PDA

View Full Version : Impedence matching



Jeff Root
2010-Sep-19, 10:15 PM
Anyone happen to know what impedance portable audio players
want for their earphones? I'm thinking of buying a small voice
recorder, and I noticed that one high-priced one specified an
impedance of only 8 ohms.

I have in-ear earphones that are 18 ohms, and headphones that
are 55 ohms. I *presume* that either of these would produce
relatively low volume when plugged into something designed
for 8 ohms. I'm hoping that audio players and voice recorders
are more typically designed for an impedence of 16-20 ohms.

-- Jeff, in Minneapolis

ShinAce
2010-Sep-20, 03:17 AM
Yes. Generally speaking, the output 'wants' an impedance over 120 ohms. Bigger loads( <120 ohms) will cause the bass to roll off. However, even using 16, or 32 ohm headphones will not cause any audible issues. If you were going to do this to work in a sound studio, you could temporarily use a headphone amplifier to 'buffer' the output of the recorder.

For example, a sound card on a computer is very similar to a CD player, which in most cases, is superior to an mp3 device. Even then, I have a homemade headphone amp that I use when listening on the PC. Generally because I'm evaluating the quality of the recording.

Do not expect the headphone impedance to affect volume. An 8 ohm headphone is simply an inefficient one. An 8 ohm loudspeaker is simply a standard one. You'll have to try them out to see which gives the loudest volume.

A portable device will safely drive an 8 ohm load, but don't expect the volume to be impressive. As the impedance of the headphones drops low, you need less voltage to drive them, yet more current. 99% of amplifiers can't compensate. They are constant voltage amplifiers. As a result, you run out of current before you deliver maximum voltage. The most efficient impedance for delivering power is about....120 ohms...for these devices.

With all that said, try all your headphones out. Compare the sound quality between them. Once that's done, see which one will tolerate the most volume before distorting. Try to stay away from the 8 ohm model if you can.

The short answer to your question is that the output impedance of thse devices is on the order of (16 ohm to 300 ohm).

The output impedance on my stereo amplifier driving my bookshelf speakers is 0.005 ohms. This is done to create a 'damping' factor. The amplifier is able to supply and sink the power necessary to the woofers so that their self generating electro motive force does not cause distortion at low frequencies.

Jeff Root
2010-Sep-20, 05:36 AM
To be clear, I'm not interested in getting volume, just the best
possible quality, which I --again-- *presume* is lost when the
impedences don't match. When the speaker impedence is too
low, I presumed that the volume would be high but distortion
would also be high. When the speaker impedence is too high,
I presumed that the volume would be low, and noise would be
high. Is that wrong?

The recorder with the 8 ohm output is the most expensive one
listed, and the only one I looked at that stated impedence.

I'm surprised that you say an 8 ohm headphone is inefficient.
I *presumed* that the lower the impedence, the more efficient.
Did I have it backward? Or is it more complicated than that?

The little 18 ohm earphones are significantly louder than the
55 ohm headphones, which seems funny, but what I expected.
More annoyingly, the headphones, which cost more than three
times as much as the earphones, don't sound any better.

-- Jeff, in Minneapolis

ShinAce
2010-Sep-20, 04:04 PM
Not quite. Distortion will be at a minimum when the headphone impedance is highest.

As an example, a power amplifier has an input imepdance of 10k - 300k ohms. Adding an amplifier will not induce distortion, it will actually lower it because of the easy load the device is now driving.

p.s. Most portable devices will handle a short circuit without problems, so there shouldn't be a risk of burning it. Use whatever headphones sound best to you.

Jeff Root
2010-Sep-20, 07:33 PM
I'm still confused.

First, is the reason the headphones don't sound as good as the
earphones that the earphones, having low impedence and not
requiring much power, don't need an amplifier, while the larger
headphones, with higher impedence and more mass to move,
do need an amplifier? So, with an amplifier, the headphones
would likely sound as good as or possibly better than the little
earphones? (Which sound pretty good for as long as they stay
in my ears.)

"Distortion will be at a minimum when the headphone
impedance is highest."

Surely you don't mean "the higher the headphone impedence,
the less distortion, regardless of the device's output." The two
do need to be matched, don't they?

Why did I choose the low-impedence headphones, then?
I had a choice of 55 ohms or 600 ohms. I forget what my
reasoning was for choosing the 55 Ohm model. I don't recall
knowing the requirements of any device I might plug them
into at the time I made the selection.

And why do ordinary speakers have such low impedence--
only 8 ohms? What is the advantage of low impedence if
distortion is lower and efficiency is greater with high
impedence?

Exactly what happens if the speaker/headphone impedence
is too low or too high for the device's output?

-- Jeff, in Minneapolis

Hornblower
2010-Sep-20, 10:52 PM
I'm still confused.

First, is the reason the headphones don't sound as good as the
earphones that the earphones, having low impedence and not
requiring much power, don't need an amplifier, while the larger
headphones, with higher impedence and more mass to move,
do need an amplifier? So, with an amplifier, the headphones
would likely sound as good as or possibly better than the little
earphones? (Which sound pretty good for as long as they stay
in my ears.)

"Distortion will be at a minimum when the headphone
impedance is highest."

Surely you don't mean "the higher the headphone impedence,
the less distortion, regardless of the device's output." The two
do need to be matched, don't they?

Why did I choose the low-impedence headphones, then?
I had a choice of 55 ohms or 600 ohms. I forget what my
reasoning was for choosing the 55 Ohm model. I don't recall
knowing the requirements of any device I might plug them
into at the time I made the selection.

And why do ordinary speakers have such low impedence--
only 8 ohms? What is the advantage of low impedence if
distortion is lower and efficiency is greater with high
impedence?

Exactly what happens if the speaker/headphone impedence
is too low or too high for the device's output?

-- Jeff, in Minneapolis

Impedance is an electrical characteristic that has nothing to do with a speaker's mass. It is a complex type of resistance, and it determines how much current flows in the coil for a given amount of output voltage from the amplifier. In theory a coil with higher impedance but more turns can generate just as loud a sound with the reduced current.

A low-impedance speaker draws a large amount of current, and it can overload an amplifer that is not designed for such a heavy load. Some types of speakers benefit from electrodynamic damping to prevent resonant distortion in the bass notes, and if I am not mistaken, properly matched low impedance is needed to make that damping happen. My educated guess is that audio designers found from experience that 8 ohms is ideal for most speakers for this purpose, and they designed the amplifiers to be able to work at that level without burning up or distorting the sound.

Headphones need only a trickle of current by comparison, and here the task is to keep the power way down. Another of my educated guesses is that the small, lightweight transducers in the headset do not have the damping needs of a large, heavy speaker cone. If that is the case, then we need only to avoid overloading the phone jack circuit with impedance that is too low. If for some reason your device cannot handle a 55 ohm headset without being overloaded, by all means try the 600 ohm set. Higher impedance in the headset should have no adverse effect.

Once again, do not take my words blindly as gospel. I welcome any response from speaker and amplifier experts.

Geo Kaplan
2010-Sep-20, 11:10 PM
Actually, the electrical impedance you measure across the speaker terminals is a function of both electrical and mechano-acoustic phenomena. The mass of the speaker, the medium it's in, etc. all do make a difference. Indeed, the more efficient the transducer, the more these will matter.

There was an earlier statement that "higher impedances will always lead to lower distortion." While frequently true, it is not necessarily always true. It depends on a variety of factors (e.g., total supply voltage, maximum current handling capability of the drivers, and the feedback topology).

As to Jeff's original question, most of these amps are designed for a pretty wide range of impedances, while delivering "good" sound. As already pointed out by others, the chief tradeoff will be in how loud you can crank up the volume, if you deviate too much from the design-center value.

G O R T
2010-Sep-21, 10:49 AM
While the internal output impedance of such devices is usually quite low, small size limits the DC decoupling capacitor size. This places a lower limit on speaker impedance for low frequency roll-off. Braided metal foil conductors also limit energy transfer at low driven impedances.
Traditionally , mini headphones have been in the 36 ohm area and there has been little reason to change that.
Don't neglect bitrates on MP3 players. I cannot tell the difference between 128 and 256 but my brother can.

jj_0001
2010-Sep-22, 09:42 PM
Ok. You don't do impedence matching with most modern amplifiers, including those cheap and lousy ones in portable players. There are two specs, one of them is "output impedence" which is often very, very small. The other one is "drive capability" and THAT is what you care about, entirely.

The drive capability in ohms, for a non-reactive load, is what you get when you divide the maximum voltage output (peak, volts) by the maximum current outout (peak, amps) that the device can muster. Reactive loads will need to be higher in impedence in all cases.

Impedence matching does not matter until transmission line length appoaches 1/10 wavelength, which at 1 foot/nanosecond, or 10^9 feet/second, is about 10^9/(2*10^4), which is about 4166 feet in wavelength, or 416 feet in air (for the 1/10 wavelength), more than half of that in a wire, before impedence matching begins to have any useful meaning.