110 is loud. It isn't what his mother is going to be listening at.
post #43 of 89
Quote:
Originally Posted by Meremoth View Post

Ummm, okay? Assuming you actually have one, what's your point?

His point is that it is perfectly possible to manufacture amps and DACs and players that have a completely uncolored output. What you put in is exactly what you get out. No added "warmth" or "veil" or "high end sparkle", just faithful sound reproduction. In fact, even inexpensive electronics can have that high level of audio fidelity.

So, if you can keep the signal totally accurate and calibrated from the player through the amp, then you are free to add coloration to taste as the last step before the speakers or headphones by using your tone controls, DSPs or EQ.

If every piece of equipment in the chain had its own coloration, the coloration of your player would cancel out the coloration in your amp, and if you got a new piece of equipment, none of the sound would be the same any more. Chaos.

Better to add the coloration at the very last step. Then you can totally control it.
post #44 of 89

MOD EDIT:  Portion removed to keep things impersonal.  Thanks fellas!

 

Quote:
Originally Posted by bigshot View Post


His point is that it is perfectly possible to manufacture amps and DACs and players that have a completely uncolored output. What you put in is exactly what you get out. No added "warmth" or "veil" or "high end sparkle", just faithful sound reproduction. In fact, even inexpensive electronics can have that high level of audio fidelity.

So, if you can keep the signal totally accurate and calibrated from the player through the amp, then you are free to add coloration to taste as the last step before the speakers or headphones by using your tone controls, DSPs or EQ.

If every piece of equipment in the chain had its own coloration, the coloration of your player would cancel out the coloration in your amp, and if you got a new piece of equipment, none of the sound would be the same any more. Chaos.

Better to add the coloration at the very last step. Then you can totally control it.

 

The worst part is, as pointed out above, those amps or pieces of equipment that claim to warm up the sound, may only do so by the placebo effect caused by appearances and claims of users, as pointed out by mikeaj.

 

If you are interested, I can discuss with you how to tailor the sound of your headphones with EQ.  I don't know why but EQ gets a vile reputation with most every subjectivist.  Maybe because they are so used to the sound being "changed" only by the power of suggestion that when a component actually *gasp* changes the sound, it is too much for their brains and ears to take?  Admittedly it is as easy to do harm as it is to do good to your sound with EQ.  But you know what doctors say?  The only kind of medicine that is risk-free is those that are effect-free (ie sham medicine)...

post #45 of 89

Equalization can do more to improve sound than anything else you can do.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Headphones NEEDING Power