PS Audio HCA-2 power amplifier Measurements

Sidebar 2: Measurements

The PS Audio HCA-2 inverted absolute polarity, and its balanced and unbalanced inputs featured almost identical voltage gains: 30.2dB and 30.4dB, respectively, into 8 ohms, both of these on the high side. As a result, it took only 1V or so to drive the amplifier to clipping. The input impedance at 1kHz measured a high 98k ohms via the unbalanced RCA jack, just under half that figure via the balanced XLR jack.

The amplifier's output impedance was moderate, at 0.2 ohm over most of the audioband, rising slightly to 0.3 ohm at 20kHz. As a result, the modification of its frequency response due to the Ohm's Law interaction between this impedance and the variation with frequency of the loudspeaker impedance will be mild. However, the low-pass filter, in series with the amplifier's output to minimize the presence of radio-frequency switching noise, also interacts with the load impedance; the top trace in fig.1 reveals that it peaks a little at 40kHz when the load is 8 ohms. As the load impedance drops, the peak disappears and the response starts its ultrasonic rolloff closer to the audioband. Into 4 ohms, the PS Audio's output is 0.5dB down at 20kHz; into 2 ohms, the 20kHz level is more than 1dB down.

Fig.1 PS Audio HCA-2, 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., right channel dashed).

The HCA-2's reproduction of a 10kHz squarewave reflected this behavior. Into 8 ohms (fig.2), a slight overshoot and one cycle of 40kHz ringing can be seen; into 4 ohms (fig.3), both the overshoot and ringing are lower in level.

Fig.2 PS Audio HCA-2, channel separation (R-L dashed, 10dB/vertical div.).

Fig.3 PS Audio HCA-2, small-signal 10kHz squarewave into 8 ohms.

Fig.4 reveals that the amplifier's channel separation is quite good, at better than 70dB below 12kHz, though some capacitive coupling between channels increasingly introduces crosstalk with rising frequency. (I first thought this measurement was being affected by the presence of ultrasonic switching noise, but looking at the crosstalk waveform on an oscilloscope revealed that true crosstalk is being shown.) However, the presence of switching noise meant that the wideband, unweighted signal/noise ratio (ref. 1W into 8 ohms) was just 45.7dB. Reducing the measurement bandwidth to the audioband improved the ratio to 72.3dB, with an A-weighting filter increasing this to 76.5dB. This is still not as good as I would have hoped, but the measurement is adversely affected by the amplifier's high gain and, therefore, sensitivity.

Fig.4 PS Audio HCA-2, small-signal 10kHz squarewave into 4 ohms.

The HCA-2's measured percentage of distortion+noise was not affected by the grounding arrangement between the amplifier and the Audio Precision System One test gear. However, it did vary significantly with both frequency and load impedance, as revealed by fig.5. This graph was taken at an output level of 2.83V, equivalent to 2W into 4 ohms. The THD rises alarmingly at higher frequencies, reaching 1% into 4 ohms at 5kHz.

Fig.5 PS Audio HCA-2, THD+noise vs frequency at (from top to bottom at 1kHz): 4W into 2 ohms, 2W into 4 ohms, 1W into 8 ohms, 2.83V into dummy loudspeaker load.