LABORATORY TEST REPORT

All measurements of the iFi Zen Can headphone amplifier were made using the unbalanced (single-ended) output.

So far as the amount of power that’s available from that socket, it delivered just under 7.5 volts RMS open circuit with a

1kHz test signal, a voltage that was reduced by only a smidgeon when loaded down with 300 and only a smidgeon more when the test frequency was dropped down to 100Hz, where it delivered 7.38 volts RMS, which equates to nearly 185mW (22.7dB). By way of comparison, into the same load at the same frequency the AudioQuest DragonFly Cobalt delivers around 12mW (11dB) and an iPhone Lightning dongle tops out at 1.6mW (2.1dB).

When tested using a 16 load, the iFi

Zen Can delivered just over 4V RMS, which translates to one watt (1,000mW), which is some 30dB above headphone sensitivity rating. Again, by way of comparison the DragonFly Cobalt managed 16mW (12dB) and the iPhone Lightning dongle 25mW (14dB) into the same load.

All of which means that the iFi Zen Cans can deliver sufficient current to drive any headphones I can think of to satisfying volume levels.

But volume level isn’t all there is to it.

There is the matter of output impedance. Again, if a headphone amplifier has a large output impedance, it acts as a voltage divider for any headphones in which the impedance varies by frequency — which is most of them. So the level of the signal fed to the headphones varies by frequency. One problem is that the majority of devices don’t specify this, so you have no idea whether you’re likely to run into such issues. That won’t be a significant issue with the iFi Zen Can, for which an impressively-low impedance of just 0.55 was measured.

So there will be no problems with unexpected frequency response variations.

Also, the output impedance has at least some effect upon the control of the amplifier on the driver. Non-linearities in a driver actually generate a back voltage, known as back-emf (where emf stands for electromotive force). A speaker — or headphone — driver is kind of like a dynamic microphone.

But any signal it produces can be defeated by shorting it out. A low impedance output provides the shorting out function. This is also known as damping factor, a calculated specification rarely mentioned these days.

The frequency response measured for the iFi Zen Can, as shown on Graphs 1 and 2, was perfect for all practical purposes, being perfectly flat and down by less than 1dB at 120kHz.

The effect of using the XBass is shown on Graph 1 and you can see that it elevates the low frequency response by 2dB at 100Hz and continues to increase it with decreasing frequency, so it was +9.0dB at 20Hz. Presumably this continues into the infrasonic frequencies, which is not what happens with, for example, a typical bass tone control. A well-designed bass tone control will limit (‘shelve’) the amount of boost it applies at a certain frequency, so no more will be applied at lower frequencies.

The iFi Zen Cans can deliver sufficient current to drive any headphones I can think of to satisfying volume levels

Typically, bass will be boosted increasingly down to, say, +8dB at 50Hz, after which that level of boost is maintained for all frequencies below 50Hz. iFi’s XBass circuit, on the other hand, uses a simpler circuit, one where the level of boost increases as the audio frequency decreases, with no shelving applied at all.

This kind of circuit is only possible on the iFi Zen Can because of the massive amount of headroom available. Most other headphone amplifiers do not have the amount of headroom available to deal with the boost and would rapidly run into clipping should there be significant bass content.

That’s an excellent result, but there are very few headphones that require anything like the full output level that’s available from the iFi Zen Can. Let’s face it, most headphones can be driven to very pleasing levels with just a volt or so.

So what the iFi Zen Can provides is heaps of headroom, which is very useful when you do such things as boost the bass by 8.5dB at 20Hz (an 8.5dB boost being 2.66 times the voltage). The overall signal-to-noise ratio measured for the iFi Zen Can is a very good 112.7dBA, but the graph of the iFi Zen Can’s noise floor (below left) showed a significant amount of 50Hz breakthrough, along with surprisingly high levels of odd-order harmonics of that mains frequency. So I started wondering what was going on. Power supply perhaps? The review unit came with the $80 low noise power supply option, rather than the standard one, and this power supply is quite highly regarded, so I unplugged the power supply completely to run some checks on it and so double-check the quality of the mains power. Finding nothing amiss with either, I reconnected it and re-ran exactly the same test. The overall S/N ratio came in just a little better, at 113.5dB, but all the noise spikes had almost disappeared. There were still noise components at 50Hz and 150Hz, but they now peaked at –123dB and –129dB respectively, compared to –101dB and –109dB in the original measurement. These before and after traces are shown in Graph3. It’s unlikely that the spurious noise noted will be present in any case, and if it is, it’s likely to be completely inaudible. But in the event that it is, it seems that simply removing the power pack from the mains socket, waiting a couple of minutes, then plugging it back in, will remove it. The measured THD of the iFi Zen Can was 0.00115% with the output level set to 1.2 volts RMS, a level selected because it’s approximately 4dBu, a standard Pro level. At 0.5 volts RMS, this time running into a 16Ω load, THD still measured only 0.0025%. That’s mighty impressive. Overall, all the test results returned by the iFi Zen Can were mighty impressive — it’s an outstandingly good headphone amplifier.

Overall, all the test results returned by the iFi Zen Can were mighty impressive — it’s an outstandingly good headphone amplifier