[setmenu]

I have some experience with the effects of high source impedance
with My Diy ribbon/planar phones.
The current model has a lot of series resistance in the planar
coil, effectively giving me little if any damping factor.
There is little or no mechanical damping with this design.
The effect is a dramatic mid-bass resonance that needs taming with an inline capacitor.
Also there is some midrange 'suck out' .

The next version of these coils will have little series Resistance so
things will be down to cable/source impedance for control.
This will need to be low.

Depending on the results this will enable some tuning of the sound to be achieved.

Like others here I also feel that the Beyer 931s benefit from a bit
of series resistance, which seems to calm the highs and improve
the bass volume[probably at the expense of some control if taken too far].

I would guess that manufactures use such effects to 'tune' their
products also, as I have noted that people mention cable changes on Senns for instance,
can brighten up the sound.


Cheers


Setmenu

 

[wallijonn]

Jona,

you could order a 120 ohm "box" from Meier-Audio (Dr. Meier) or you could make your own.

To make your own you'll need a Radio Shack box, a 1/4" stereo plug and a 1/4" stereo receptecle. You'd unscrew the plug, make a hole which it can be threaded into, add a nut on the inside, then put the receptecle on the inside. Add a ground wire between the two grounds.

Now you can solder and unsolder any resistance you want.

Once you find the right resistance (Radio Shack sells 100 ohm'ers and not 120s, but you can parallel their 220s for 110 ohms; or you can get a pack from other stores).

make sure that you have no shorts (especially if you tighten down the receptecle too far down) before inserting into the amp or connecting the headphones.

So yes, you can tailor it to anything you want. Just a slight tweak may be enough to make a bright can acceptable. you'll end up turning the volume higher, but the overall tonne will change accordingly.

And no, they are not reversible. But I have an idea to add "resistor insertion pins" on the outside whereby a resistor or a straight piece of wire can be added. It will be unobtusive. Damn, what a great idea!

Steve999,

there's a section on driving headphones directly from a DVD/CD line out in the amp, source and members area. The problem is that you need a DVD/CDP with variable line outs or you add a potentiometer. Some say that it is much clearer sounding. It would emphasis the treble and de-emphasise the the bass. typically you have 2v @ 600 ohms.

 

[robert]

rather than start yet another thread:

my tube amp has a barely noticeable hum with 32 ohm grados; none with 60 ohm KSC-50 or 100 ohm ETY-ER4.

a suggestion is to add about 50 ohm of load with a resistor between the lug of each output transformer and the jack.

so: does this 50 ohm load raise the impedence seen by the amp, and is thus a Good Thing; or does it raise the output impedence of the amp, thus lowering the damping factor and is a Bad Thing???

 

[Steve999]

My overall impression is that a) this stuff is very complicated and differences are headphone-specific based on impedence vs. frequency characteristics of the headphones; and b) I don't have a clue about it, and most of what I read here doesn't seem to hold water.

I'm sure that many here have a better understanding of it than I do, but I don't see any lucid, coherent understanding of the subject reflected in any of these posts.

Further confusing me is Jazz's assertions that he hears no difference running headphones straight out of a line output, which as I understand it has a huge impedence (perhaps 600 ohms at a minimum and possibly thousands of ohms). Now, Jazz is a little on the theoretically zany side at times (sorry Jazz, but I do find you tremendously entertaining at all times, and I'm equally zany if not moreso, so don't take it too hard), IMHO, but Jazz seems to have very good ears (for example, he held his ground on the PX200s and turned out to be right) in any event, and if he says he hears little or no difference on A/B testing between a line out and a headphone amp, I absolutely believe him. Further, I "tried this at home" myself, and it seems to me he may well be right. Which seems to throw this whole impedence-matching business into grave doubt, or at best, mass confusion.

So, WHAT IS THE DEAL HERE? ISN'T THERE SOME SUPER-EDUCATED LURKER WHO SITS AROUND AND LAUGHS AT US AND JUST PLAIN KNOWS THE ANSWER??? IF SO, SPEAK UP!!!!

 

[pedxing]

This stuff is hard to understand. Don't feell bad. I know electrical engineers who can't do this stuff either, but that might be becaues today's curriculums in certain colleges are more about digital designs and less about analog stuff.

 

[PeterR]

Quote:

Originally posted by Steve999 Maybe you could show me the right math, because I find it confusing myself. Best I can tell, this theory would always warrant a 0 ohm output jack for any phone (0 divided by anything equals zero, a low ratio indeed) so the mathematical model has no explanatory value whatsoever, in fact, it would appear to border on a nonsequitar.

Yes, as far as the damping factor is concerned, the lower output impedance the better, 0 Ohm being the ideal case.

Quote:

Further, the formula would be contradictory to the concept of most phones being designed for the industry-standard 120 ohm jack, and we do know some phones are designed to be driven by a 120 ohm jack, and that many of these phones (Beyers, for example) tend to have high impedences; this is real-world evidence against the mathematical/sonic theory that a high impedence phone will tend to sound better out of a low ohm jack. What am I missing?

Let's start with what the output impedance does. The output impedance of the amplifier (think of an ideal amplifier with a resistor in series with its output) forms a voltage divider with the impedance of the headphone. Therefore the signal into the headphone will be attenuated compared to the ideal 0 Ohm amp. Now the problem is that the impedance of the headphone over frequency is not constant. There's the resonance peak in the bass, and because of the inductivity of the voice coil there will also be a rise towards higher frequencies. Therefore the attenuation is not the same at all frequencies. The attached image shows the electrical response for a Sony 7506 driven from 120 Ohm output impedance (the response for zero source impedance would be a flat line at 0dB). So the frequency response will be deformed according to the impedance response when the phone is driven from an amp with an output impedance >0, and the higher the output impedance, the stronger this effect.
Now if a phone is designed with 120 Ohm output impedance in mind, this deformation of the frequency response is taken into account and required to achieve the desired response. Driving such a phone from a 0 Ohm jack will therefore result in a different response than the designer had in mind, and probably sound worse.

 

[Steve999]

Thanks PeterR, now that I actually sort of understand. That was tremendously helpful. I imagine you do sit around and laugh at us.

Seriously, I really, really appreciate it.

HOWEVER, it seems to me to indicate that the 7506 out of a 120 ohm jack will be about 1 db BRIGHTER at 20 khz, as well as introduce a 1 db increase in the LOW bass. (Not a big change, but not a good thing for the V6s!) Am I missing something?

Is this the general pattern?

Would running the HD580s at 120 ohms then likely make them a just a little brighter in the upper frequencies (10khz to 20 khz) as well as a just little more pronounced in the LOW bass? Or is this an overgeneralization, that is, do headphones vary greatly in their impedence vs. frequency characteristics?

I am also inferring that the response curve differences from a 0 ohm output vs. a 32 ohm output would be quite small but measurable.

Would running headphones out of a line out or variable line out cause much greater distortions of intended frequency response, or do resistance specifications for those outputs, typcially 600 ohms to 10 kilo-ohms, have a different meaning?

Am I catching on?

 

[PeterR]

Quote:

Originally posted by Steve999 HOWEVER, it seems to me to indicate that the 7506 out of a 120 ohm jack will be about 1 db BRIGHTER at 20 khz, as well as introduce a 1 db increase in the LOW bass. (Not a big change, but not a good thing for the V6s!) Am I missing something?

No, you aren't, that's exactly what would happen

Quote:

Is this the general pattern?

Yes, the basic shape of the impedance curve is the same for all dynamic headphones, but by how much the impedance varies is different. The impedance plot for Grados for example looks essentially flat, so there should be little influence of the source impedance.
http://www.headphone.com/graph.php?graphID=133
Quote:

Would running the HD580s at 120 ohms then likely make them a just a little brighter in the upper frequencies (10khz to 20 khz)

A little, but already at the upper edge of your hearing.
Quote:

as well as a just little more pronounced in the LOW bass?

Around 70Hz IIRC, so not really low.

Quote:

I am also inferring that the response curve differences from a 0 ohm output vs. a 32 ohm output would be quite small but measurable.

In any case smaller than from 120 ohms, yes.

Quote:

Would running headphones out of a line out or variable line out cause much greater distortions of intended frequency response,

Depends

I've just flipped through a few Stereofool back issues and looked up the output impedances of some sources:

Classé CDP-10 47.8 ohms
Sony SCD-XA9000ES 109 ohms
Burmester 001 66.2 ohms
Krell SACD 119 ohms
Linn Unidisk 209 ohms
Mark Levinson N0.390S 24 ohms
Theta Digital Generation VIII 11.5 ohms
Esoteric DV-50 448 ohms
Gamut CD 1 56 ohms
Ayre CX-7 48 ohms
Musical Fidelity Tri-Vista 58 ohms

So you see there's quite some variation, some of them have usefully low values - if the current capabilities are sufficient, some sources could very well be suitable to drive headphones directly.
Quote:

or do those resistance specifications have a different meaning?

You need to be careful, the specs often state the lowest impedance the output is able/recommended to drive, not the output impedance.

Quote:

Am I catching on?

Guess so

 

[Steve999]

Awesome, PeterR, I think I get it! Thank you so much!

 

[JaZZ]

Starting from this impedance curve you can calculate the distortion of the amplitude response when e.g. a 600-ohm resistor is switched in series to the HD 600.

The bass resonance peak at 100 Hz amounts to 540 ohm; the lowest impedance appears around 2 kHz, with 310 ohm, and the voice coil's inductivity increases the impedance at 20 kHz to 375 ohm. The task is calculating the amplitude damping at the specific frequencies caused by the 600-ohm resistance, by adding 600 ohm to the impedance amount at the specific frequency and divide the original impedance by this sum.

100 Hz:
540 : (540 + 600) 1140 = 0.474

2 kHz:
310 : (310 + 600) 910 = 0.34

20 kHz:
375 : (375 + 600) 975 = 0.385

The resulting numbers, multiplied by 100, show the percentage of the remaining sensitivity: 47.4%, 34%, 38.5%. Now what's most interesting is the relation between the amplitudes at the different frequencies. Given the 2-kHz value as a reference...

47.4 : 34 = 1.4
38.5 : 34 = 1.13

...the amplitude at 100 Hz is 1.4 times higher than the one at 2 kHz, and the amplitude at 20 kHz is 1.13 times higher. This makes 3 dB for 100 Hz and ~1 dB at 20 kHz.

Peter, correct me if I used a wrong basis for the calculation, but I think one gets the point.

Of course this is an extreme example to demonstrate the effect, but even considerably lower serial impedances have their audible effects too.

As to driving headphones directly: I'm one of the few who does or did that, but meanwhile I had to learn that unlike my own gear most of the affordable players have output impedances of 200 ohm and more.

 

[Steve999]

Thanks again for the help PeterR, and thanks for the reply, JAZZ. That pretty much reconciles everything and it all makes a lot of sense to me now.

Quote:

Originally posted by JaZZ As to driving headphones directly: I'm one of the few who does or did that, but meanwhile I had to learn that unlike my own gear most of the affordable players have output impedances of 200 ohm and more.