[Donald North]

 
Quote:

I think the recording industry design standard (don't ask me to name the agency, I don't remember) is actually 120 ohm, but no one actually goes by that.  

I believe you're referring to the IEC 61938 standard for headphone output, specifying 5Vrms with 120 ohm output impedance. The purpose was to provide a more similar power delivery into both low and high impedance headphones.
 
Beyerdynamic's own A1 amplifier has 100 ohm output impedance. Some higher end Meier Audio amplifiers offer 0 and 120 ohm output impedance options. My Sonett also offers 120 ohm per the IEC spec. In my experience, the 250 and 600 ohm impedance beyerdynamic headphones sound better balanced when driven by higher output impedance than 0.

 

[nikongod]

 
Quote:

In my experience, the 250 and 600 ohm impedance beyerdynamic headphones sound better balanced when driven by higher output impedance than 0.

I share this opinion of the 250ohm beyers. I dont have enough personal experience with the 600ohm headphones, but it makes sense.

 

[xnor]

Quote:

sorry, you didnt mention that its YOUR OPINION that sony is crap. this is not the rite forum for that.

I didn't say sony is crap, don't put words in my mouth. I was referring to the specific model ("the sony") in question. English's not my first language, I'm sorry.. And yes, it is my opinion. Why is this not the right forum for that?
 
Let's see what you just posted:
 
Quote:

Originally Posted by Donald North
 
In my experience, the 250 and 600 ohm impedance beyerdynamic headphones sound better balanced when driven by higher output impedance than 0.
 
Originally Posted by nikongod
 
I share this opinion of the 250ohm beyers. I dont have enough personal experience with the 600ohm headphones, but it makes sense.

 
(emphasis by me)
... and I don't like your tone, but maybe you just had a bad day, dunno.
 
And please define "sound better", else I think "this is not the rite forum for that". Jeez!

 

[xnor]

Quote:

All this jibjab in the science forum and no one has mentioned impedance matching yet? In simple terms, you get optimal power transfer from your amplifier when your headphone impedance matches your amp impedance. I think the recording industry design standard (don't ask me to name the agency, I don't remember) is actually 120 ohm, but no one actually goes by that.
 
It's an ISO standard. Some manufacturers go by it, some don't. The Beyerdynamic A1 amp is close (heh) with Zs=100 ohm.
 
I don't think impedance matching is something you want to do with headphones, instead you want impedance bridging.
("since it is more important to accurately control the loudspeaker than to drive it with the maximum amount of power available")
 
Here's my take on things:
- you want optimal power transfer
 
Maximum efficiency (Zs~=0) instead.
 
[...]
 
If I had the knowhow to build a voicecoil myself, I would... but I don't. I did however make myself a couple impedance adapters (like the Etymotic 4P/4S adapter) in 90 and 260 ohm flavours and plugging into my Bottlehead Crack amp which is designed for high impedance cans.
 
Here's the problem. It's not designed for such a load.
 
If I plug in a straight 32 ohm can, it's hard to control and I can barely turn the volume knob without it going too loud, but no only that it doesn't sound very good. The high range is ok, but the low end sounds all flabby and undefined. Plug in an adapter and the most noticeable effect is volume attenuation. My listenable levels are now within a comfortable 10-2 o'clock sweep and the definition has increased noticeably (I hesitate to say "significantly" since we all hear things differently), especially in the low end.
 
Putting resistors in series with speakers .. is not a good "solution". In fact, the result of doing so is flabby and uncontrolled bass.
 
Now this is just my experience in adding resistance to the output line. The only effect here should be noise rejection and impedance load matching to optimize power transfer.
 
Noise rejection isn't needed with the right amp/device that's been designed for such a load.
Power transfer: see above
 
In the end, you want to headphone to see the "same" voltage the amp is seeing on the input (bridging), and not a "distorted" version of it (power transfer), right?

Answers in bold.

 

[mikeaj]

Below I quoted my previous blurb about the difference in signal seen across the different impedance headphones.  However, if indeed the higher impedance headphones have drivers with less mass, does that make an appreciable difference in the mechanical behavior in tracking the input signal?
 
Taking a naïve view of the headphones being a classical second-order mass-spring-damper system--in reality, the general principles hold even for systems not second order, so bear with me--the mass is related to the damping ratio zeta.  This comes from the equation mx''+ cx' + kx = y, where y is the input signal, x is the location, m is the mass, c is the damping coefficient, and k is the spring coefficient.  The damping ratio is proportional to the inverse of the square root of the mass, so with less mass there should be a greater damping ratio.  Based on the HeadRoom graphs the headphones all seem to be underdamped (damping ratio between 0 and 1).
 
Based on the step response, which is similar to the portion of the square wave response following a switch in polarity, the headphones with a higher damping ratio should have less overshoots, among other things. Does this make any sense, or did I make a mistake somewhere?  I realize there were a lot of simplifications made here.  I'm not thinking clearly today with regards to the EMF induced relating to the motion of the transducers.  In any case, it's all difficult to interpret without knowing exactly the mechanical differences between the drivers.
 
Speculation aside, if there were a large difference in sound resulting from this difference in mechanical quality, you would expect it to show up in the square wave response graphs.  People explicitly mention "speed" and "faster transient response."  This is for the 250 and 32 ohm versions.  Is there a significant difference?

 

 
 
Quote:

With higher impedance, you get less current requirements, higher damping factor, higher signal-to-noise ratio, less crosstalk, and probably some other things I am forgetting.  Amps aren't quite perfectly linear, so I wouldn't be surprised to see different responses to different loads.  In many applications (I mean not audio in general), impedance in the low hundreds and especially in the tens is often considered unusual and difficult to deal with.  Some of the highest-fidelity line-driver op amps aren't spec'd for impedances below 600 ohms, so a more "typical" load might be better driven.  As mentioned earlier, many volume potentiometers have issues with channel balance in low settings, which a high impedance headphone would not need to be operated at.  Non-class A amplifiers have more trouble at around zero output, which happens more at a lower gain setting required by lower impedances.
 
Of course, these differences are most likely very marginal between 250/600 versions, though the 32 ohm version could be somewhat different because of these things.  Just take these as possible factors.

 

[nikongod]

Quote:

 
(emphasis by me)
... and I don't like your tone, but maybe you just had a bad day, dunno.
 
And please define "sound better", else I think "this is not the rite forum for that". Jeez!

White noise played through my DT880/250 with an IEC output source has fewer obvious dips and bumps VS the same general chain but with a near-zero ohm output impedance.
 
Why not add a way to verify or disprove the OP's idea? So far we have strong reliance on 1 source of information VS a large amount of anecdotal evidence to the contrary.
 
Regarding the head-room graphs: have you ever seen 2 frequency graphs of 2 things that claim to be different that are so similar? 8 & 16 ohm versions of the same lowther speakers dont measure this similarly... 
 
 
I should also add: 
Im not a fan of the square wave outputs from the headroom system, and have not been for a while. Not a single one is even close to smooth. Every square wave graph shows a similar ring with only the slightest variations. from every headphone from every MFR... 
 
Try it out. there are a FEW headphones that dont ring like the vast majority of the others, but they have obviously rolled off treble response... 
 

 
There may be a flaw in headroom's measurement techniques.

 

[mikeaj]

Yes, it'd be nice to have similar results published by anybody else, or at least a more detailed description of HeadRoom's methodology and equipment.  After going through some more of those, I can agree that they look suspicious.  You might say that HeadRoom, as a commercial and not an academic entity, has incentives to make equipment look more different from each other--if the square wave responses look the same, then they must be! 
 
But I think incompetence (improper methods or equipment) may be the simpler and more likely accurate explanation.
 
Okay, I kid, the motivation is to have their logo watermarked on as many graphs around the Internet as possible.
 
Quote:

I should also add:  Im not a fan of the square wave outputs from the headroom system, and have not been for a while. Not a single one is even close to smooth. Every square wave graph shows a similar ring with only the slightest variations. from every headphone from every MFR... 
 
Try it out. there are a FEW headphones that dont ring like the vast majority of the others, but they have obviously rolled off treble response... 
 
There may be a flaw in headroom's measurement techniques.

 

[leeperry]

well, yeah, those headroom measurements are meaningless...and yes the 600Ohmers sound much "better" than the 250, their mass is also lower just like BD said. BD are not much of bs tellers IME.

 

[Catharsis]

Quote:

well, yeah, those headroom measurements are meaningless...and yes the 600Ohmers sound much "better" than the 250, their mass is also lower just like BD said. BD are not much of bs tellers IME.

Leeperry, can you verify the mass of the 600ohm drivers and the 250ohm drivers with empircal measurements (i.e. use a weight scale)?  And can somebody verify his claims?
 
FWIW leeperry, headroom graphs are not meaningless.  Regardless of what logic you attempt to use to dismiss them, all of the graphs at headroom use a standard mic and dummy head setup which offers a consistent HFTF.  While your head and my head may differ, you can still compare 1 headphone to another using this standard.
 
Another headphone measurement website (japanese) with waterfall  / impulse response graphs.
 
http://www.geocities.jp/ryumatsuba/review.html
 
$By your logic, all lighter diaphragms are fast, and all heavy diaphragms are slow.  Have you ever seen an impulse response / waterfall plot for an electrostatic (light) driver? By that logic we would expect that all loudspeakers would be orders of magnitude slower than even the heaviest headphone driver.  We seem to forget about the force that is applied to the mass.
 
STAX 4040 ("super light driver")
http://www.geocities.jp/ryumatsuba/srs-4040.html
 
DENON D5000 ("big fat heavy driver")
http://www.geocities.jp/ryumatsuba/ah-d5000.html
 
Though I know you'll find the achilles heel in these graphs and call them meaningless, I must say your alternatives have been less than underwhelming.
 
 
Leeperry, if you spend a few minutes looking at the data, you'll notice startling positive correlations between flat FQ response and impulse response.  Both are of the "time" domain. 
 
This is like arguing evolution with a creationist. 

 

[leeperry]

yay, circular logic again.."all lighter diaphragms are fast, and all heavy diaphragms are slow"...I never talked about diaphragms, I talked about voice coils, completely different thing. Let me google it for ya: http://en.wikipedia.org/wiki/Headphones
 
"The diaphragm is actuated by the attached voice coil, when an audio current is passed through the coil. The alternating magnetic field produced by the current through the coil reacts against the static magnetic field in turn, causing the coil and attached diaphragm to move the air, thus producing sound"
 
AFAIK the diaphragm is identical in all those drivers and yes, the lighter the voice coil in the BD, the faster they are(ever realized that the T1 is 600Ω?)...anyway, that's really my last post in your failtastic thread. Good luck, then.

 

[Catharsis]

 
Quote:

yay, circular logic again.."all lighter diaphragms are fast, and all heavy diaphragms are slow"...I never talked about diaphragms, I talked about voice coils, completely different thing. Let me google it for ya: http://en.wikipedia.org/wiki/Headphones
 
"The diaphragm is actuated by the attached voice coil, when an audio current is passed through the coil. The alternating magnetic field produced by the current through the coil reacts against the static magnetic field in turn, causing the coil and attached diaphragm to move the air, thus producing sound"
 
AFAIK the diaphragm is identical in all those drivers and yes, the lighter the voice coil in the BD, the faster they are(ever realized that the T1 is 600Ω?)...anyway, that's really my last post in your failtastic thread. Good luck, then.

Goodness Leeperry, your evidence is so compelling.  The judge and jury would bend over backwards in headphone court. 
 
I see you're going for the "if I restate what I've already stated over and over and over, eventually people will start to support my arguement" tactic.  How's that working for you?

 

[Anaxilus]

Quote:

FWIW leeperry, headroom graphs are not meaningless.  

They are when every variance is explained away by you as sampling error.

 

[Catharsis]

Quote:

They are when every variance is explained away by you as sampling error.

It is possible to have anomalies; especially when one measurement is out of character with others.  It may be true that the headroom measurements of the DT770 are accurate, but it remains inconsistent with the results posted from different impedance DT880 / DT990.
 
I am aware that headroom has edited / altered / remeasured headphones in the past and that they use a HRTF function to translate the raw data.  I made it clear that the DT770 anomaly could be a measurement error, (or not), and that it may be redone by headroom in the future.
 
I hope this doesn't turn into a "measurements are meaningless" argument when applied to audio.  Science can and does apply to everything natural and testable and that includes audio, and requires objective measurement. 

 

[Catharsis]

Thanks everybody for your contributions to this thread.  It appears as though the objectivist / calculationist approach would demonstrate / predict that there is in fact very little difference (inaudible) between 250ohm and 600ohm impedance variations of both the DT880 and DT990 headphones and could explain why some people have not noticed a sonic difference between the two in level-matched testing.  Of course other people have noticed a difference, but that's no great feat considering how many people hear differences when clearly there isn't (cables...cough...)
 
Still to my amazement, even on a sound science forum, there are those that still shoot holes in any shred of evidence (no matter how scant) and insist that we resort to using our ears to hear the differences or to hold to sets of drivers in our hands to feel the difference in weight.  With all we know about level matched blind testing (ABX), psychoacoustics, expectation bias, and the limits of human perception, I don't know why some people even venture into this forum, other than to dilute its integrity.  I am perhaps a bit bitter as my tone will show - I deal with science on a daily basis and when people meddle with it, my job just gets harder, and the world gets a little dumber.  

Of course, I'm still dying to see some evidence (in the form of real measurements or accurate calculations) that would nullify my hypothesis.  I hope this discussion continues!

 

[HariBhushan]

Catharsis, I'm going to agree with you about this "600ohm so-called conspiracy theories"
 
Right now, I own both DT880/600 and DT990/600 (herp, derp///  I actually wonder how i ended up with 600ohm version of these phones/// herp derp) 
 
I have tried the DT880/600 and DT990/600 from just abit every source and amp i could find(but mostly on my Woo Audio 3) and tried many many sources in the past, but no matter what, the brightness/ sharp agonizing treble spikes were still there[oh...but i thought 600ohm beyers corrected all the problems and flaws?]. Youtube videos with the slightest amount of errors are heard so easily that you don't even want to go near them. Forget playing any mainstream albums, it's physically impossible to go on for 20 min straight without your ears bleeding out.
 
I remember owning a DT880/250 a while back, and i remember that had very soft ear pleasing treble IIRC. I really want to re-visit the dt880/250 because of my good experience with them in the past. I honestly think that 600 ohm is bs,  it's probably meant to be used with high end studio equipment or something. I bet beyer are probably laughing their assess off on this one.  I do plan on owning either a dt990/250 or a dt880/250 again in the future so i can do some more A/B testing. You really can't trust anyone's ears except yourself.