Bryston 9B-THX five-channel power amplifier Measurements
The five-channel Bryston 9B-THX is the first multichannel amplifier I've measured for Stereophile, so I had to scramble a little to get five identical dummy loads. Even then, the Audio Precision System One I use can measure only two channels at a time, so I generally looked at just one channel. However, for the power testing I did drive all five channels for the 8 ohm measurement, and two for the 4 ohm result. Although the 9B has a tip-ring-sleeve balanced input connector for each channel, I tested it only using its singled-ended RCA inputs. I also didn't measure channel separation—mea culpa—but, given the amplifier's independent-module construction, I would be astonished to find any significant crosstalk present.
The 9B-THX was preconditioned by being driven (all five channels) at one-third power for one hour. This maximally stresses an amplifier with a class-B output stage, and, after the hour, the exposed edges of the Bryston's internal heatsinks were too hot to keep my hand on, implying a temperature higher than 60 degrees C (140 degrees F). The chassis, however, while hot, was not too hot to be touched, implying sensibly arranged heatsinking.
The amplifier was non-inverting via its single-ended input. The voltage gain measured 29.2dB into 8 ohms, a 100mV input resulting in a power delivery of just over 1W into 8 ohms. The input impedance was suitably high at 47.6k ohms from the bass through the mid-treble, this dropping to 41.5k ohms at 20kHz. (On the suggestion of a reader, we are now measuring input impedance across the audioband rather than just at 1kHz, as we used to.) The output impedance was a very low 0.04 ohms except at 20kHz, where it rose to a still impressive 0.07 ohms.
As a result, there will be hardly any resistive divider interaction between the amplifier and the speakers it drives. This can be seen in fig.1, where the small-signal frequency response was hardly affected by the simulated loudspeaker load. The bass is extended to the 10Hz lower limit of the graph, while the highs roll off above the audioband, reaching their 3dB-down point around 180kHz. Not surprisingly, the small-signal 10kHz squarewave response (fig.2) was essentially perfect.
Fig.1 Bryston 9B-THX, frequency response at (from top to bottom at 20kHz): 1W into 8 ohms, 2W into 4 ohms, and 2.83V into dummy loudspeaker load (0.5dB/vertical div.).
Fig.2 Bryston 9B-THX, small-signal 10kHz squarewave into 8 ohms.
As can be seen in fig.3, the 9B offered very low distortion, except at higher frequencies into 2 ohms. What distortion was present was primarily the innocuous second harmonic (fig.4), but observe the very high power required to lift the distortion waveform out of the residual noise. Both the low level of the distortion and its second-harmonic nature are confirmed by fig.5, while fig.6 shows that intermodulation distortion products are conspicuous almost by their absence, even at the very high power used to generate this graph (just below visible clipping with this demanding signal). Note, however, the suspicious rise in the spectral noise floor around the peaks that represent the 19kHz and 20kHz tones. I have no idea what this will mean subjectively, but I'd prefer that it wasn't there.
Fig.3 Bryston 9B-THX, THD+noise (%) vs frequency at (from top to bottom at 4kHz): 4W into 2 ohms; 2.83V into simulated loudspeaker load; 2W into 4 ohms; and 1W into 8 ohms.
Fig.4 Bryston 9B-THX, 1kHz waveform at 200W into 4 ohms (top), distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.5 Bryston 9B-THX, spectrum of 50Hz sinewave, DC-1kHz, at 200W into 4 ohms (linear frequency scale).
Fig.6 Bryston 9B-THX, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 128W into 4 ohms (linear frequency scale).