Audio Alchemy HPA v1.0 headphone amplifier Measurements

Sidebar 3: Measurements

All of the following measurements were made from the front-panel headphone output. The polarity of the Audio Alchemy v1.0 was noninverting. Its input impedance was 88.2k ohms, its output impedance 0.27 to 0.29 ohms, depending on frequency. S/N measured an excellent 105dB (22Hz–22kHz, unweighted, ref. 1V). DC offset measured 12.5mV in the left channel, 11.5mV in the right. The tracking of the Audio Alchemy's volume control was reasonably good, with a maximum left/right deviation of 0.5dB at half rotation and less. Maximum voltage gain was 8.9dB.

The Audio Alchemy's frequency response into a high impedance load is shown in fig.1. The response with the HeadRoom processing selected is plotted at the top; the latter is shaped in an attempt to compensate for the typical ear/headphone response. The frequency response is plotted again in fig.2; this time the output of the left channel response is shown with the left and right channel inputs both in phase and out of phase. Here the processing is seen to be quite similar to that used by HeadRoom (see JA's review of the HeadRoom headphone amplifier in the January 1994 Stereophile, p.173). The 1kHz squarewave response is nearly ideal and is not shown. The 10kHz squarewave (fig.3), taken in the unprocessed mode (as were the remainder of the measurements below), is very good, with a good risetime, just a slight rounding of the leading edge, and no overshoot.

Fig.1 Audio Alchemy HPA, frequency response at 1V into 100k ohms with HeadRoom circuitry engaged (top below 1kHz) and bypassed (0.5dB/vertical div.).

Fig.2 Audio Alchemy HPA, frequency response at 1V into 100k ohms with HeadRoom circuitry engaged and driven by L+R signal (top below 1kHz) and L–R signal (0.5dB/vertical div.).

Fig.3 Audio Alchemy HPA, small-signal 10kHz squarewave into 100k ohms.

The Audio Alchemy's crosstalk is shown in fig.4. Notice that the channel separation is considerably reduced in the processed mode (in much the same manner as it is reduced by loudspeakers in a room).

Fig.4 Audio Alchemy HPA, crosstalk (from bottom to top at 1kHz): L–R, R–L, processing off; L–R, R–L processing on (10dB/vertical div.).

Fig.5 plots the Audio Alchemy HPA v1.0's THD+noise vs its output voltage. Note that the maximum output is about the same with either a 40 ohm or a 150 ohm load, though there is a small (and probably insignificant) difference between the two loads just before the point at which the distortion begins to rise rapidly. The output is about 6.5V at clipping (1% THD+noise). This is over 1 watt into 40 ohms—more than enough to induce permanent ear damage with virtually any headphones.

Fig.5 Audio Alchemy HPA, distortion (%) vs output voltage into (from bottom to top at 5V): 150 ohms and 40 ohms.

The variation in THD+noise vs frequency for an output of 4V into 40 ohms is shown in fig.6—an excellent result. The distortion waveform (40 ohm load, 2V or 100mW output, 1kHz input) is shown in fig.7. It is very low in level and largely second-harmonic plus noise. Driving the Audio Alchemy with a frequency of 50Hz at an output of 4V into a 40 ohm load results in the distortion spectrum shown in fig.8. Even at this very high level, the distortion products are all below –90dB (0.003%). Similarly, fig.9 shows the output intermodulation spectrum for an input of 19+20kHz at a high output voltage; all the by-products are under –80dB (0.05%). The test-bench measurements of the HPA v1.0 were first-rate in all respects.—Thomas J. Norton

Fig.6 Audio Alchemy HPA, THD+noise vs frequency at 4V into 40 ohms.

Fig.7 Audio Alchemy HPA, 1kHz waveform at 4V into 40 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.8 Audio Alchemy HPA, spectrum of 50Hz sinewave, DC–1kHz, at 4V into 40 ohms (linear frequency scale). Note that the fifth harmonic at 250Hz is the highest in level.

Fig.9 Audio Alchemy HPA, HF intermodulation spectrum, DC–22kHz, 19+20kHz at 4V into 40 ohms (linear frequency scale).

Audio Alchemy
no longer trading (2006)