6.35mm home theatre receiver headphone output

Are you dissatisfied with how your headphones sound? There’s a good chance that the device driving them is making them sound bad. I’m not talking about airy fairy matters of high-end sound quality. I’m talking about actual, measurable colouration of the signal.

What are these source devices which are so deficient? No, I’m not talking about your laptop computer, although it probably suffers from the same problem. Nor am I talking about your phone – many don’t have this problem at all. I’m talking about respectable, sometimes very highly regarded, audio gear from manufacturers large and small.

If you listen with headphones plugged into an integrated amplifier or a home theatre receiver, there’s a very good chance that the amp is colouring the signal.

What is wrong with those headphone outputs?

It comes down to one simple measurement: output impedance. To understand what effect this has we’ll need to briefly discuss voltage dividers.

You can think of any voltage source as consisting of a pure voltage, plus an inline resister. A headphone output is a voltage source and because we live in the real world, all headphone outputs have some level of inline resistance to the flow of current. If we measure the voltage across the headphones, we’ll find it less than the pure source voltage. Perhaps only very slightly less, perhaps a lot less, depending on how large the internal resistance is.

Final Audio B3 IEMs

The higher the internal resistance, the lower the voltage applied to the headphones. When you plug headphones into a headphone socket, the amp will be producing a certain voltage. Some of that voltage will be applied to the notional internal resistance. What's left over will be applied to your headphones. The amounts are proportional to the resistances.

So, if your headphones are a 50-ohm load and the internal resistance of the headphone amplifier is 100 ohms, your headphones will receive precisely one third of the voltage that the amplifier is providing. The other two thirds is burnt up as heat in that internal resistance.

Alternating signals

We don’t normally talk about the electrical resistance of headphones, but their impedance. The word takes into account the complications of alternating current signals, which is what audio signals are. In most headphones the electrical resistance varies by frequency. At some frequencies it will be relatively low, while at others it will be high. Consider the Focal Elear headphones. These are a fairly straightforward design of dynamic headphones – one single driver per side, no crossover – with a nominal impedance of 80 ohms.

The former website Inner Fidelity – now part of Stereophile – tested their impedance. The graph (PDF - I’m not reproducing it here for reasons of copyright) starts at 10 hertz, at which point the headphone impedance is 100 ohms – already higher than the nominal 80 ohms. The impedance rises sharply to hit a peak of 310 ohms at 55 hertz, then falls away to around 80 ohms at 400 hertz, where it remains.

If the headphone output has a high impedance, then the source will be delivering a significantly higher voltage to the headphones at 55 hertz than at, say, 1000 hertz.

What is the impedance of those headphone outputs?

Over the past few years I’ve measured the inline impedance of quite a few headphone outputs. Major brand home theatre receivers all have a high output impedance. The best in this regard is Yamaha. Its receiver headphone outputs are typically just under 100 ohms. Yes, I know, and that’s the best! Other characteristics of their performance suggests to me that the company uses a dedicated op amp to drive the headphones. But a recent entry-level Yamaha stereo receiver had around 460 ohms in line. And that suggests to me that the headphone output is simply connected to the main amplifier, with a 460-ohm limiting resistor placed in line.

The good thing about that design is that it tends to provide plenty of power, especially to high impedance headphones. But we’ll see in a moment the negative impact it has on other aspects of performance.

Sennheiser HD 560S headphones

Home theatre receivers from Marantz, Denon and NAD also appear to use that inline resistor dodge on their headphone outputs, with typically 470 to 480 ohms of inline resistance. By comparison, the neat little Marantz M-CR612 mini system only places 75 ohms in line. One multi-thousand dollar integrated amplifier from a highly reputable brand passed through my office a few months ago. Oh, why be coy? It was the Rega Aethos. It had a bit more than 100 ohms inline. An Electrocompaniet integrated amp from Norway was significantly better at only 11 ohms, suggesting the use of a separate headphone amp. A combination streamer/amplifier from the famed British firm Linn placed 57 ohms in line.

I think all those impedance levels are unacceptably high. But are they inevitable?

Low impedance headphone outputs

The other day I reviewed the ALO Audio Pilot portable DAC and headphone amplifier. Its output impedance? Around 1.2 ohms. That low an impedance should have effectively no affect upon the sound. How about the Astell&Kern &Futura SE200 portable player? Just under one ohm for the ESS output, and 0.75 ohms for the AKM output. Astell&Kern seems to make a habit of this. The markedly lower-cost SR25 came in at 0.9 ohms. The iFi ZEN CAN headphone amplifier imposes an output impedance of just 0.55 ohms.

iFi ZEN CAN headphone amplifier

The AudioQuest DragonFly Red and Cobalt come in at just 0.2 ohms, which must be close to a record.

But surely all these are expensive. Well, no, most are only a few hundred dollars. But even really cheap stuff – the $15 Lightning headphone adaptor for iPhones has only 1.5 ohms of inline resistance.

So, no, a high impedance on the headphone output is not at all inevitable. It’s just that it’s cheaper to put a resistor in line with the existing amplifiers than to install a dedicated headphone amplifier.

What does high inline resistance do?

But does it matter all that much?

Well, if your headphones have an impedance that varies by frequency, having a high inline resistance will change the frequency balance of the signal that the amplifier is delivering to the headphones.

Some headphones have an even impedance across the range. Planar Magnetic ones commonly do, so they won’t be much troubled by this issue. But the great majority of headphones and earphones do have a variable impedance. And the lower the nominal impedance, the more effect it has.

I tested two sets of headphones and one set of earphones I have on my desk at the moment: the aforementioned Focal Elear – now out of production – the new Sennheiser HD 560S and the Final Audio B3 in-ear monitors. I faked a high impedance output by putting 466 ohm resistors in line with the low impedance output of the iFi ZEN CAN headphone amplifier, and measured the signal as delivered to the headphones. This is not the sound produced by the headphones. It’s the signal that the headphones are receiving.

Measurement results

Remember the big bass bump in impedance for the Focal Elear headphones mentioned above? Well, that translates into a big bump in the bass output: a more than 10dB lump at 70 to 80 hertz.

Don’t like the sound of Focal Elear headphones when listening to your home theatre receiver? Well, that’s why.

The Sennheiser HD 560S headphones apparently have a more even impedance curve and a higher nominal impedance – 120 ohms – so they’re less effected. Still, there’s a 3.5dB swing in signal input level, with a bass boost at around 60 hertz.

The Final Audio B3 IEMs use two balanced armature drivers for each unit and have a nominal impedance of just 19 ohms. I think we can see from the graph where the crossover is. With a home theatre receiver, expect a generally quiet bass, a nasty 5dB peak at 3kHz and that rising treble.

Here's the graph showing the results:

Signal delivered via inline resistance to three headphones

The solution

There are no real standards for the output impedance of headphone amplifiers. If there are, they are certainly not adhered to. They are only specified by some manufacturers of higher end gear, so much of the time you just won’t know what affect it’s likely to have, at least until you’ve listened.

And without extensive testing, there’s no way to be certain whether any set of headphones is relatively immune thanks to an even impedance across the audio band, or will be wildly influenced by inline resistance. Nor, indeed, how it’s likely to be influenced. As we saw, some headphones had a lumpy bass, while the earphones had a mid-treble peak and rising high end.

The best solution is the one we recommend for most things: take your favourite music with you to a reputable dealer who stocks the headphones and amplifier on your shortlist, and listen. If you’re looking at specifications, check for an output impedance specification on the headphone amplifier. Anything below two or three ohms should be fine, but of course the lower the better.

Read reviews that actually measure this specification. That’s what we’ll be doing here. And put on your shortlist brands with a habit of producing gear with low output impedances. That includes such brands as FiiO, iFi and Astell&Kern.

Or, for that matter, the earphone adaptor for your iPhone!