Bel Canto e1X power amplifier Measurements

Sidebar 3: Measurements

I measured the Bel Canto e1X using my Audio Precision SYS2722 system (see the January 2008 "As We See It"). As the Bel Canto is a class-D design, it wasn't necessary to precondition the amplifier by using it to drive a 1kHz tone at one-third power into 8 ohms. Nevertheless, before doing any testing I ran it for an hour at a moderate power level, to ensure that it was fully warmed up. Because class-D amplifiers emit relatively high levels of ultrasonic noise that would drive my analyzer's input into slew-rate limiting, all measurements were taken with Audio Precision's auxiliary AUX-0025 passive low-pass filter, which eliminates noise above 200kHz. Without the filter, there was 398mV of ultrasonic noise present at the loudspeaker terminals.

The voltage gain at 1kHz into 8 ohms was the specified 27dB from both the balanced and single-ended inputs. Internal switches should increase the gain by 6dB; I measured 5.1dB. (Removing the top cover to reset the gain revealed the two Hypex Ncore class-D output modules and the switch-mode power supply.) The amplifier preserved absolute polarity (ie, was noninverting) from both sets of inputs. The single-ended input impedance measured the specified 10k ohms at low and middle frequencies, dropping very slightly to 9.2k ohms at the top of the audioband. The balanced input impedance was exactly twice these values, again as specified.

The Bel Canto's output impedance, including the series impedance of 6' of loudspeaker cable, was a very low 0.06 ohm at 20Hz and 1kHz, increasing very slightly to 0.07 ohm at 20kHz. The modulation of the amplifier's frequency response, due to the Ohm's law interaction between this source impedance and the impedance of our standard simulated loudspeaker, was therefore minuscule, at ±0.1dB (fig.1, gray trace). The response into resistive loads was flat to 10kHz into 8 ohms, then down by 0.6dB at 20kHz and 3dB at 38kHz (fig.1, blue and red traces). The restricted ultrasonic response is responsible for slightly lengthened risetimes with the amplifier's reproduction of a 10kHz squarewave (fig.2). There is also a faint hint of overshoot with the Bel Canto's squarewave response, this presumably due to the class-D output stage's integral low-pass filter.


Fig.1 Bel Canto e1x, frequency response at 2.83V into: simulated loudspeaker load (gray), 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (green) (1dB/vertical div.).


Fig.2 Bel Canto e1x, small-signal, 10kHz squarewave into 8 ohms.

Channel separation was superb at 105dB in both directions below 1kHz and still 82dB at 20kHz. With the Audio Precision ultrasonic filter, the gain set to 27dB, and the analog inputs shorted to ground, the wideband, unweighted signal/noise ratio (ref. 2.83V into 8 ohms) measured 73.7dB in the left channel and 71.4dB in the right. Restricting the measurement bandwidth to 22kHz increased the ratio to 90.8dB, left, and 85.5dB, right, and an A-weighting filter increased it further, to an excellent 100dB in both channels. Spectral analysis of the Bel Canto's low-frequency noise floor (fig.3) revealed spuriae related to the AC power-line frequency to be very low in level.


Fig.3 Bel Canto e1x, spectrum of 1kHz sinewave, DC–1kHz, at 1W into 8 ohms (left channel blue, right red; linear frequency scale).

The e1X is specified as delivering a maximum output power, at 1% THD+noise, of 250Wpc into 8 ohms, and 500Wpc into 4 ohms,3 both equivalent to 24dBW (footnote 1). I found that the Bel Canto amplifier easily exceeded its specified power, clipping at 340Wpc into 8 ohms (fig.4, 25.3dBW) and at 530Wpc into 4 ohms (fig.5, 24.2dBW), both powers at 1% THD+noise.


Fig.4 Bel Canto e1x, distortion (%) vs 1kHz continuous output power into 8 ohms.


Fig.5 Bel Canto e1x, distortion (%) vs 1kHz continuous output power into 4 ohms.

Distortion levels in the bass and midrange were low (fig.6) and didn't increase by much into 2 ohms (gray trace). The THD was higher in the treble, especially into lower impedances (cyan, magenta, and gray traces). The shape of the THD+N spuriae waveform at a moderate power into 8 ohms (fig.7, bottom trace) suggests that the distortion signature is primarily third harmonic in nature, though spectral analysis reveals that the second harmonic is also present at a slightly lower level (fig.8). Higher-order harmonics all lie below –110dB (0.0003%). Into lower impedances (not shown), the second harmonic rises to the same level as the third, but at –90dB (0.003%), neither will have any effect on the amplifier's sound. Despite the rise in THD in the top audio octaves, intermodulation distortion with an equal mix of 19kHz and 20kHz tones was very low, even at high powers into 4 ohms (fig.9).


Fig.6 Bel Canto e1x, THD+N (%) vs frequency at 20V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), and 2 ohms (gray).


Fig.7 Bel Canto e1x, 1kHz waveform at 30W into 8 ohms, 0.0035% THD+N (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).


Fig.8 Bel Canto e1x, spectrum of 50Hz sinewave, DC–1kHz, at 50Wpc into 8 ohms (left channel blue, right red; linear frequency scale).


Fig.9 Bel Canto e1x, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 100Wpc peak into 4 ohms (left channel blue, right red; linear frequency scale).

Bel Canto's e1X performed well on the test bench.—John Atkinson

Footnote 1: Our dBW figures follow the convention established by Martin Colloms and the late Peter W. Mitchell in the 1980s: they are referred to 2.83V (1W into 8 ohms). Thus an amplifier that behaves as a perfect voltage source (the ideal) offers the same dBW rating into all loads, an easier paradigm to grasp than having to remember to add 3dB every time the load impedance halves.
Bel Canto Design
221 1st St. North, Suite 300
Minneapolis, MN 55401
(612) 317-4550

Bogolu Haranath's picture

Mr.Tom Gibbs has good size listening room suitable for reviewing large size loudspeakers :-) ........

Bogolu Haranath's picture

Mr.TG could review the new, more powerful, same price as the Bel Canto, PS Audio Stellar M1200 mono-blocks, $6,000/pair ...... They are hybrid tube/Class-D design amps :-) .......

Ortofan's picture

...Bryston 4B3. Slightly higher output power capability and a 20-year warranty, instead of only 5 years from Bel Canto.

Otherwise, the still more powerful Parasound Halo A 21+ costs only $3,000.

The yet again more powerful Rotel RC-1590 costs $3,500.

OTOH, if you need a fancy brand name to impress your fellow audiophiles, then the Mark Levinson No. 532H, with an output capability similar to that of the Bel Canto, is now available for $6K.

None of these amps requires an external filter to limit ultrasonic garbage at their output.

Bogolu Haranath's picture

Parasound JC-5 can put out more power and costs the same, $6,000 :-) .........

Bogolu Haranath's picture

However, if you are one of the 'green new deal' supporters, Bel Canto and PS Audio Stellar amps are good choices :-) ........

Kal Rubinson's picture

I will not argue any of your suggestions but I believe that more apt comparisons might be among amps that can be as easily lifted as the BC. :-)

Bogolu Haranath's picture

May be Ortofan's 'Total-T' levels are higher than many of us .......... Just kidding Ortofan :-) .......

Ortofan's picture

... been my first suggestion, except that Benchmark does not specify an output power rating for operation in bridged mono mode into a 4Ω load.

If spending $6K on a power amp, I'd try to scrape up another $1K and buy a McIntosh MC312. 385W into either 8Ω or 4Ω should be sufficient. If not, the power guard feature will prevent clipping. The dealer will deliver it and the amp need never again be moved until after I die. Plus, it has those blue lit meters.

Bogolu Haranath's picture

Recently reviewed Primare A35.2 Class-D amp is slightly less powerful, and costs $3,500 :-) .......

Bogolu Haranath's picture

Stereophile Class-A, PS Audio Stellar M-700 Class-D monos are more powerful and cost $3,000/pair :-) ........

John Atkinson's picture
Ortofan wrote:
None of these amps requires an external filter to limit ultrasonic garbage at their output.

Neither does the Bel Canto when used to drive a pair of loudspeakers. But as I have explained, I need to use such a filter when measuring class-D amplifiers to avoid driving my analyzer's input stage into slew-rate limiting and thus give inaccurate distortion readings.

John Atkinson
Technical Editor, Stereophile

jeffhenning's picture

You and your crazy fear of input stage slew-rate limiting. The next thing you are going to say is that your measurements are accurate and relate to the sound you can hear.

You nutty kid!

a.wayne's picture

Hello John ,

So whats causing the poor Squarewave responses ..?


John Atkinson's picture
a.wayne wrote:
So whats causing the poor Squarewave responses?

The Bel Canto's squarewave response isn't "poor." In the 10kHz squarewave (fig.2), the lengthened risetime is associated with the ultrasonic rolloff (-3dB at 38kHz) seen in fig.1. There is a very slight overshoot on the waveform's leading edges but this is not accompanied by any ringing.

John Atkinson
Technical Editor, Stereophile

stereophileuser2020's picture

In the year 2020, you should really start thinking about the impact of power efficiency on the planet. Class D with its 90% efficiency is the only way to go.

Also, class D is much, much lighter in weight.

  • Bel Canto e1x amp (this review): 13 lbs
  • Bryston 4B3: 42 lbs
  • Parasound A 21+: 71 lbs
  • Rotel RB-1590: 84 lbs
Ortofan's picture

... NAD C 298 amp with the Purifi Eigentakt amp modules.

dc_bruce's picture

It's probably worth noting that when Jim Whinney introduced his "Magneplanar" speakers in the early 1970s, he demoed them with a 60 watt Audio Research tube amp. I first heard the Magneplanar (a 3-panel speaker, IIRC) demonstrated in 1972. I believe the transistor amps of the time were rated at about the same power output and clipped pretty nastily, unlike the ARC tube amps. Supposedly, the Magneplanar was much easier load, not being particularly reactive like cone speakers. I think people's expectations about loudness was less than it is today, and the Magneplanar was very listenable, as it remains today.

jeffhenning's picture

Thinking of the white papers and reviews of nCore amps I've seen in the past, they usually have a lot less distortion in the treble than this amp. In fact, they usually have very little rise in distortion as the frequency goes north.

That makes me wonder if this is a "feature" of Bel Canto's input stage.

Regardless, for around $1,800 or so, you can get Nord to make you an amp with one of Bruno's latest Purifi amp modules with a choice of input op-amps.

I think that says all you need to know on this subject.

David Harper's picture

While it is particularly important (with maggies)that an amp be a good match for them (I have the LRS speakers) it has nothing to do with the price of the amp. Rather it is about impedance,power output, current capability, etc. Correctly designed high-quality power amplifiers, when not overdriven, do not vary in sound quality.I'm driving my maggies with a Schiit Vidar amp which cost $700 and the sound quality (again, when not overdriven) is equal to that of any other amp regardless of price.

Bogolu Haranath's picture

In a recent review of Primare Class-D amp, HR used the Primare and Bel Canto Class-D amps to drive his Magnepan .7 speakers and got good results :-) ........

Tom Gibbs's picture

A couple of things; yes, $6000 is a lot for an amplifier to drive a $650 pair of loudspeakers, but it makes a couple of things really very clear. 1) When Wendell Diller says the Magneplanar LRS is built to show its mettle with higher-quality electronics, he's not overstating his position. The LRS will handle high input power levels and reproduce impressive dynamics with the Bel Canto e1X without flinching. 2) Sometimes it's possible to get a significant portion of a manufacturer's higher-end excellence at a very reasonable price point. The LRS is an almost unbelievable example of this, and also, though at a significantly higher price point, so is the e1X; it gives you so much more than just a taste of Bel Canto's top-line design implementations at a significantly reduced price.

Bogolu Haranath's picture

May be Mr.TG could review the new Yamaha NS-5000 speakers? :-) ........

Ali's picture

Class D, it seems, getting momentum!

Ric Schultz's picture

Amps using the same Ncore module and even the slightly better Purifi modules are available from VTV Audio, Nord, Apollon, etc. For instance, from VTV you can get a stereo amp with two Purifi/or NC500 modules, one 1200 watt power supply, discrete input stage using Sparko labs input op amps and Sparko labs regulators for $1350! Mono blocks are $1800. Less than one third the price and you get mono blocks with 1200 watt power supply in each amp. Direct sales, 30 day trial period. You can also get mono block amps using the NC1200 (bigger, more powerful and better sounding than NC500 modules) modules for about $2600. And these much cheaper amps can be modded for even better sound.