Headphone & Amp Impedance Questions? Find the answers here!

I've changed the 'required' to 'desired'. Makes more sense.

 

Quote:

Hmm... not sure if you're still interested in this question but... let's give it a try...
 
Yes, there is a theoretical, ideal ratio. Zearphone/Zout should be ideally infinite.
Other words, Zout should be ideally zero. This holds on the earphone end,
but not for all circuitry, meaning that some mid-stages (some modern ic amplifiers)
are current driven, not voltage driven, and in these cases that is not necessarily true.
 
In practice, however, you see amp specs like this: I can deliver 30mW to a 32ohm headphone,
but only 10mW to a 120ohm load, for example. The maximum power an amp will deliver
is, of course, defined by design, and the delivered power at any specific load goes by V^2/Z. 
That's a nonlinear relationship, but you can see the bigger the earphone impedance, the lower
the power the amp will give. You can see nothing was said about the output impedance, 
but that's implied in the 30/32 10/120 spec.
 
 
Now, one should understand that the output impedance of an active element (such
as an amplifier) is a v/i, dynamic, frequency dependant, relationship, not a physical resistor.
This is important because, if one thinks about maximum power transmission through the amplifier,
the ideal ratio would be 1, that is, the earphone impedance should be exactly the
same as the output one. For example, with a fixed frequency source (your power outlet
is an example), suppose you connect a 6ohm resistance in series (that would be your
output impedance), and after it you will try some lamps, say, a 3ohm, 6ohm, 12ohm, 1000ohm.
The lamp that will bright the most is the 6ohm one.

 

Yeah but the power transfer theorem is about choosing a load resistance with a given (fixed) source resistance. In your lamp example you'd be best off not adding the resistor at all (assuming the lamp can handle that).

 

Quote:

Hi Mr. Threepwood , yes, that was for the sake of the example... the resistance would be
the output impedance of an amplifier, in a system with fixed frequency. The resistor would
already be part of the system, say, inside the power outlet.

 

First I'd like to say thank you for this very helpful thread.
 
I'm still struggling to understand how to figure out how much power I need.
 
Say I'm comparing the headphones Hifiman he-500 (38Ω and 89db) and the amp fiio e07k (> 220 mW@32Ω /> 250 mW@16Ω).
 
Using the formula you provided: inverse log ( (desired db - headphone db) / 10 ) I got 0.398 (if I desired 85b) or 0.398 x 10^(-3) (if I desired 65db). 
 
I assume the answers are in watts: so 85db meant 398mw and 65db meant 3.98mw? I'm really confused. 
 
Also, would it matter if the resistance of the headphone is out of the range of the resistance of the amp? Or is this not how it works?
 
Thanks.

 

The headphone datasheet at IF says it needs 0.31 V to reach 90 dB SPL. That's about 87 dB/mW (using the measured impedance at 1 kHz of 47 ohms) or 100 dB/V.
 
10^((85-87)/10) = 0.631 mW, yes that is milliwatts, in watts: 0.000631 W
 
10^((65-87)/10) = 0.00631 mW

 

nice thx for the reference

 

Quote:

 
Sure np.

 

hi there i was just wondering if there is even a need to amp a low impedence headphone like the noontec zoros? if so how much of a difference does it make? i'm currently thinking about getting a fiio e 7 or 11 and maybe he fire eye mini
 
this is from innerfidelity
Impedance and phase plots show a 20 Ohm headphone with some modest resonances at 2kHz and 6kHz, likely originating from resonances behind the driver. With 16mVrms needed to achieve 90dB at the ear this is a very efficient headphone, and will easily be driven to loud levels from portable players. But the isolation plots shows that this headphone does not isolate well at all making this a headphone that can be used portably, but not for listening in loud environments. On the other hand, you will be able to remain aware of your surroundings, so using these cans portably in quite neighborhoods will work well.

 

Quote:

 
Innerfidelity is pretty spot on with their analysis. From the data it seems it doesn't really require an amp.

 

Quote:

thanks, for your reply, what would an amp even do anything? change in sq?

 

Louder (so you can go deaf faster). If your current "thing" is noisy or has massive channel imbalance (you'd know about both of these, no question about it) it might change that, depending on how you source the signal into it - but by and large: louder.

 

Quote:

thats perfect response! even better i can save some cash!
 
i always thought an amp would boost all the frequencies (bass,mids and treble) ( the noontec zoros have a sharp notch at about mid treble) figured an amp could fix that
i was wondering then wouldn't the noontec zoros be a gem then? its low impedence and its relatively flat signature perfect for portable use!
 
why would anyone want headphones that have a high impedence even if its for home use ? if all it does it suck up more power? Hypothetically two headphones were identical in sound quality but the difference in impedence, logically the lower impedence is more economically friendly?

 

If it boosted all frequencies it would just increase the overall signal (= louder). But it will do *nothing* of the sort (at least not as you're proposing) - all an amplifier will (should) do is apply gain to the signal (that is, make the signal larger); it does not magically improve (or transfigure) the sound quality. Depending on the interaction with the load it is presented, and relatively speaking, it may act on the frequency domain to an extent, but an EQ is a better bet if you want "big, dramatic, night and day" changes, and if you're looking to change something on the time or radiative domain - look at the speakers/headphones.

Impedance is not a "quality" or "performance" spec, despite what a lot of marketing suggests. It's just a characteristic of a given driver - some will be high impedance by nature, some will be low impedance by nature, and there is no rule of thumb that one is better than another. Higher impedance doesn't actually even require more power (power requirement is determined by sensitivity and your target SPL (which should always be <85)), but it *may* require relatively proportionally more voltage (which is why some portable devices have problems with some high impedance headphones).

Generally with headphones sold for at-home use, the designer/manufacturer doesn't have to care either way, because the assumption is (or at least was, for many years) that you're plugging them into a stereo receiver or integrated amplifier which will drive its headphone jack with the "main" amplifier section built into it (with resistors in the way, to protect both ends). Usually these devices will put out a few watts per channel for headphones (which is more than any ten people should ever need; seriously). However that's changed in recent years, and people have started to want to take their headphones with them and plug them into things like CD players and iPods (which have inherently limited voltage swing, due to their battery power). So this is where low impedance headphones have an edge.

It's all relative, basically - if you're talking from the perspective of an AC powered integrated amplifier or CD player, it really truly doesn't matter what specifications your headphones have; if you're talking from the perspective of an iPod, it's better go with low impedance and high sensitivity, to get the best out of the device (and the less it has to put out, the longer the battery lasts); if you're talking from the perspective of a marketier - oh good lordy, your headphones better be at least one trillion ohms, and your amplifier's output better have negative impedance across it, so that you can show fantasy numbers with a ton of 0's and tell the customers about specifications (that have no actual bearing on their lives) and move more units. :tongue_smile:

 

Quote:

thanks for your time in writing that  response, i really enjoyed reading it =) i'm still new to the specs on headphones and been only watching some youtube videos about them saying things like (bigger drivers= more bass) i now feel like an idiot listening to that. 
 
i am using an iphone 3gs and i'm trying to look for an eq that fixes the noontecs notch at about 4k cause otherwise i'm very pleased with all other aspects of the headphone except the lack of isolation
 
the iphone eq is pretty hard bad, you can't make a custom one, so i found myself leaving it alone (the knotch at 4k doesn't really matter too much to me but i would like the possiblity to fix that if i so choose)