Audiopax Model Eighty Eight monoblock power amplifier Measurements
The Audiopax Model Eighty Eight's two Timbre controls make possible a large number of combinations. I decided, therefore, to simplify matters by looking at the drop-dead-gorgeous amplifier's measured performance with both controls set to their fully clockwise positions, which I assumed was maximum bias current, and set as Bob Deutsch preferred: the left control to "11," the right to "3."
The input impedance at 1kHz was a usefully high 229k ohms, and the amplifier inverted absolute polarity. The voltage gain into 8 ohms varied somewhat according to the positions of the Timbre controls. With them set fully clockwise, the gain was a low 17.15dB, dropping by 0.5dB with them rotated to "3" and "3," and reaching -0.85dB at "8" and "8." The reason for this is that the output impedance also varied with the control settings. It was lowest with them at full clockwise, but was still very high in absolute terms at 3.5 ohms across most of the audioband. Unusually, the source impedance dropped at low frequencies, reaching 1.8 ohms at 20Hz.
As a result of this high source impedance, there will be considerable and audible modification of the amplifier's frequency response due to the Ohm's Law interaction between it and the speaker impedance. This can be seen in fig.1 as fluctuations of ±2dB in the response taken with the Audiopax driving our simulated speaker load. Other than that, note the wide bandwidth. The amplifier's small-signal response is down 0.5dB at 25Hz and 75kHz, the latter correlating with the excellent shape of a 10kHz squarewave (fig.2).
Fig.1 Audiopax Eighty Eight, frequency response at (from top to bottom at 2kHz): 2.83V into dummy loudspeaker load, 1W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms (0.5dB/vertical div.).
Fig.2 Audiopax Eighty Eight, small-signal 10kHz squarewave into 8 ohms.
Despite its low gain, the Eighty Eight's noise floor was not quite as low as I was expecting. Referenced to 1W into 8 ohms, the unweighted, wideband signal/noise ratio was 72.8dB, this figure improving to 88.4dB when A-weighted. The amplifier's linearity depended on both the load and the setting of the Timbre controls. The traces in fig.3 were taken at 2.83V output with the controls fully clockwise, and plot the THD+noise percentage against frequency into resistive loads ranging from 16 to 2 ohms, as well as into our simulated speaker. The measured figure drops below 0.1% only in the low treble and into 16 ohms. Under all other conditions, the distortion at this output voltage is high enough to be audible, depending on its spectrum.
Fig.3 Audiopax Eighty Eight, Timbre controls fully clockwise, THD+N (%) vs frequency (from bottom to top at 1kHz): 2.83V into 16 ohms, dummy loudspeaker load, 8 ohms, 4 ohms, 2 ohms.
The picture is even worse with the Timbre controls set to "11" and "3," which is how RD used the amplifier (fig.4). Why, then, was he not bothered by this nonlinearity in his auditioning? First, his Avantgarde horn speakers play very loud when driven by only a few milliwatts, and, as will be shown later, the Audiopax's linearity improves significantly as the output level drops. Second, the Eighty Eight is least linear at low frequencies, which is the region covered by the Avantgarde's powered woofers. And third, the distortion content at these levels is almost pure second harmonic (fig.5), to which the ear is relatively tolerant.
Fig.4 Audiopax Eighty Eight, Timbre controls set to "11" and "3," THD+N (%) vs frequency (from bottom to top at 1kHz): 2.83V into 16 ohms, dummy loudspeaker load, 8 ohms, 4 ohms, 2 ohms.
Fig.5 Audiopax Eighty Eight, Timbre controls fully clockwise, 1kHz waveform at 1W into 8 ohms (top), 0.45% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
At low frequencies (fig.6), the second harmonic is joined by a regularly descending series of higher harmonics, this spectrum not being significantly altered by the Timbre controls. At higher frequencies, however, the Timbre controls did introduce changes in the amplifier's distortion signature. Fig.7 was taken with the controls set fully clockwise; as expected from fig.5, the second harmonic is the highest in level, the third 20dB lower, and all other harmonics are either absent or approach -100dB. At RD's preferred control settings (fig.8), the second harmonic has doubled in level, going from -46dB to -40dB (0.5-1%), and has been joined by higher-order components, though it is fair to note that these are still very low in level.
Fig.6 Audiopax Eighty Eight, Timbre controls fully clockwise, spectrum of 50Hz sinewave, DC-1kHz, at 1W into 8 ohms (linear frequency scale).
Fig.7 Audiopax Eighty Eight, Timbre controls fully clockwise, spectrum of 1kHz sinewave, DC-10kHz, at 1W into 8 ohms (linear frequency scale).
Fig.8 Audiopax Eighty Eight, Timbre controls set to "11" and "3," spectrum of 1kHz sinewave, DC-10kHz, at 1W into 8 ohms (linear frequency scale).
The benign nature of added second-harmonic content will depend on how much intermodulation distortion accompanies it. On this account, the Eighty Eight performs quite well. Fig.9 reveals that, even at a power level well above what would have been used with the Avantgarde horns, the difference product resulting from an equal mix of 19kHz and 20kHz tones lay at -46dB (0.5%). Again, however, the level of this beat tone was affected by the Timbre controls, RD's preferred setting increasing it to 1%.
Fig.9 Audiopax Eighty Eight, Timbre controls fully clockwise, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 5W into 8 ohms (linear frequency scale).
Finally, fig.10 plots the percentage of distortion and noise in the Audiopax's output against power into 16, 8, 4, and 2 ohms. Only into 16 ohms (lowest trace) did the Eighty Eight offer relatively linear output to levels that will be usefully high with anything other than horn speakers, reaching the traditional 1% THD clip point at 17W (15.3dBW). Into lower impedances, the amplifier's "bent" transfer function resulted in steadily increasing levels of distortion with increasing power. At 1% THD, 5.2W was available into 8 ohms (7.2dBW), while relaxing the definition of clipping to 3% THD resulted in 28W (14.5dBW). However, the distortion into lower impedances can be seen to rise to high levels at modest power levels.
Fig.10 Audiopax Eighty Eight, Timbre controls fully clockwise, distortion (%) vs 1kHz continuous output power into (from bottom to top at 1W): 16 ohms, 8 ohms, 4 ohms, 2 ohms.
The Audiopax Model Eighty Eight offers acceptable measured performance only when used in the midrange and above and only at low levels into high impedances. This pretty much mandates its use with horn speakers that have powered woofers. The three Avantgarde speaker models would therefore seem to be its natural match.—John Atkinson