56 December 2014
protective film on both sides.
Removing the film was
probably the most annoying part of
building the kit — it was easy on
the flat surfaces, but the curved
portions were a real challenge due
to the cuts. It probably took me 20
minutes of picking little bits of the
protective wrapper off to get the
acrylic panels ready to finally
After I got the film off, the
residual static charge it left on the
acrylic was enough to attract every
speck of dust from around the
room onto its surface, so I wiped it
down with a clean (just barely)
damp cloth and that seemed to
take care of it.
Getting the front and rear
panels to mate neatly was a bit
challenging too. It felt like it
required three hands to snap
everything together. However, I did
get all the tabs lined up correctly in the end, and bolted
them down with the small securing screws.
The lead dress isn’t perfect
inside, but it’ll be easy to go back
in and clean it up after the fact. I
did a few listening tests with it
finally assembled. This amp doesn’t
put out a huge amount of power,
so it’s best with high efficiency
speakers. I used a set of Klipsch
bookshelf speakers with about 94
dB efficiency, and was able to drive
them louder than I would want to
listen to at my desk. So, there’s
plenty of power available.
As far as the sound, it was
accurate and neutral, almost clinical
even. The limiting factor of the
sound quality seemed to be the
source material. Since I’ll be using it
largely as a desktop amplifier, I
used a selection of lossless digital
audio files and a FiiO external DAC.
I could hear every note and pick
out every instrument with perfect
clarity. It did manage to sound
decent with a Pandora stream, but it wasn’t hard to pick
out some compression artifacts which the amp faithfully
reproduced if you listened for it. It’s going to sound as
good as the signal you feed it.
After the listening test, I took some measurements.
Based on my analysis, I found the amp’s power bandwidth
was within 1 dB from 20 kHz; 33 kHz and down was only
- 3 dB at 10 Hz. Distortion was very low throughout the
frequency range, too. It was a little higher at the ends and
lower in the middle, but all below 0.06% THD. That’s a
little bit higher than the 0.015% theoretical performance
of the chip; I expect I’m a bit limited by my test setup in
this case. Regardless, total harmonic distortion below
0.06% is very, very good.
It took me about six hours from start to finish to build
the kit, and other than some trouble prepping the case
and attaching some of the stranded wires, it was fun and
straightforward. If you’re a new builder, it might take a
The modular design and well thought-out instructions
make each step a rewarding experience. Since the kit is
modular, it’s easy to make some progress even if you only
have a few minutes to dedicate to projects in an evening.
I’d recommend this kit if you’re looking for a fun
project that delivers great results. I’m planning to use the
completed amplifier as my bench amp to use while
working on line-level audio devices like tuners and preamps since the low distortion will ensure I’m hearing the
device under test, and not the amp itself. It’d be just as
great as a desktop amp for daily listening, too. And, it
looks great! NV
Figure 22. Frequency response (0 dB = 12W into 8 ohms).
Figure 21. Test stack.
Figure 23. Total harmonic distortion (N = 5 at 12W into 8 ohms).