Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Measurements were taken of both channels of the Parasound Halo A 23+ at 120V AC line voltage, both channels driven, at the balanced and unbalanced inputs. Except as noted, discussed below are the results for the unbalanced inputs and the left channel.

Power output (stereo mode) @ 1kHz

• Power output at 1% THD+N: 191.9W @ 8 ohms, 275.1W @ 4 ohms
• Power output at 10% THD+N: 229.3W @ 8 ohms, 337.6W @ 4 ohms

Power output (mono mode) @ 1kHz

• Power output at 1% THD+N: 526.1W @ 8 ohms
• Power output at 10% THD+N: 693.4W @ 8 ohms

Additional data

• Input/output polarity: noninverting
• AC-line current draw at idle: 60.0W, 0.73A, 0.68PF
• Gain: output voltage divided by input voltage, 8-ohm load
  • Stereo mode
    • Balanced input: 28.5X, 29.1dB
    • Unbalanced input: 29.2X, 29.3dB
  • Mono mode
    • Balanced input: 57.6X, 35.2dB
    • Unbalanced input: 56.4X, 34.7dB
• Input sensitivity for 1W output into 8 ohms
  • Stereo mode
    • Balanced input: 99.2mV
    • Unbalanced input: 96.9mV
  • Mono mode
    • Balanced input: 49.1mV
    • Unbalanced input: 50.1mV
• Output impedance @ 50Hz
  • Stereo mode: 0.005 ohm
  • Mono mode: 0.009 ohm
• Input impedance @ 1kHz
  • Stereo mode
    • Balanced inputs: 97.0k ohms
    • Unbalanced inputs: 50.0k ohms
  • Mono mode
    • Balanced input: 97.0k ohms
    • Unbalanced input: 50.0k ohms
• Output noise (stereo mode), 8-ohm load, balanced inputs terminated with 600 ohms, Lch/Rch
  • Wideband: 0.335mV/0.244mV, -78.5dBW/-82.0dBW
  • A weighted: 0.116mV/0.085mV, -87.7dBW/-90.4dBW
• Output noise (stereo mode), 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 0.322mV/0.228mV, -78.9dBW/-81.9dBW
  • A weighted: 0.116mV/0.082mV, -87.7dBW/-90.8dBW
• Output noise (mono mode), 8-ohm load, balanced inputs terminated with 600 ohms
  • Wideband: 0.454mV, -75.9dBW
  • A weighted: 0.189mV, -83.5dBW
• Output noise (mono mode), 8-ohm load, unbalanced inputs terminated with 1k ohms
  • Wideband: 0.438mV, -76.2dBW
  • A weighted: 0.179mV, -84.0dBW

Measurements summary

The Halo A 23+, an update of the Halo A 23, is the lowest-powered amplifier model in Parasound’s Halo line. Because the A 23+ can be switched between stereo and bridged mono, some measurements were taken in both modes. In the charts below, the suffix A indicates measurements taken in stereo mode, the suffix B measurements taken in bridged mono.

Charts 1A and 1B show the Halo A 23+’s frequency response with varying loads. Chart 1A shows both channels, as between them there is a small difference in gain. Because the A 23+’s output regulation is so tight, only the 8-ohm loading is shown; the other loadings look the same. Note the greater high-frequency rolloff in bridged mode, due to the series connected nature of the load in bridged-mono mode.

Chart 2A illustrates how the Halo A 23+’s total harmonic distortion plus noise (THD+N) vs. power varies for 1kHz and SMPTE intermodulation test signals and amplifier output load with loads of 8 and 4 ohms. Chart 2B shows only the results in bridged mode for a load of 8 ohms -- the Halo A 23+ is not specified for use with 4-ohm loads in bridged mode.

The Halo A 23+’s THD+N as a function of frequency at several different power levels is plotted in Charts 3A and 3B. The increase in high-frequency distortion with frequency is moderate, and distortion remains reasonably low through most of the power range in both stereo and bridged-mono modes.

Chart 4 shows that the Parasound’s damping factor vs. frequency is of a value and nature typical of many solid-state amplifiers: high up to 1-2kHz, then rolling off with frequency. Not unexpectedly, the damping factor was lower in bridged mono, due to the fact that in this mode each channel’s output impedance is doubled.

Chart 5, the spectrum of the residue of harmonic distortion and noise of a 10W, 1kHz test signal, reveals a lot of line-frequency harmonics of significant amplitude. The signal harmonics are dominated by the second harmonic, with higher harmonics of significantly lower amplitude.

Chart 1 - Frequency response of output voltage as a function of output loading

Chart 1A - stereo mode

Stereo mode
Red line = Lch 8-ohm load (open circuit and 4-ohm loading look the same)
Blue line = Rch 8-ohm load (open circuit and 4-ohm loading look the same)

Chart 1B - mono mode

Mono mode
Red line = 8-ohm load (open circuit and 4-ohm loading look the same)

Chart 2 - Distortion as a function of power output and output loading

Chart 2A - stereo mode

Stereo mode
(Line up at 70W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 4-ohm THD+N distortion
Third line = 8-ohm THD+N
Bottom line = 4-ohm SMPTE IM distortion

Chart 2B - mono mode

Mono mode
(Line up at 70W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

Chart 3A - stereo mode

Stereo mode
(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 160W

Chart 3B - mono mode

Mono mode
(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Green line = 200W
Yellow line = 400W
Cyan line = 600W

Chart 4 - Damping factor as a function of frequency

Stereo mode
Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

Stereo mode
1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Unless otherwise noted, all measurements were taken using the balanced input of the Bryston 4B³’s left channel, with 120V AC line voltage driving both channels at 29dB gain.

Power output (stereo mode) @ 1kHz

• Power output at 1% THD+N: 354.0W @ 8 ohms, 552.0W @ 4 ohms
• Power output at 10% THD+N: 424.0W @ 8 ohms, 626.0W @ 4 ohms

Power output (mono mode) @ 1kHz

• Power output at 1% THD+N: 1190W @ 8 ohms
• Power output at 10% THD+N: 1440W @ 8 ohms

Additional data

• Input/output polarity: noninverting
• AC-line current draw at idle: 81.0W, 1.09A, 0.62PF
• Gain: output voltage divided by input voltage, 8-ohm load
  • Stereo mode
    • Balanced input (29dB gain): 28.26X, 29.0dB
    • Balanced input (23dB gain): 14.13X, 23.0dB
    • Unbalanced input (29dB gain): 27.27X, 28.7dB
    • Unbalanced input (23dB gain): 14.12X, 23.0dB
  • Mono mode
    • Balanced input (29dB gain): 27.84X, 28.9dB
    • Balanced input (23dB gain): 13.91X, 22.9dB
    • Unbalanced input (29dB gain): 27.65X, 28.8dB
    • Unbalanced input (23dB gain): 13.79X, 22.8dB
• Input sensitivity for 1W output into 8 ohms
  • Stereo mode
    • Balanced input (29dB gain): 100.1mV
    • Balanced input (23dB gain): 203.3mV
    • Unbalanced input (29dB gain): 102.3mV
    • Unbalanced input (23dB gain): 200.3mV
  • Mono mode
    • Balanced input (29dB gain): 101.6mV
    • Balanced input (23dB gain): 203.3mV
    • Unbalanced input (29dB gain): 102.3mV
    • Unbalanced input (23dB gain): 205.1mV
• Output impedance @ 50Hz
  • Stereo mode: 0.008 ohm
  • Mono mode: 0.018 ohm
• Input impedance @ 1kHz
  • Stereo mode
    • Balanced input: 20k ohms
    • Unbalanced input: 29k ohms
  • Mono mode
    • Balanced input: 10.5k ohms
    • Unbalanced input: 7.8k ohms
• Output noise (stereo mode), 8-ohm load, balanced inputs terminated with 600 ohms, Lch/Rch
  
  • 29dB gain
    • Wideband: 413.9uV/406.6uV, -76.7dBW/-76.9dBW
    • A weighted: 98.69uV/96.98uV, -89.2dBW/-89.3dBW
  • 23dB gain
    • Wideband: 215.9uV/209.0uV, -82.4dBW/-82.6dBW
    • A weighted: 53.04uV/49.13uV, -95.0dBW/-95.2dBW
• Output noise (stereo mode), 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • 29dB gain
    • Wideband: 419.7uV/410.5uV, -76.6dBW/-76.8dBW
    • A weighted: 99.43uV/98.98uV, -89.1dBW/-89.3dBW
  • 23dB gain
    • Wideband: 219.8uV/213.0uV, -82.2dBW/-82.5dBW
    • A weighted: 54.99uV/51.09uV, -94.2dBW/-94.9dBW
• Output noise (mono mode), 8-ohm load, balanced inputs terminated with 600 ohms
  
  • 29dB gain
    • Wideband: 413.9uV, -76.7dBW
    • A weighted: 98.69uV, -89.2dBW
  • 23dB gain
    • Wideband: 215.9uV, -82.4dBW
    • A weighted: 53.04uV, -94.5dBW
• Output noise (mono mode), 8-ohm load, unbalanced inputs terminated with 1k ohms
  • 29dB gain
    • Wideband: 419.7uV, -76.6dBW
    • A weighted: 99.43uV, -89.1dBW
  • 23dB gain
    • Wideband: 219.8uV, -82.2dBW
    • A weighted: 54.99uV, -94.2dBW

Measurements summary

The Bryston 4B³ is a high-power stereo power amplifier capable of operating in bridged-mono mode; most measurements were also made in the mono mode.

Chart 1 shows the 4B³’s frequency response with varying loads. The output impedance over most of the audioband is low enough that the IHF load would not show any appreciable difference within that bandwidth. In mono mode (not shown), the amount of change with load at 200kHz was about double.

Chart 2A illustrates how the 4B³’s total harmonic distortion plus noise (THD+N) vs. power varies for 1kHz and SMPTE IM test signals into 8 and 4 ohms. Chart 2B shows the results for the mono mode into 8 ohms. The levels of distortion are very low.

THD+N as a function of frequency at several different power levels is plotted in Chart 3. The degree of increase in distortion at high frequencies is admirably low.

Chart 4 shows the 4B³’s damping factor vs. frequency: though high at low frequencies, it starts to decline at a lower frequency than in most solid-state designs. The damping factor for the mono mode (not shown) is similar in shape but about half the magnitude.

The spectrum of the harmonic distortion and noise residue of a 10W, 1kHz test signal is plotted in Chart 5. The magnitude of the AC line harmonics is reasonably low, and dominated by odd harmonics of 60Hz. The test-signal harmonics are very low, consisting mostly of the second and third harmonics.

Chart 1 - Frequency response of output voltage as a function of output loading

Stereo mode
Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 2 - Distortion as a function of power output and output loading

Chart 2A

Stereo mode
(Line up at 100W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 4-ohm THD+N
Third line = 8-ohm SMPTE IM distortion
Bottom line = 8-ohm THD+N

Chart 2B

Mono mode
(Line up at 100W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

Stereo mode
(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 100W
Green line = 300W

Chart 4 - Damping factor as a function of frequency

Stereo mode
Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

Stereo mode
1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Unless otherwise noted, measurements were made at 120V AC line voltage, using the balanced inputs, and with the left channel.

Power output (stereo mode) @ 1kHz

• Power output at 1% THD+N: 414.8W @ 8 ohms, 665.9W @ 4 ohms
• Power output at 10% THD+N: 510.0W @ 8 ohms, 800.0W @ 4 ohms

Power output (mono mode) @ 1kHz

• Power output at 1% THD+N: 1328W @ 8 ohms
• Power output at 10% THD+N: 1450W @ 8 ohms

Additional data

• Input/output polarity: noninverting
• AC-line current draw
  • Standby: 2.1W, 0.04A, 0.44PF
  • Operate: 267.0W, 3.1A, 0.72PF
• Gain: output voltage divided by input voltage, 8-ohm load
  • Stereo mode, Lch/Rch
    • Balanced inputs: 28.04X, 29.0dB / 27.06X, 28.7dB
    • Unbalanced inputs: 28.35X, 29.1dB / 27.36X, 28.7dB
  • Mono mode
    • Balanced input: 59.94X, 34.8dB
    • Unbalanced input: 54.30X, 34.7dB
• Input sensitivity for 1W output into 8 ohms
  • Stereo mode, Lch/Rch
    • Balanced inputs: 100.7mV / 104.5mV
    • Unbalanced inputs: 99.8mV / 103.4mV
  • Mono mode
    • Balanced input: 51.5mV
    • Unbalanced input: 52.1mV
• Output impedance @ 50Hz
  • Stereo mode: 0.0033 ohm
  • Mono mode: 0.0065 ohm
• Input impedance @ 1kHz
  • Stereo mode, Lch/Rch
    • Balanced inputs: 210k ohms / 219k ohms
    • Unbalanced inputs: 105k ohms / 109k ohms
  • Mono mode
    • Balanced input: 152k ohms
    • Unbalanced input: 78k ohms
• Output noise (stereo mode), 8-ohm load, balanced inputs terminated with 600 ohms, Lch/Rch
  • Wideband: 337.2uV/323.5uV, -78.5dBW/-78.8dBW
  • A weighted: 133.1uV/97.9uV, -86.6dBW/-89.2dBW
• Output noise (stereo mode), 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 333.3uV/325.4uV, -78.6dBW/-78.9dBW
  • A weighted: 130.1uV/97.9uV, -86.7dBW/-89.2dBW
• Output noise (mono mode), 8-ohm load, balanced inputs terminated with 600 ohms
  • Wideband: 501.0uV, -73.0dBW
  • A weighted: 188.5uV, -83.5dBW
• Output noise (mono mode), 8-ohm load, unbalanced inputs terminated with 1k ohms
  • Wideband: 498.3uV, -75.1dBW
  • A weighted: 190.7uV, -83.4dBW

Measurements summary

Parasound’s Halo JC 5 is one of the latest results of John Curl’s long-established expertise in power-amplifier design. Weighing a hefty 90 pounds, it’s a beautiful unit, impressive indeed.

The frequency responses for the Halo JC 5’s stereo and mono modes are plotted in Charts 1A and 1B. Notable are the smooth high-frequency rolloff and the good regulation with load changes. The greater rolloff in mono mode is essentially caused by the outputs of the two stereo channels in series into the loads.

The JC 5’s total harmonic distortion (THD) plus noise and SMPTE intermodulation distortion are shown in Charts 2A and 2B for, respectively, the stereo and mono modes. (In mono mode, the JC 5 is specified for 8-ohm loads.)

Charts 3A and 3B plot the Halo JC 5’s THD+N vs. frequency and power. The rise in high-frequency distortion is typical of most of the power amplifiers I test.

The Halo JC 5’s damping factor vs. frequency in stereo mode, plotted in Chart 4, is typical of solid-state amplifiers: high at low frequencies, and falling above a few hundred hertz. Not surprisingly, the result in mono mode (not shown) looked similar, but with about half the damping factor. Again, this is because, in mono mode, the output is the series output of the two stereo channels.

Chart 5, the spectrum of a 10W, 1kHz signal into 8 ohms, shows low amounts of signal harmonics, mainly the second and third harmonics -- a good thing for sound. Also visible, however, are large amounts of line-based hum harmonics; these are of high enough magnitude to contribute to the total harmonic measurement amount.

Chart 1 - Frequency response of output voltage as a function of output loading

Chart 1A - stereo mode

Stereo mode
Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 1B - mono mode

Mono mode
Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 2 - Distortion as a function of power output and output loading

Chart 2A - stereo mode

Stereo mode
(Line up at 100W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 4-ohm THD+N
Bottom line = 8-ohm THD+N

Chart 2B - mono mode

Mono mode
(Line up at 100W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

Chart 3A - stereo mode

Stereo mode
(4-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 75W
Cyan line = 200W
Green line = 400W
Yellow line = 600W

Chart 3B - mono mode

Mono mode
(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 100W
Cyan line = 600W
Green line = 1200W

Chart 4 - Damping factor as a function of frequency

Stereo mode
Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

Stereo mode
1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Unless otherwise noted, measurements were taken at the Devialet Expert 130 Pro’s left-channel Coax 1 digital input, at a sampling rate of 96kHz and with 120V AC line voltage, both channels driven.

Power output

• Power output at onset of limiter: approximately 100W @ 8 ohms, 200W @ 4 ohms

Additional data

• Input/output polarity (digital and analog inputs): noninverting
• AC-line current draw
  • Standby: 6.0W, 0.13A, 0.40PF
  • Operating: 33.0W, 0.35A, 0.82PF
• Gain: output voltage divided by input voltage, 8-ohm load (Lch/Rch)
  • Analog unbalanced inputs (volume full up at +30dB): 284.6X, 48.6dB / 262.3X, 48.3dB
  • Digital input (-20dBFS input with volume at 0dB): 2.483V / 2.480V
• Input sensitivity for 1W output into 8 ohms (Lch/Rch)
  • Analog unbalanced inputs: 9.9mV / 10.8mV
• Output impedance @ 50Hz: 0.01 ohm
• Input impedance @ 1kHz
  • Analog unbalanced inputs: 16.5k ohms
• Output noise, volume at 30dB (Lch/Rch)
  
  • Wideband: 6.44mV/5.82mV, -52.8dBW/-53.7dBW
  • A weighted: 0.063mV/0.060mV, -93.0dBW/-93.5dBW
• Output noise, volume at 0dB (Lch/Rch)
  
  • Wideband: 6.28mV/5.55mV, -53.1dBW/-54.1dBW
  • A weighted: 0.040mV/0.032mV, -97.0dBW/-98.9dBW
• Output noise, volume at -20dB (Lch/Rch)
  • Wideband: 5.76mV/5.47mV, -53.8dBW/-54.3dBW
  • A weighted: 0.071mV/0.063mV, -92.0dBW/-93.0dBW
• Output noise, volume at -30dB (Lch/Rch)
  
  • Wideband: 5.81mV/5.49mV, -53.8dBW/-54.2dBW
  • A weighted: 0.075mV/0.063mV, -91.5dBW/-93.0dBW

Measurements summary

The Expert 130 Pro is an unusual and clever combination of class-A and class-D amplification: The class-A section swings the output voltage, and the class-D section supplies most of the output current. Unique among designs using class-D circuitry, there is no output low-pass filter. As a result, the Devialet’s high-frequency response is quite independent of load. The Expert 130 has a limiter circuit that prevents it from clipping, which meant I couldn’t run my usual test of clipping behavior along with the power output at 1% and 10% THD+N. The Devialet’s digital switching frequency noise was low enough that I didn’t need to use Audio Precision’s AUX-0025 external low-pass filter in my testing, as I usually do.

I tested the Devialet’s volume-control tracking using a 1kHz test tone with the reference volume being that of the 5W output with a 500mV signal input. Volume-control tracking was within less than 0.1dB with volume settings of +30dB to -30dB.

Chart 1A shows the frequency response of the Expert 130 Pro with varying loads and with my usual vertical scale: the curves of the open circuit and 8- and 4-ohm loads are direct overlays; that is, they’re identical. The analog input frequency response with a sampling frequency of 96kHz looks about the same. Chart 1B plots the Expert 130 Pro’s frequency responses for sample rates of 44.1, 96, and 192kHz. For these kinds of response curves driven by the Audio Precision digital generator, it’s not possible to follow the curves very far into the high-frequency cutoff region, due to those frequencies approaching one-half the sampling frequency.

Chart 2 illustrates how the Expert 130 Pro’s total harmonic distortion plus noise (THD+N) vs. power varies for 1kHz and SMPTE intermodulation test signals and amplifier output for loads of 8 and 4 ohms. The amount of distortion is quite low, and is dominated by noise rather than by distortion per se over much of the power range. This test yielded very similar results when taken at the Devialet’s analog input.

The Expert 130 Pro’s THD+N as a function of frequency at several different power levels is plotted in Chart 3. This amplifier’s levels of distortion are so low that measuring its THD+N with an 80kHz filter bandwidth, as I usually do, didn’t actually reveal the distortion, which remains largely below the level of the noise. To compensate for this, in Chart 3 I reduced the measurement bandwidth to 22kHz, at the cost of showing an increase in distortion in the last two octaves of the audio range. Nonetheless, the readings are still largely dominated by noise, and show the distortion’s tendency to rise at higher frequencies.

The Devialet’s damping factor vs. frequency is plotted in Chart 4. These data were taken as a function of the Devialet’s volume-control setting, and surprisingly showed this variation in the high-frequency shape. Note that is with the digital input signal set to Off, and the output of the channel measured driven differentially from a separate power amplifier. A regulated 1A of current as a function of frequency is applied to each phase of the tested amplifier’s output as a function of frequency.

Chart 5 plots the spectrum of the Expert 130 Pro’s harmonic distortion and noise residue of a 10W, 1kHz test signal. The line harmonics are visible but low in magnitude. The signal harmonics are mainly the second, with lower-level, higher-order harmonics above that. These data show just how low the Devialet’s distortion is, and the values in Chart 3 approach this.

Chart 1A - Frequency response of output voltage as a function of output loading

Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 1B - Frequency response of output voltage as a function of sample rate

Red line = 44.1kHz
Magenta line = 96kHZ
Blue line = 192kHz

Chart 2 - Distortion as a function of power output and output loading

(Line up at 20W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 4-ohm THD+N
Bottom line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 60W
Green line = 80W

Chart 4 - Damping factor vs. frequency and volume control

Damping factor = output impedance divided into 8

Red line: +10dB
Magenta line: 0dB
Blue line: -10dB
Cyan line: -20dB

Chart 5 - Distortion and noise spectrum

1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Measurements of the NAD M32’s left channel were taken at its AES digital and Line 1 analog inputs, at a line voltage of 120V AC, both channels driven. The M32 is a Direct Digital switching amplifier, a technology developed by NAD. As usual, I had to use (except as noted) Audio Precision’s AUS-0025 external low-pass filter, to keep the out-of-band noise of the tested amplifier from corrupting the measurements taken with AP’s SYS-2722 measuring system.

Power output

• Estimated power output at 1% THD+N: 193.0W @ 8 ohms, 190.0W @ 4 ohms
• Estimated power output at 10% THD+N: 200.0W @ 8 ohms, 193.0W @ 4 ohms

Additional data

• Input/output polarity (digital and analog inputs): noninverting
• AC-line current draw: 35.0W, 0.456A, 0.63PF
• Gain: output voltage divided by input voltage, 8-ohm load (Lch/Rch)
  • Analog unbalanced inputs (volume full up at +10dB): 64.85X, 36.2dB / 64.06X, 36.1dB
  • Digital input (-20dBFS input with volume at 0dB): 4.103V / 4.057V
• Input sensitivity for 1W output into 8 ohms (Lch/Rch)
  • Analog unbalanced inputs: 43.6mV / 44.1mV
• Output impedance @ 50Hz: 0.1 ohm
• Input impedance @ 1kHz
  • Line 1 input: 9.92k ohms
• Output noise, Line 1 analog input, volume at 10dB (Lch/Rch)
  
  • Wideband: 40.8mV/39.3mV, -36.9dBW/-37.1dBW
  • A weighted: 0.36mV/0.36mV, -77.9dBW/-77.9dBW
• Output noise, Line 1 analog input, volume at -4dB (Lch/Rch)
  
  • Wideband: 42.4mV/40.3mV, -36.5dBW/-36.9dBW
  • A weighted: 0.113mV/0.113mV, -88.0dBW/-88.0dBW
• Output noise, Line 1 analog input, volume at -20dB (Lch/Rch)
  • Wideband: 40.3mV/39.5mV, -36.9dBW/-37.1dBW
  • A weighted: 0.086mV/0.086mV, -90.3dBW/-90.3dBW
• Output noise, AES/EBU digital input, volume at 10dB (Lch/Rch)
  
  • Wideband: 40.5mV/39.6mV, -36.9dBW/-37.1dBW
  • A weighted: 0.082mV/0.082mV, -90.8dBW/-90.8dBW
• Output noise, AES/EBU digital input, volume at -4dB (Lch/Rch)
  
  • Wideband: 40.7mV/39.6mV, -36.9dBW/-37.1dBW
  • A weighted: 0.086mV/0.082mV, -90.3dBW/-90.8dBW
• Output noise, AES/EBU digital input, volume at -20dB (Lch/Rch)
  
  • Wideband: 40.5mV/38.5mV, -36.9dBW/-37.3dBW
  • A weighted: 0.086mV/0.082mV, -90.3dBW/-90.8dBW

Measurements summary

Volume-control tracking, tested using a 1kHz test tone, was within less than 0.1dB in the range of volume settings of +10dB (max) to -40dB.

Chart 1A shows the frequency response of the M32 with varying loads and with my usual vertical scale. This was done by feeding the Line 1 analog inputs an internal sampling rate of 96kHz. As can be seen, the curves of the open circuit and the 8-ohm, 4-ohm, and NHT dummy-speaker loads deviate considerably from flatness in the high frequencies. NAD has a speaker-impedance setting that’s supposed to correct for the effects of variable loading on the output LCR low-pass filter in the amp. I tested it in both settings, and there was a subtle difference, but it really did nothing to improve the out-of-band response with loading. With a digital input, I plotted the output response at sample rates of 44.1, 96, and 192kHz. Again, an out-of-band rise in response appears, and with response beyond that of Chart 1B. This rise in out-of-band response could have an effect on the sound of high-frequency music information, as opposed to a flat response.

The frequency-response curve in Chart 1C is that of the RIAA equalization error of the analog phono input. This is quite good for this measurement.

Chart 2 illustrates how the M32’s total harmonic distortion plus noise (THD+N) vs. power varies for 1kHz and SMPTE intermodulation test signals and amplifier output load of 8 and 4 ohms. The amount of distortion is reasonably low in this design, and through much of the power range is dominated by noise rather than distortion per se. This test, measured at the analog input, yielded very similar results.

The M32’s THD+N as a function of frequency, at several different power levels, is plotted in Chart 3. Because this amplifier generates quite a bit of out-of-band noise, using my usual 80kHz low-pass filter to make it possible for me to measure the harmonics of 20kHz would have caused the measurements to be far too dominated by noise. So for this chart I reduced the measurement bandwidth to 22kHz, though this shows an increase in distortion in the last two octaves of the audioband. Nonetheless, the readings are still largely dominated by noise, and show the distortion’s tendency to rise at higher frequencies.

The M32’s damping factor vs. frequency is plotted in Chart 4. As is typical of many solid-state power amplifiers, the damping factor begins to fall off above about 1kHz.

Chart 5 plots the spectrum of the M32’s harmonic distortion and noise residue when fed a 10W, 1kHz test signal. The line harmonics are visible but low in magnitude. The signal harmonics are numerous and complex, with even and odd harmonics out to 20kHz.

A comment on the noise measurements above with both analog and digital data: The analog noise is higher at the higher volume settings due to the noise of the M32’s analog electronics feeding the A/D converter. In the digital data, the noise is independent of the volume setting.

Chart 1A - Frequency response of output voltage as a function of output loading

Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load
Cyan line = NHT dummy-speaker load

Chart 1B - Frequency response of output voltage as a function of sample rate

Red line = 44.1kHz
Magenta line = 96kHZ
Blue line = 192kHz

Chart 1C - Frequency response of output voltage as a function of phono RIAA equalization error

Red line = left channel
Magenta line = right channel

Chart 2 - Distortion as a function of power output and output loading

(Line up at 100W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 4-ohm THD+N
Bottom line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 70W
Yellow line = 180W

Chart 4 - Damping factor vs. frequency and volume control

Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Unless otherwise noted, these measurements were taken at the left-channel unbalanced analog inputs of the Exogal Ion PowerDAC at 120V AC line voltage, both channels driven. I used the old IHF standard for integrated amplifiers, in which the volume control is set so that a 500mV analog input signal produces a nominal output of 5W into 8 ohms. Unless stated otherwise, I used Audio Precision’s Aux-0025 external filter.

Power output

• Power output at 1% THD+N: 80.0W @ 8 ohms, >100.0W @ 4 ohms
• Power output at 10% THD+N: 100.0W @ 8 ohms, >100.0W @ 4 ohms

Additional data

• Input/output polarity
  • Analog input: inverting
  • Digital input: noninverting
• AC-line current draw
  • Standby: 9.0W, 0.17A, 0.44PF
  • Operating: 25.0W, 0.31A, 0.68PF
• Gain: output voltage divided by input voltage, 8-ohm load (Lch/Rch)
  • Unbalanced inputs: 36.70X, 37.7dB / 36.18X, 37.7dB
• Input sensitivity for 1W output into 8 ohms (Lch/Rch)
  • Unbalanced inputs: 77.1mV / 78.3mV
• Output impedance @ 50Hz: 0.065 ohm
• Input impedance @ 1kHz
  • Unbalanced inputs: 1.87k ohms
• Output noise at reference conditions, 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 0.352mV/0.350mV, -78.1dBW/-78.2dBW
  • A weighted: 0.142mV/0.142mV, -86.0dBW/-86.0dBW
• Output noise with volume at maxium, 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 127.2mV/128.3V, -26.9dBW/-26.8dBW
  • A weighted: 1.95mV/1.89mV, -63.2dBW/-63.5dBW
• Output noise with volume at minimum, 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 0.352V/0.350V, -78.1dBW/-78.2dBW
  • A weighted: 0.142mV/0.142mV, -86.0dBW/-86.0dBW

Measurements summary

Exogal’s Ion PowerDAC is an unusual combination of a DAC and line stage (the Comet) coupled to a DAC and switching power amp (the Ion). I used the unbalanced analog inputs of the Comet to test the combination as an integrated amplifier.

It was difficult to measure some things on the Ion PowerDAC, including its maximum power and damping factor. The Exogal’s very sensitive protection circuit frequently shut it down, requiring a restart, and the damping factor was affected by some interference from the untested right channel.

Chart 1A shows the frequency response of the Ion PowerDAC with varying loads and with my usual vertical scale. The FR is strongly dependent on the load, as is typical of switching-amp designs. With resistive loading, the -3dB bandwidth for an 8-ohm load extends slightly higher than 20kHz; with a 4-ohm load, the bandwidth is 11-12kHz. Chart 1B shows the Exogal’s high-frequency response out to 30kHz: It rolls off pretty quickly and, with the open circuit loading, peaks at over +10dB. This will probably not be a problem with non-inductive tweeters, but with a typical moving-coil tweeter, the Ion PowerDAC may produce a high-frequency rise in the top octave of the audioband: 10-20kHz.

To test the tracking of the Exogal’s volume control, I used a 1kHz test tone; the reference volume was the 5W output with a 500mV input signal. The volume-control tracking was within a few tenths of a dB down to -60dB.

Chart 2 illustrates how the Ion PowerDAC’s total harmonic distortion plus noise (THD+N) vs. power varied for 1kHz and SMPTE intermodulation (IM) test signals and amplifier output for 8- and 4-ohm loads. As mentioned above, it was not possible to get full distortion data because of the behavior of the Exogal’s protection circuit. The THD+N of the 1kHz test signal was pretty good, but the IM distortion was unusually high throughout the range I was able to measure.

The Ion PowerDAC’s THD+N as a function of frequency at different power levels is plotted in Chart 3. To reduce the out-of-band noise, I had to use the regular Audio Precision low-pass filter set to 30kHz instead of the usual 80kHz, and a 40kHz sharp cutoff filter to keep the distortion measurement uncontaminated by noise throughout most of the audioband. Any rise in high-frequency distortion is therefore masked above about 10kHz. At the 70W level, the amount of distortion rises precipitously below 100Hz and above about 15kHz.

It wasn’t possible to take my usual measurements of damping factor vs. frequency. What can be deduced from Chart 1A is that the damping factor is reasonably high at low frequencies, at around 123, and decreases rapidly above about 1kHz -- as can be seen in the divergence of the curves above that frequency.

Chart 5 plots the spectrum of the harmonic distortion and noise residue of a 10W, 1kHz test signal sent through the Exogal. The magnitudes of the AC line harmonics are not visible in what is a relatively high level of background noise throughout the audioband. The signal harmonics are mainly the third, fifth, and seventh, with lower-level, higher-order harmonics above 10kHz.

Chart 1 - Frequency response of output voltage as a function of output loading

1A

Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load
Cyan line = NHT dummy-speaker load

1B

Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load
Cyan line = NHT dummy-speaker load

Chart 2 - Distortion as a function of power output and output loading

(Line up at 20W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 4-ohm THD+N
Bottom line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 70W

Chart 5 - Distortion and noise spectrum

1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Unless otherwise noted, measurements were taken at a line voltage of 120V AC at the left-channel balanced input, with the Audio Precision AUX-0025 filter engaged.

Power output

• Power output at 1% THD+N: 164.7W @ 8 ohms, 302.4W @ 4 ohms
• Power output at 10% THD+N: 230.0W @ 8 ohms, 395.0W @ 4 ohms

Additional data

• Input/output polarity: noninverting
• AC-line current draw at idle: 14.0W, 0.23A, 0.51PF
• Gain: output voltage divided by input voltage, 8-ohm load (Lch/Rch)
  • Balanced inputs: 15.36X, 23.7dB / 15.28X, 23.7dB
  • Unbalanced inputs: 30.81X, 29.8dB / 30.48X, 29.7dB
• Input sensitivity for 1W output into 8 ohms (Lch/Rch)
  • Balanced inputs: 184mV / 185mV
  • Unbalanced inputs: 91.8mV / 92.8mV
• Output impedance @ 50Hz: 0.018 ohm
• Input impedance @ 1kHz
  • Balanced inputs: 21.4k ohms
  • Unbalanced inputs: 10.3k ohms
• Output noise, 8-ohm load, balanced inputs terminated with 600 ohms, Lch/Rch
  
  • Wideband: 0.849mV/0.857mV, -70.5dBW/-70.4dBW
  • A weighted: 0.050mV/0.050mV, -95.0dBW/-95.0dBW
• Output noise, 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  • Wideband: 0.882mV/0.890mV, -70.1dBW/-70.0dBW
  • A weighted: 0.065mV/0.066mV, -92.8dBW/-92.6dBW
• Output noise, 8-ohm load, without AP AUX-0025 filter, unbalanced and balanced inputs
  
  • Wideband: 498mV, -15.1dBW

Measurements summary

The Wadia a315 is a medium-power, two-channel, switching power amplifier that includes the latest Hypex Electronics Ncore technology.

Chart 1, which shows the a315’s frequency response into varying loads, was not sent through the Audio Precision AUX-0025 low-pass measuring filter. One of the interesting aspects of the Ncore technology is that the out-of-band high-frequency response is controlled by an output LCR filter in the overall feedback loop. Of interest in the shapes of the curves is that, just above 20kHz, the output impedance goes from normal positive (output drops with increasing load) to negative, with the greatest deviation of response at 40kHz, then passes through a close-to-zero output-impedance point at 60kHz, above which the output impedance becomes positive again. Within the audioband, the output impedance is low enough that there is negligible variation with the NHT dummy speaker load.

Chart 2 illustrates how the a315’s total harmonic distortion plus noise (THD+N) vs. power varies with 1kHz and SMPTE intermodulation test signals, and with 8- and 4-ohm loads. The level of distortion indicated is reasonably low.

THD+N as a function of frequency at different power levels is plotted in Chart 3. The increase in distortion with frequency is moderate. Still, the level of distortion is low over most of the power and frequency range.

Damping factor vs. frequency is shown in Chart 4. The damping factor remains moderately high throughout the audioband, and, unlike with most power amps, does not roll off severely above 500-1000Hz.

Chart 5 plots the spectrum of the harmonic distortion and noise residue of a 10W, 1kHz test signal. The AC line harmonics are low, their principal component being 60Hz. The signal harmonics are dominated by the second and third harmonics, with higher harmonics of rapidly decreasing magnitude.

Chart 1 - Frequency response of output voltage as a function of output loading

Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 2 - Distortion as a function of power output and output loading

(Line up at 50W to determine lines)
Top line = 4-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 4-ohm THD+N
Bottom line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

(8-ohm loading)
Red line = 1W
Magenta line = 10W
Blue line = 30W
Cyan line = 70W
Green line = 130W
Yellow line = 150W

Chart 4 - Damping factor as a function of frequency

Stereo mode
Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

1kHz signal at 10W into an 8-ohm load

Details

Parent Category: Products

Category: Amplifier Measurements

All amplifier measurements are performed independently by BHK Labs. All measurement data and graphical information displayed below are the property of the SoundStage! Network and Schneider Publishing Inc. Reproduction in any format is not permitted.

Note: Unless otherwise noted, measurements were taken on the left-channel balanced input at 120V AC line voltage.

Power output (stereo mode)

• Power output at 1% THD+N: 130.0W @ 8 ohms, 195.9W @ 4 ohms
• Power output at 10% THD+N: 157.1W @ 8 ohms, 241.0W @ 4 ohms

Power output (mono mode)

• Power output at 1% THD+N: 405.0W @ 8 ohms
• Power output at 10% THD+N: 480.0W @ 8 ohms

Additional data

• Input/output polarity: noninverting
• AC-line current draw at idle: 31.0W, 0.53A, 0.48PF
• Gain: output voltage divided by input voltage, 8-ohm load (balanced/unbalanced inputs)
  •      Stereo mode: 32.2X, 30.2dB
  •      Mono mode: 64.5X, 36.2dB
• Input sensitivity for 1W output into 8 ohms (balanced/unbalanced inputs)
  •      Stereo mode: 87.7mV
  •      Mono mode: 43.8mV
• Output impedance @ 50Hz
  •      Stereo mode: 0.003 ohm
  •      Mono mode: 0.0035 ohm
• Input impedance @ 1kHz (stereo/mono modes)
  •      Balanced input: 90.0k ohms
  •      Unbalanced input: 45.0k ohms
• Output noise (stereo mode), 8-ohm load, balanced inputs terminated with 600 ohms, Lch/Rch
  
  •      Wideband: 0.15mV/0.15mV, -85.6dBW/-85.8dBW
  •      A weighted: 0.069mV/0.060mV, -92.3dBW/-93.5dBW
• Output noise (stereo mode), 8-ohm load, unbalanced inputs terminated with 1k ohms, Lch/Rch
  
  •      Wideband: 0.18mV/0.15mV, -83.9dBW/-85.5dBW
  •      A weighted: 0.12mV/0.062mV, -87.4dBW/-93.1dBW
• Output noise (mono mode), 8-ohm load, balanced inputs terminated with 600 ohms
  
  •      Wideband: 0.20mV, -82.9dBW
  •      A weighted: 0.083mV, -90.7dBW
• Output noise (mono mode), 8-ohm load, unbalanced inputs terminated with 1k ohms
  •      Wideband: 0.23mV, -81.9dBW
  •      A weighted: 0.13mV, -84.0dBW

Measurements summary

The Simaudio Moon Neo 330A is a medium-power stereo amplifier with a conventional linear power supply and bipolar output devices. It can be bridged to function as a mono power amp by using Simaudio’s special balanced bridge cable, which reverses the signal phase of the left channel to make the outputs of the two stereo channels out of phase. As usual in mono mode, the load is connected between the two plus outputs of the stereo channel outputs.

Chart 1 shows the frequency response of the Moon Neo 330A with varying loads. The output impedance is low enough that there was negligible variation with the NHT dummy speaker load.

Chart 2A illustrates how total harmonic distortion plus noise (THD+N) vs. power varies for 1kHz and SMPTE intermodulation (IM) test signals and amplifier output load for loads of 8 and 4 ohms. The output power is a bit shy of specification into 4 ohms. In contrast to most amplifier measurements at the lower levels that are noise dominated, the Moon Neo 330A has actual distortion at 100mW that remains reasonably constant over most of the power range.

Chart 2B is a plot of THD+N for 1kHz and SMPTE IM test signals into 8 ohms. The Moon Neo 330A was not measured into 4 ohms in mono mode, as the load per channel would have been 2 ohms. This surely would have triggered its protection circuit, and the stock 5A, normal-blow AC line fuse would likely have blown.

THD+N as a function of frequency at several different power levels is plotted in Chart 3. The amount of increase in distortion with frequency is typical of many amplifiers. Also, the low-frequency region begins to be more distorted at the higher power levels. The protection circuit was triggered at the points on the graph at the low-frequency area where the trace stops shy of reaching 10Hz -- these sweeps begin at the high-frequency end and end up at 10Hz.

The Moon Neo 330A’s damping factor vs. frequency (Chart 4A) is of a shape typical of most power amplifiers. Measured in mono mode, the shape was quite a bit different for reasons I don’t understand (Chart 4B).

A spectrum of the harmonic distortion and noise residue of a 10W, 1kHz test signal is plotted in Chart 5. The AC line harmonics are a bit high and complex. The right channel (not shown) was quite a bit better. Signal harmonics are dominated by the second, third, fourth, and fifth harmonics, with higher harmonics of decreasing magnitude.

Chart 1 - Frequency response of output voltage as a function of output loading

Stereo mode
Red line = open circuit
Magenta line = 8-ohm load
Blue line = 4-ohm load

Chart 2 - Distortion as a function of power output and output loading

Chart 2A

Stereo mode
(Line up at 100W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 8-ohm SMPTE IM distortion
Third line = 8-ohm THD+N
Bottom line = 4-ohm THD+N

Chart 2B

Mono mode
(Line up at 100W to determine lines)
Top line = 8-ohm SMPTE IM distortion
Second line = 8-ohm THD+N

Chart 3 - Distortion as a function of power output and frequency

Stereo mode
(4-ohm loading)
Red line = 1W
Magenta line = 30W
Blue line = 75W
Cyan line = 150W
Yellow line = 200W

Chart 4 - Damping factor as a function of frequency

Chart 4A

Stereo mode
Damping factor = output impedance divided into 8

Chart 4B

Mono mode
Damping factor = output impedance divided into 8

Chart 5 - Distortion and noise spectrum

Stereo mode
1kHz signal at 10W into an 8-ohm load