Jeff Rowland Design Group Model 2 power amplifier 1997 Measurements

Sidebar 4: 1997 Measurements

All measurements on the Rowland Model 2 were made with the amplifier in its high-impedance input setting at a gain of 26dB, unless otherwise noted. All of the measurements were, of course, balanced. And all of the measurements, again except as noted, were made with the battery functioning normally (battery connected to the line for constant recharge, amp operating off the battery).

Following its 1/3-power, one-hour preconditioning test, the Model 2's heatsinks were very warm but not hot. The Rowland Model 2's input impedance measured 44k ohms (720 ohms in its low-input impedance setting). The output impedance measured under 0.025 ohms at 20Hz and 1kHz, increasing to a maximum of 0.35 ohms at 20kHz. Voltage gain into 8 ohms measured 26dB (32.1dB at the 32dB setting). Signal/Noise ref. 1W into 8 ohms measured 83dB over a 22Hz-22kHz bandwidth, unweighted; 72.2dB over a 10Hz-500kHz bandwidth, unweighted; and 88dB, A-weighted. Like other Rowland gear, the Model 2 is pin-3 positive at its inputs. DC offset measured a negligible 0.2mV in the left channel, 0.1mV in the right.

Fig.1 shows the frequency response of the Model 2, which is virtually flat across the audible range (it was identical with direct-line operation; ie, battery off). The 10kHz squarewave in fig.2 shows just a slight overshoot on the leading edge, but is otherwise very good. The 1kHz squarewave, not shown, is also excellent, and also barely shows this overshoot. The squarewave response is not changed when the amplifier is switched to direct-line power (battery off). The Model 2's crosstalk (not shown) was very low, and the two channels are very closely matched—certainly indicative of careful design, though probably not significant at such low crosstalk levels.

Fig.1 Rowland Model 2/BPS-2, frequency response at 1W into 8 ohms (right channel dashed, 0.5dB/vertical div.).

Fig.2 Rowland Model 2/BPS-2, small-signal 10kHz squarewave into 8 ohms.

The THD+noise vs frequency curves are plotted in fig.3. The result was identical with the amplifier powered directly from the line (battery off). The waveform of the distortion at 2W into 4 ohms is shown in fig.4. It is heavily third-harmonic, though there are clearly some higher-order components, plus the usual noise visible with such low levels of distortion. The waveforms into 8 and 2 ohms (not shown) were similar.

Fig.3 Rowland Model 2/BPS-2, THD+noise vs frequency at (from top to bottom at 10kHz): 4W into 2 ohms, 2W into 4 ohms, 1W into 8 ohms, 2.83V into simulated speaker load (right channel dashed).

Fig.4 Rowland Model 2/BPS-2, 1kHz waveform at 2W into 4 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

The Model 2's output spectrum reproducing a 50Hz input at 100W into 4 ohms is shown in fig.5 The distortion products are all extremely low—below -90dB, or 0.003%. Into our simulated loudspeaker load at a level of 21.9V, only the third harmonic (at -75dB, or 0.17%) was in any way relevant, though other artifacts at less than -80dB are visible.

Fig.5 Rowland Model 2/BPS-2, spectrum of 50Hz sinewave, DC-1kHz, at 100W into 4 ohms (linear frequency scale).

Fig.6 shows the amplifier's output spectrum reproducing a combination of 19+20kHz signals at 66.4W into 4 ohms (visible clipping was present above these output levels with this input signal). While there are more intermodulation products visible with a 4 ohm load, in all cases the artifacts are very low—a maximum of -70dB (0.03%) at 21kHz. (Note that the trace in this graph has been plotted 6dB too high in level.)

Fig.6 Rowland Model 2/BPS-2, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 66.4W into 4 ohms (linear frequency scale)

The 1kHz, THD+noise vs output power for the Model 2 are shown in fig.7. The discrete clipping levels for the Model 2 are shown in Table 1, along with the results for direct-line operation (battery off).

Fig.7 Rowland Model 2/BPS-2, distortion (%) vs output power into (from bottom to top at 40W): 8 ohms, 4 ohms, and 2 ohms.

Table 1 Jeff Rowland Design Group Model 2

Discrete Clipping levels (1% THD+noise at 1kHz)
 Both Channels DrivenOne Channel Driven
ImpedanceW (dBW)W (dBW)
8 (battery)74.8 (18.7)74.9 (18.7)81.3 (19.1)
(no battery)99 (19.2)99.2 (20)121.6 (20.8)
4 (battery)116 (17.6)118 (17.7)134 (17.7)
(no battery)126.2 (18)126 (18)170.3 (19.3)
2 (battery)  184.5 (16.7)
(no battery)  199.6 (17)

Note that the Model 2 puts out less power under battery drive, and does not quite meet its 4 ohm specification of 150W under any conditions except that of line drive, one channel driven. I also noticed that the distortion increased gradually at high output when the amplifier was driven through the battery. For example, with one channel driven into 4 ohms, the power listed above at 1% distortion was only good for a few seconds before the distortion would start to increase gradually and constantly. After 30 seconds, it was up to 1.25% and still increasing. On line drive under the same conditions, the distortion remained stable. This suggests that the power available from the battery is slightly degraded if called upon to drive the amplifier at high power for an extended period—not typical of music.

Nevertheless, the battery-powered Rowland Model 2 produced a solid set of measurements especially notable for low distortion, noise, and crosstalk.—Thomas J. Norton