Help choosing project.

[dsavitsk]

yes but we arent talking about amps tuned to be current sources here; running on current drive, are we?

I was merely using it as an example of how mechanical and electrical damping both matter, which I was told above did not.

Anyhow, I think the black and white nature of this argument is pretty stupid. As are the ad hominem attacks. The "objectivists" are correct that measurements matter. They matter a lot. However, the problem is that they have not identified how measurements correlate with sound. There is a temptation to view THD, or IMD, or square wave response, or whatever as the end all and be all of amplifier quality. But, this is simply not the case. So, until they can more clearly identify how exactly measurements matter (did you know that dissonant signals are amplified in the brain stem, while consonant ones are attenuated? that the sound preceding a sound influences how it is processed, and that musical experience influences all of this? http://www.soc.northwestern.edu/brainvolts/publications.php ), and identify some that do, "good" objective measurements can at best be seen as a necessary, but not sufficient, condition for good sound.

Similarly, "subjectivists" need objective measurements as otherwise they are simply guessing. Correlating changes that produce improvements with objective standards is the essence of learning.

Most of us who design audio equipment are somewhere between these extremes. Yes, you need to measure it, but in the end, it needs to sound good, too.

Back to the point, we are not driving theoretical resistors -- these are dynamic systems. And, the damping of the speaker driver has a lot to do with what amplifier is appropriate for driving it. If you don't believe this, hook up a Krell power amp to a Lowther speaker some time. It will sound awful, and have no bass. This is not because either is bad, but because they are incompatible. And, the argument here is that the drivers in Grado headphones are more like Lowthers than other more traditional speakers, which is really not surprising. So, a little less damping seems to go a long way. And, to bring it full circle, Grados do great with high impedance, "transconductance" amplifiers.

 

[wakibaki]

Can I first of all say that I was over-hasty in how I went about dismissing  the question of over-damping. I apologise for that
 
Can I then say that over-damping is being overplayed as a performance issue in headphone reproduction and audio reproduction in general. The worst consequence of over-damping is a slight roll-off in low frequency response.
 
Amplifier output impedance is only one of several features in a system contributing to damping. The impact of reducing amplifier output impedance is trivial in most instances in terms of introducing over-damping of the system as a whole. The likely impact of a low  amplifier output impedance is tight, well-controlled bass because the driver is well-damped not over-damped. A driver which will be over-damped with an amplifier output of zero impedance will be rare.
 
The issues which are solved by reducing amplifier output impedance such as a peaky frequency response are far more obvious and intrusive than any which might be caused by reducing amplifier output impedance, such as a slight roll of in the bass.
 
For this reason no lower limit is specified for amplifier output impedance when manufacturers publish data for their drivers, because is is generally presumed that the lower the better, and in fact it turns out that the lower the amplifier output impedance, the flatter the frequency response, which is the single most important arbiter of driver quality. A low amplifier output impedance makes a driver look good.
 
On the other hand, under-damping (caused by a too-high output impedance) not only contributes to a peaky FR, it can result in an ill-defined, wooly or boomy bass, and frequently does.
 
Everybody knows that, so cautioning against a low output impedance on the basis that it causes high-end harshness in Grados when the OP has Ultrasone Pro 750s is hardly going to encourage me to have patience with the intervention.
 
@tomb
 
Just because things have become accepted here over a period of years does not mean that they are right.
 
Some things here have become completely closed off to discussion, which is a definite indicator that something is seriously wrong. People are attempting to insulate themselves from the truth. The harder they try, the more difficult it gets.
 
 
Loyalty is nice to see, but take care not to sacrifice your integrity on the altar of your loyalty.
 
w
 
Don't mention the elephant.
 
I did, but I think I got away with it.

 

[nikongod]

Quote:

Can I first of all say that I was over-hasty in how I went about dismissing  the question of over-damping. I apologise for that
 
Can I then say that over-damping is being overplayed as a performance issue in headphone reproduction and audio reproduction in general. The worst consequence of over-damping is a slight roll-off in low frequency response.
 
Amplifier output impedance is only one of several features in a system contributing to damping. The impact of reducing amplifier output impedance is trivial in most instances in terms of introducing over-damping of the system as a whole. The likely impact of a low  amplifier output impedance is tight, well-controlled bass because the driver is well-damped not over-damped. A driver which will be over-damped with an amplifier output of zero impedance will be rare.
 
The issues which are solved by reducing amplifier output impedance such as a peaky frequency response are far more obvious and intrusive than any which might be caused by reducing amplifier output impedance, such as a slight roll of in the bass.
 
For this reason no lower limit is specified for amplifier output impedance when manufacturers publish data for their drivers, because is is generally presumed that the lower the better, and in fact it turns out that the lower the amplifier output impedance, the flatter the frequency response, which is the single most important arbiter of driver quality. A low amplifier output impedance makes a driver look good.
 
On the other hand, under-damping (caused by a too-high output impedance) not only contributes to a peaky FR, it can result in an ill-defined, wooly or boomy bass, and frequently does.
 
Everybody knows that, so cautioning against a low output impedance on the basis that it causes high-end harshness in Grados when the OP has Ultrasone Pro 750s is hardly going to encourage me to have patience with the intervention.
 
@tomb
 
Just because things have become accepted here over a period of years does not mean that they are right.
 
Some things here have become completely closed off to discussion, which is a definite indicator that something is seriously wrong. People are attempting to insulate themselves from the truth. The harder they try, the more difficult it gets.
 
 
Loyalty is nice to see, but take care not to sacrifice your integrity on the altar of your loyalty.
 
w
 
Don't mention the elephant.
 
I did, but I think I got away with it.

Half of what you wrote here is flat out wrong or half right at best. 
 
Starting at the top...
It is impossible to make a broad generalization of how a given driver will respond to a variety of output impedances. One could generalize based on brand or house sound, but to make an accurate statement covering all dynamic headphone drivers is impossible. 
 
Depending on the drivers, overdamping (electrically) can lead to peaky highs on some headphones. Put another way, reducing damping factor (increasing output impedance) can reduce/subdue peaky highs on certain headphones. 
 
Your statement that a given headphone will measure flatter with a lower output impedance is hopelessly broad and inaccurate. 
 
The rest is really stating that a few more times, a few different ways. 
 
The rule for low output impedances stems from multi-driver speakers where the CROSSOVERS do funky things when the amp's output impedance is not right. When applied to single driver speakers without crossovers or "compensation networks"* the "0ohms or bust" rule for amplifier output impedance is meaningless. 
 
*Would you even need the compensation network if the amp had the right (non-zero) output impedance? I have heard a few single driver speakers go from absolute garbage to something quite nice with no change but a 3ohm resistor in series with them.

 

[Willakan]

I've taken my discussion to PM. I think perhaps others should do likewise/start a new thread on the damping of dynamic headphone drivers. Come on, if I can resist the urge to respond to the criticisms of objectivism you can all surely resist the urge to argue about damping in a thread about what kind of amp someone should build?

 

[nikongod]

 
Quote:

I've taken my discussion to PM. I think perhaps others should do likewise/start a new thread on the damping of dynamic headphone drivers. Come on, if I can resist the urge to respond to the criticisms of objectivism you can all surely resist the urge to argue about damping in a thread about what kind of amp someone should build?

Is your stance objectitivism or just the blind following of rules? 

 

[qusp]

ok dsavitsk: you know what, i cannot disagree with anything you say right there and you have provided reference that makes more sense to me (and i'm in a less than argumentative mood). it depends on what plane you are viewing the said improvements, or where the 'fault' lies (in between by the sounds of it) a zenlike path of least resistance so to speak, talking in a holistic sense. my main question was of the profundity of the effect on headphone drivers vs speaker drivers with considerably higher inertia and physical impedance rather than just electrical with minimal resistance wrt pressure being open, light and small, but i'm happy to learn something new. the way i framed that statement you quoted was quite flawed reading it back...
 
i've never been a grado man, nor do i spend much time in the fullrange forums, being more of a sennhieser or jh13 guy for headphones and a digitally crossed classic brittish monitor/bass reflex guy with speakers (dont have the room for OB, transmission line, or horn) i'll concede my own opinion (later clarified) was perhaps too rigid as to not include more pure musical experiences, that are not perhaps as 'accurate', but preferred by some as more tonally rich/correct
 
i'm aware of Gm amps as have done some study on them wrt dac iv stages. 
 
in between lies the middle
 
my amps (most currently in progress to match with new speakers funnily enough)
 
the wire hybrid chip/lateral fet hybrid poweramp
the wire IC based ridiculously accurate 16W amps
aleph JX
Semisouth jfet circlotron
 
conflicted
 
also i have all the wire headamps, one using a modded iv stage with current gain, a mostly finished DAO bar casework and a few other bits and pieces
 
all are balanced input, weeell the AJX hmm hard to call that one

 

[qusp]

Quote:

I've taken my discussion to PM. I think perhaps others should do likewise/start a new thread on the damping of dynamic headphone drivers. Come on, if I can resist the urge to respond to the criticisms of objectivism you can all surely resist the urge to argue about damping in a thread about what kind of amp someone should build?

agreed, but actually where is the OP? sorry for the mess :redface:

 

[wakibaki]

Quote:

It is impossible to make a broad generalization of how a given driver will respond to a variety of output impedances. One could generalize based on brand or house sound, but to make an accurate statement covering all dynamic headphone drivers is impossible. 
 
Depending on the drivers, overdamping (electrically) can lead to peaky highs on some headphones. Put another way, reducing damping factor (increasing output impedance) can reduce/subdue peaky highs on certain headphones. 
 
Your statement that a given headphone will measure flatter with a lower output impedance is hopelessly broad and inaccurate. 

 
Sorry ng, and I'll grant you that it is a broad generalisation, but it's an accurate generalisation. I'm not insisting that it's true in every case, just the very large majority.
 
under-damping (caused by a too-high output impedance) not only contributes to a peaky FR, it can result in an ill-defined, wooly or boomy bass, and frequently does.
 
This is not a controversial assertion. It's a trifle loose, since the underdamping contributes the problems in the bass and the high output impedance the peaky FR, but we're accustomed to discussing these issues in the same breath.
 
It's a good generalisation and one with which only a person desperate to preserve an approach with no rational foundation would disagree.
 
In general, a low amplifier output impedance results in a performance preferred by most people, i.e. an uncoloured performance.
 
It's not rocket science. 

 Damping factor (and its only-halfway appropriate name) was being discussed in hi-fi magazines of the '60s. Damping factor is only 1 of 3 principal results of a low output impedance.
 
You will find it impossible to assemble a substantial body of evidence (even anecdotal) to support the contention that anything but a very few, if any, headphones benefit from a high amplifier output impedance. I know, I've searched already. Perhaps, while you're at it, you might try to work up a description of the  mechanism  underlying this beneficial effect.
 
On the other hand, you will not find it difficult to find plentiful verification even in graphical form of my generalisation (yes, plots of the same headphone driven by different impedances), in a place to which I am prohibited from linking. You will also find descriptions in the text of the underlying mechanisms.
 
My recommendation to the OP stands, especially, as I have pointed out more than once already, since the contrary suggestion was based on the putative behaviour of Grados, which the OP does not own.
 
You simply want to disagree with me, or you would be raising that particular objection yourself.
 
I made a simple, sensible suggestion as a starting point on which the OP might build his experience of amplifiers and headphones. 
 
I have been doing my best to help a fellow member. Perhaps I should have made a greater effort to express myself more precisely, but I didn't start out with the expectation that I would have to refute a slew of ill-conceived objections. 
 
You (all) have been unable to let an unadorned piece of common sense pass without raising obscure and trivial objections and interjecting irrelevances based neither on equipment owned by the OP, or equipment in general, but on specific equipment which he definitely does not own, for motivations at which I can only guess and none of which conceivably redound to your credit, presumably because you felt your cages had been rattled.
 
If you people have anything constructive to suggest to the OP, out with it, and stop exercising your hobby-horses on my lawn.
 
w

 

[qusp]

well i guess my wire amps wont be great with grados, specific output Z into 8ohms was posted today as 0.003R/3mOhms for the LPULD, 0.012R/12mOhms for the SE-SE headamp and 0.0062R/6.2mOhms for the bal-bal headamp. tested with an AudioPrecision System 2
 
conditions
 
with the headamps, most detailed results were for the SE-SE seen below for the output terminals, not taking the DCR of the headphone cable or driver into account.
 
6VRMS output into 37.4ohm load using dV/dI method
 
 

V (open load) = 5.998 VRMS (measured with AP and a calibrated Agilent meter) V (37.4 ohm load) = 5.996 VRMS (measured with AP and a calibrated Agilent meter) I (open load) = 0A I (37.4 ohm load) = 0.16018 ARMS (measured with a calibrated Agilent meter) So we have 2mV of voltage delta and 0.16018A of current delta. This gives: output impedance = dV/dI = 0.002V/0.16018 = 12.5 milliohms. That gives a corresponding damping factor of 2560 with a 32 ohm load or 48000 with a 600 ohm load.

 
 
not bad =) but apparently no good for Grados

 

[ExecuteMethod]

Quote: Originally Posted by Willakan  I've taken my discussion to PM. I think perhaps others should do likewise/start a new thread on the damping of dynamic headphone drivers. Come on, if I can resist the urge to respond to the criticisms of objectivism you can all surely resist the urge to argue about damping in a thread about what kind of amp someone should build? agreed, but actually where is the OP? sorry for the mess :redface:

I'm here, I was just eating some popcorn.

Seriously though, I have been so busy with work that I haven't really been following the discussion, and honestly, it's a bit over my head. I am not a designer, and I haven't heard enough hi-fi systems to judge subjectively, and haven't studied enough audio to comment objectively.

Essentially, I would like to build an amp as a project that would subjectively and/or objectively sound better on these specific headphones than my current setup.

Are the drivers in the Ultrasones similar enough to those of Grados to look at amps with, as dsavitsk put it, a slightly higher impedance and transconductance?

 

[ExecuteMethod]

 
 

This was actually my first thought. I'm really wish I had somewhere near me that I could demo some different amps.

 
 
Where are you from?

I am in VA, near Bristol.

 

[nikongod]

 
Quote:

Are the drivers in the Ultrasones similar enough to those of Grados to look at amps with, as dsavitsk put it, a slightly higher impedance and transconductance?

Try them with various series resistors on each channel. Very simple test. 
 
Id try values from about 7 to about 500ohms. 
 
If you like them with a few ohms say 5 to 20, try transformer coupled Single ended triode
 
If you like them with "who a lot" of ohms you have all sorts of options. 

 

[qusp]

Quote:

I'm here, I was just eating some popcorn.
Seriously though, I have been so busy with work that I haven't really been following the discussion, and honestly, it's a bit over my head. I am not a designer, and I haven't heard enough hi-fi systems to judge subjectively, and haven't studied enough audio to comment objectively.
Essentially, I would like to build an amp as a project that would subjectively and/or objectively sound better on these specific headphones than my current setup.
Are the drivers in the Ultrasones similar enough to those of Grados to look at amps with, as dsavitsk put it, a slightly higher impedance and transconductance?

ultrasones are utterly different headphones, drivers are totally different, titanium most likely, sealed synthetic cups. couldnt be more different, but as Nikongod says, you can try it out with a resistor to test, but only to test, if you liike it i would look into another method to achieve the same result, as he suggests. its not just the drivers at least with speakers its the whole system of electrical and acoustical impedance.
 
this is how i'm led to believe it works anyway =P
 

 

[wakibaki]

Quote:

not bad =) but apparently no good for Grados

Why not take a look at this thread:- http://www.diyaudio.com/forums/headphone-systems/162962-maximum-allowable-headphone-amp-output-impedance-7.html
 
...where it is clearly explained why the nonlinear mechanical damping of output transducers makes low output impedances and the accompanying enhanced control of the diaphragm desirable and how modern transducers take advantage of advances in materials technology and SS design to provide better bass response via acoustic suspension in part leading to speaker drivers being designed to anticipate low (zero) output impedances, which is seen as being a desirable goal for 'phones too. 
 
Grados are apparently designed to suit a zero ohm output impedance...
 
w

 

[nikongod]

I would say that "the wire" is exactly as equally well suited to a Grado headphone as any other measurements first amp. 
Can you *objectively* say that its even different than the O2, M3, Pimetta, b22, or other measurements first amp? No? So since you really cant prove that it even sounds different through measurements (all measure similarly enough that differences should be inaudible) or DBT all you can really say is that its an expensive amp that you like. 
 
There are a lot of other amps & techniques mentioned in this thread that don't measure as well BUT at least they sound different. And differences are what we are here to talk about, aren't they? Not everyone has the same subjective opinions of what sounds nice. The alarming differences between various headphones clearly shows that. 
 
How many speaker drivers do you know of driven by a zero ohm source impedance? Please consider the series resistance of inductors & ESR of caps at audio frequencies. For sure there are a couple in speakers with active crossovers, but in anything with a passive crossover I have to say: Zero ohms, what?
 
Perhaps it would be faster to just tell people what headphones to listen to and how than to tell them how they should be enjoying what they are enjoying their own way.