avhub.com.au 35
LAB REPORT
(0.00562%). The fourth, fi fth and sixth
harmonics are all more than 110dB down
(0.00031%) with a seventh harmonic at
–100dB (0.001%). There are some higher-or-
der harmonics, but all are more than 100dB
down. At 600-watts into 4Ω loads, distor-
tion was lower again, as you can see from
Graph 4. There’s a second harmonic at –82dB
(0.00794%), a third at –84dB (0.00630%),
then effectively only four higher-order com-
ponents, all of which are more than 110dB
down (0.00031%). Considering the power
level involved, and the low impedance, this
is an excellent result.
You can see from Graphs 1 through 4
that overall noise levels are not quite as low
across the audio hand as I’d expect from a
‘traditional’ amplifi er design, and there’s a
signifi cant increase in noise at low frequen-
cies. However, if you discount low-frequency
noise, noise was generally more than 120dB
down across the audio band, which is
excellent performance and about on par
with Bel Canto’s A-weighted, band-limited
specifi cation. As for wideband noise, Newport
Test Labs it as 87dB unweighted and 92dB
A-weighted when referenced to an output of
1-watt, and as 96dB unweighted and 100dB
A-weighted when referenced to rated output.
Intermodulation distortion was very
low, as you can see from Graph 5, which
shows CCIF distortion, using 19kHz and
20kHz test tones at –6dB. You can see there’s
almost no regenerated difference signal at
1kHz, which is excellent. (It is present, but
it’s 100dB down). There are some sidebands
either side of the test signals, the two high-
est, at 18kHz and 21kHz, are around –75dB
(0.01778%), the next pair more than 90dB
down (0.00316%), then the next more than
100dB down (0.001%). Again, excellent
performance.
Square wave performance was exemplary,
though assessing it is made diffi cult because
Newport Test Labs has not fi ltered out the Bel
Canto’s carrier frequency (though it’s almost
invisible on the 100Hz wave due to scaling)
from the oscillograms. You can see that
there’s no phase shift on any of the graphs,
plus good low-frequency and high-frequency
extension, and the underlying shape of the
1kHz square wave is excellent.
You can also see that when the ampli-
fi er is driven into a highly reactive load, it is
completely stable... though there is consider-
able ringing present. It would seem that Bel
Canto is only doing minimal fi ltering of the
carrier, reasoning that it’s high enough in
frequency that it would be totally inaudible,
and low enough in level that it would not
bother any tweeter.
Input sensitivity was satisfactorily low,
with 126mV required at the unbalanced
input to cause an output of 1-watt at the
speaker terminals, and a 2.3V input required
to deliver rated output. This means most
devices you could use with the REF600M will
have suffi cient voltage present at their out-
put to utilise the amplifi er’s full rated output.
As you’d expect of a Class-D design, it’s very
energy-effi cient, requiring only 28.9-watts to
deliver 1-watt of audio, and only 396-watts
to deliver rated output (into 8Ω).
In sum, I think this may be the best-per-
forming Class-D amplifier I’ve ever seen.
Steve HoldingBel Canto 600M Mono Power Amplifier – Lab Test Results
Test Measured Result Units/Comment
Frequency Response @ 1 watt o/p 11Hz – 22kHz –1dB
Frequency Response @ 1 watt o/p 6Hz – 66kHz –3dB
THD+N 0.005% / 0.014% @ 1-watt / @ rated output
Signal-to-Noise (unwghted/wghted) 87dB / 92dB dB referred to 1-watt output
Signal-to-Noise (unwghted/wghted) 96dB / 100dB dB referred to rated output
Input Sensitivity (CD Input) 126mV / 2.3V (1-watt / rated output)
Output Impedance 0.06Ω OC = V
Damping Factor 133 @1kHz
Power Consumption N/A / 8.37 watts (Standby / On)
Power Consumption 28.9 / 396 watts at 1-watt / at rated output
Mains Voltage Variation during Test 235–253 Minimum – MaximumBel Canto 600M Mono Power Amplifi er – Lab Test Results
Channel Load (Ω) 20Hz
(watts)20Hz
(dBW)1kHz
(watts)1kHz
(dBW)20kHz
(watts)20kHz
(dBW)
1 8 Ω 338 25.2 338 25.2 328 25.1
1 4 Ω 557 27.4 600 27.8 645 28.1
1 2 Ω 650* 28.1 650* 28.1 650* 28.1
Note: Figures in the dBW column represent output level in decibels referred to one watt output.
*Protection circuit cuts in at this output level.