Onkyo A-9555 integrated amplifier Measurements
Looking first at the Onkyo A-9555's phono stage, this was suitable for moving-magnet cartridges, with a voltage gain a little lower than normal of 35.6dB (measured at the amplifier's Tape Out jacks) and an input impedance of 45k ohms. The RIAA equalization featured a treble region that was shelved down by 0.75dB (fig.1), which will be audible as a slight dulling of the balance. The two channels matched very well, however.
Fig.1 Onkyo A-9555, phono-stage RIAA error at 5mV input at 1kHz (0.5dB/vertical div., right channel dashed).
The phono input's unweighted, wideband signal/noise ratio (ref. 5mV at 1kHz) was a good 61.1dB, this improving to 75.4dB when A-weighted. The overload margin was excellent, at 23.5dB across the audioband.
Turning to the Onkyo's line-stage performance, I preconditioned the amplifier by running both channels for 60 minutes at one-third the specified power into 8 ohms. This is thermally the worst case for an amplifier with a class-B or-AB output stage, but it should cause no problems for an amplifier with a class-D output stage, such as the Onkyo. After an hour at 35Wpc into 8 ohms, the amplifier's chassis was slightly warm, however.
The maximum voltage gain into 8 ohms via the CD input in Pure Direct Mode was a high 45dB. The line input reserved absolute polarity; ie, was non-inverting. The input impedance was a usefully high 103k ohms at low and middle frequencies, dropping slightly but inconsequentially to 88k ohms at 20kHz. The output impedance at low and middle frequencies was a little higher than usual for a solid-state amplifier, at 0.24 ohm. At 20kHz, however, it rose to almost 2 ohms, due to the presence of the essential low-pass filter between the amplifier's switching output stage and the speaker terminals.
Unfortunately, the tuning of this filter depends on the load impedance. Fig.2 reveals that while it does roll off the amplifier's response at radio frequencies, it also peaks the response just prior to that rolloff. Both the highest-level peak and the lowest in frequency occur into 8 ohms (fig.2, top pair of traces), while a 2 ohm load rolls off the audioband response by more than 3dB at 20kHz, before peaking at 100kHz (bottom trace). This ultrasonic peak in the response results in a significant overshoot on the A-9555's reproduction of a 10kHz squarewave (fig.3), though this overshoot is critically damped; ie, it returns to the waveform top and bottom without ringing.
Fig.2 Onkyo A-9555, frequency response at 2.83V into (from top to bottom at 2kHz): simulated loudspeaker load, 8, 4, 2 ohms (2dB/vertical div., right channel dashed).
Fig.3 Onkyo A-9555, small-signal 10kHz squarewave into 8 ohms.
Figs.2 and 3 were taken with the tone controls switched out of circuit. To look at the effect of these controls, I reduced the level by 20dB and ran a set of response curves into 8 ohms with the controls set to their maximum and minimum positions (Loudness was set to Off). The results are shown in fig.4. The central two traces are with the controls centered; they basically echo the 8-ohm trace in fig.2, but with a 0.75dB channel imbalance, this due to the volume control being set to 12:00 rather than its maximum. With both the Bass and Treble controls set to their maximum (top pair of traces), the low and high frequencies are boosted by 17dB at 20Hz and 14dB at 20kHz, both higher than the 10dB specified, though the response continues to climb above 20kHz, due to the filter interaction examined earlier. With the controls set to their minimum positions (bottom pair of traces), the amplifier's output was now –13dB at 20Hz and –11dB at 17kHz. Peculiarly, this applied only to the left channel (bottom solid trace). The right channel's output (bottom dashed trace) was reduced only slightly. Perhaps I should have rebooted the amplifier's control section (by holding down the Pure Direct button, then the Loudness button, then releasing both) before performing the tone control measurements, but circumstances intervened (see later).
Fig.4 Onkyo A-9555, effect of tone controls set to their maximum and minimum positions, into 8 ohms, volume control at 12:00, Loudness control off (5dB/vertical div., right channel dashed).
Channel separation (not shown) was almost 90dB at 1kHz, but decreased above that frequency due to the usual capacitive coupling. Amplifiers with class-D output stages always have poor wideband S/N performance, due to the residual RF energy getting past the output filter. With its volume control at its maximum but the input short-circuited, the A-9555's wideband, unweighted S/N ratio was 41.6dB ref. 1W into 8 ohms, due to 25mV or so of RF noise present on the output. Reducing the measurement bandwidth to the audioband improved the measurement to a good 72.8dB, while switching in an A-weighting filter improved it even further, to 84.2dB.
The presence of this RF noise makes measuring a class-D amplifier's distortion problematic, because the RF content can drive the analyzer's input stage into slewing overload. To assess the Onkyo A-9555's circuit linearity, I inserted an Audio Precision AUX-0025 passive low-pass filter between the test load and my System One analyzer. Fig.5 shows how the percentage of THD+noise in the Onkyo's output varies with output power into loads ranging from 8 to 2 ohms. Defining clipping as 1% THD, the A-9555 comfortably exceeded its rated power, clipping at 110Wpc into 8 ohms (20.4dBW), 175Wpc into 4 ohms (19.4dBW), and, with one channel driven rather than two, 245W into 2 ohms (17.9dBW) (footnote 1). There is a peculiar rise in the A-9555's THD+N percentage when the output reaches a few watts; if you look at the output current into the three impedances when this occurs, it happens just above 1A in each case.
Fig.5 Onkyo A-9555, with AP AUX-0025 low-pass filter, distortion (%)vs 1kHz continuous output power into (from bottom to top at 10W): 8, 4, 2 ohms.
I examined how the Onkyo's THD+N percentage changed with frequency at a level of approximately 5V (fig.6). The amplifier has moderately low distortion across the audioband, but is less comfortable driving 2 ohms. This is graphically shown in figs. 7 and 8, which show the distortion+noise residual waveform with the amplifier driving 8 and 4 ohms, respectively. Into 8 ohms (fig.7), the distortion content is predominantly the subjectively innocuous second harmonic, but at almost the same voltage into 4 ohms (fig.8), the fifth harmonic makes a strong appearance.
Fig.6 Onkyo A-9555, with AP AUX-0025 low-pass filter, THD+N (%)vs frequency at 8V into (from bottom to top): 8, 4, 2 ohms (right channel dashed).
Fig.7 Onkyo A-9555, with AP AUX-0025 low-pass filter, 1kHz waveform at 4W into 8 ohms (top), 0.086% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.8 Onkyo A-9555, with AP AUX-0025 low-pass filter, 1kHz waveform at 7.55W into 4 ohms (top), 0.112% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
Looking at the Onkyo A-9555's distortion from a different direction, fig.9 shows the spectrum of the amplifier's output while it reproduced a 50Hz sinewave at high power into 8 ohms. The second harmonic is still the highest in level, but has now been joined by the third, with some higher-order components also visible. But I was puzzled by the appearance of a spectral line at 170Hz, reaching –80dB (0.01%). This frequency is the sum of the signal frequency and the 120Hz power-supply frequency, suggesting that some supply modulation is taking place. Repeating the spectral analysis with a 1kHz tone at the same level gave the graph shown in fig.10. The harmonic components are very similar in nature and level to those accompanying the low-frequency tone in fig.9, but the higher frequency of the test tone reveals that, yes indeed, there are 120Hz-spaced sidebands present around its spectral line. Again, the highest in level lie at –80dB, which might be thought to be negligible, but this behavior does suggest that the amplifier is working hard at this level, even though it lies 3dB below the actual clipping point.
Fig.9 Onkyo A-9555, with AP AUX-0025 low-pass filter, spectrum of 50Hz sinewave, DC–1kHz, at 50W into 8 ohms (linear frequency scale).
Fig.10 Onkyo A-9555, with AP AUX-0025 low-pass filter, spectrum of 1kHz sinewave, DC–10kHz, at 50W into 8 ohms (linear frequency scale).
Finally, these sidebands can also be seen in the spectral analysis of the A-9555's output when it drove an equal mix of 19 and 20kHz tones into 8 ohms (fig.11), just below the point where visible waveform clipping could be seen on the oscilloscope. The actual signal intermodulation, however, is quite low in level, the 1kHz difference component lying at –68.5dB (0.04%).
Fig.11 Onkyo A-9555, with AP AUX-0025 low-pass filter, HF intermodulation spectrum, DC–24kHz, 19+20kHz at 60W peak into 8 ohms (linear frequency scale).
Amplifiers featuring class-D output stages are "green" in that they are very efficient at turning current from the wall outlet into sound. However, they also offer a set of performance attributes different from those of a conventional amplifier, not least because they have to incorporate a low-pass filter between the output stage and the speaker terminals. The Onkyo A-9555 offers quite respectable measured performance for its price, but it is clearly more comfortable driving higher impedances.—John Atkinson
Footnote 1: I always perform this test last, because it is the most stressful for an amplifier. The Onkyo switched itself off when the output level reached 300W into 2 ohms, and the power switch glowed a constant red. Following the advice in the excellent manual, I unplugged the amp from the wall and let it stand for a couple of minutes. When I plugged it back in and pressed the power switch, the amplifier did turn on but the blue LED on the volume control flashed continually, the mute wouldn't lift, and no signal was passed. I suspect an internal fuse had blown.—John Atkinson