Amplifier Output Impedance – It Matters
When looking at getting a new (or your first) headphone amp, it can often become rather overwhelming when attempting to sift through reviews, specs and other online posts.
The specifications, in particular, can be a daunting thing to look at if you don’t know how to make sense of them. Which specs are relevant in your particular use? How can you tell, from looking at the specs, what makes one amp better than another?
Also, just for the sake of clarity, when we talk about headphone amps, we don’t mean just dedicated headphone amps, as this applies to the headphone output of DAC/AMPs as well as music players.
A couple of things are pretty self-explanatory, such as the power output. Generally, the more power output is available, the more demanding headphones the amp can drive. Some headphones require very little power to get their drivers moving, whilst others can require a substantial amount more to get to the same volume level. Then there’s also the voltage swing to consider, meaning when the amp switches the power output between negative and positive, how much of a voltage differential is there between each swing? Again, broadly speaking, more voltage swing often equates to better dynamics since dynamics, by definition, is how big of a difference there is between the quietest parts and the loudest parts of music.
But then there’s another specification which is often referenced – the output impedance.
Usually, this spec is brought up when taking into consideration high-efficiency and low-impedance headphones/earphones. The reason for this is because we need to consider the damping factor.
The damping factor is not the main topic of this conversation, but it’s worthwhile just briefly going over it.
As I understand it, the damping factor is the ratio of the impedance of the headphones to the output impedance of the amplifier, and the general rule-of-thumb here is that you’d ideally want the amplifier’s output impedance to be no more than 1/8th of the headphone’s impedance.
So, if you have a set of headphones with a 32-ohm impedance, then ideally, you’d want to stick to a headphone amp which has an output impedance of no more than 4.
So, why is the damping factor important? Well, this has to do with controlling unwanted movement of the driver. The entire purpose of a driver is to move air, and since moving air is what we perceive as being sound, and unwanted driver movement may result in what we hear as certain sounds being added to the music.
Think of it a bit like the difference in bounce between a bouncing-ball and perhaps a tennis-ball. If you were to throw a tennis ball as hard as you can at the floor, and then did the same for a bouncing-ball, you’d notice that the bouncing-ball will bounce higher than what the tennis-ball would simply because the difference in their materials and structure don’t allow for the same amount of conservation of kinetic energy. So, think of that difference in materials and structure as being the damping factor in this discussion.
When the amp outputs a quick electrical signal (impulse) to the headphones, what follows is the driver moving accordingly to the impulse, and then there are a few little “bounces” of the driver after that. This is called ringing. The less ringing you have, the less those unintended driver movements can interfere with the next impulse. Music is essentially just a bunch of impulses, so it’s easy to see then how the unintended driver movement immediately following one impulse can then influence the next impulse, and thus have an effect on the music overall.
So, the better the drivers are dampened, theoretically the more accurately the driver can convert the electrical signal being fed to it into kinetic energy and thus sound energy.
That’s just a brief overview of what the damping factor but there is a whole lot more that can be covered on that topic. But, as I said, this main topic of this discussion is not the damping factor, but rather about how the output impedance of the headphone amp can actually change the frequency response of the headphones/earphones.
This is a concept which really does not get talked about
enough. But, my biggest gripe with this is that, because this factor is not
considered, users and reviewers alike can far too often mistake the change
brought about because of the output impedance as being a change brought about
by the amp, DAC, or music player itself.
To illustrate what I mean, let’s take a look at some frequency response graphs
which can be found on https://reference-audio-analyzer.pro.
Here we’ve got the response of the Shure SE535, and as we
can see, it’s got a pretty flat response. Now let’s say that you wanted to
compare the sound of, for example, the Hiby R6 and maybe the FiiO M11. I’m sure
that you would agree that, for the purpose of hearing the differences between 2
players, it would make sense to use the same headphones to listen to both
devices, right? You can’t have one set plugged into the one player and then
have a different set of headphones plugged into the other player, and then swap
between them because you’d be far more likely to be hearing differences between
the headphones, not the players.
And yet, if you don’t pay attention to the output impedance, and in particular
how your headphones may react to different impedance, you might essentially end
up doing pretty much just that – listening to 2 different frequency responses,
rather than actually hearing differences between the players.
Above is a graph showing how the frequency response of the
SE535 changes as the output impedance of the device its connected to changes. With
reference to the examples used here, the M11 and R6 have output impedances of 1-ohm
and 10-ohm respectively.
By looking at the graph, we can see how someone might listen to the R6 with the
SE535 and describe it as sounding less clear, or perhaps less detailed given
the change in the frequency response between 4kHz and 10kHz.
In much the same way as you have to pay attention to properly volume-match (not
by ear) the source devices when making comparisons, it’s just as critical to
use headphones/IEMs that have a stable frequency response, i.e one which is not
affected by the output impedance of the device.
This may also, to some extent, explain why some users might have different
experiences/perceptions with particular headphones, as the sources devices they
use are different.
Below are a couple of more graphs of some more popular headphones/IEMs which are often referenced as being “revealing” when in fact the effect of the output impedance of a device should immediately disqualify them form being used in such comparisons.
Then the last thing I wanted to mention are tube amplifiers, in particular with reference to headphones such as Sennheiser’s HD6-series. We’ve often heard people say that the 600 and 650 sound better when connected to a tube amp…like it makes it just seem fuller and that the bass is more textured. Granted, tube amps can certainly have a noticeably different sound to solid state amps…but you know what else tube amps tend to have? Higher output impedances. Below are the responses of the HD600 and HD650 which shows why using a device (amp) with a higher output will alter the sound character slightly, irrespective of whether it’s a tube-based design or solid state.
So, what’s the point of all of this? Does this mean that you
should stay away form any headphones/IEMs that have unstable and unreliable
frequency responses? Absolutely not…not for the purpose of just listening to
music anyways.
However, if you are a reviewer, or when simply making comparisons between different
source devices, I strongly urge you to only use headphones/IEMs that are immune
to such fluctuations in frequency response. As with any experiment, it’s vital
to identify constants and variables, and to eliminate as many variables as
possible in order to only test one at a time. A headphone/IEM which does not
have a stable frequency response is just another variable, one which can make
it extremely difficult for you to determine if what you’re hearing is a change
in which ever device you’re trying to test (player, DAC/AMP, amplifier, etc) or
if the changes you are hearing are actually mostly (or even entirely) due to the
change in frequency response.
Well, I hope that made sense and that it will be of some use
to someone. As far as I am concerned, the job of a reviewer is to inform the
public, much like a journalist. As such, I feel reviewers, all of us, have a
duty to ensure that we are not just being honest with our readers/viewers, but
that we are also factually correct to the best of our knowledge.
Being honest does not mean you are correct, and so it’s imperative to constantly
reconsider your methods in order to identify inefficiencies and ultimately
improve and refine that method.
