Rogue Audio Magnum M-120 monoblock power amplifier Measurements
Due to a miscommunication, I didn't measure the Rogue Audio Magnum M-120's performance with its output stage set to ultralinear mode. However, my experience with other tube amplifiers that can be switched between ultralinear and triode operation is that the higher power available from ultralinear is associated with a higher source impedance and higher levels of distortion. So please note that all of the following measurements relate to the M-120 run in triode mode.
The Rogue's input impedance was commendably high, approaching 1 megohm over most of the audioband. While it did drop at 20kHz, this was only to 500k ohms—still very high. (Note, however, that there is a high margin of error in these measurements of very high input impedances.) The M-120 didn't invert signal polarity, and the voltage gain was a high 33.2dB from the 8 ohm output transformer tap, a still-high 30.3dB from the 4 ohm tap. Probably as a result of the high gain, the M-120's signal/noise ratio was good rather than excellent at 70dB wideband, unweighted (ref. 1W into 8 ohms). It did improve to 83dB when A-weighted.
The M-120's output impedance in the midrange was on the low side for a tube design, at 0.3 ohm (8 ohm tap) and 0.2 ohm (4 ohm tap). It rose at the edges of the audioband, however, to 0.5 ohm (8 ohm) and 0.33 ohm (4 ohm). As a result of the generally low value, any modification of the amplifier's frequency response due to impedance interactions was mild. The top trace in fig.1, for example, shows that variations with our simulated speaker load are no greater than ±0.2dB from the 4 ohm tap. The response into a resistive load was fundamentally flat over the main audioband, with slight top- and bottom-octave rolloffs of around 0.5dB. The ultrasonic -3dB point was a respectable 70kHz, which correlates with the relatively short risetime seen in the amplifier's reproduction of a 10kHz squarewave (fig.2). There is also no overshoot or ringing to be seen in this graph.
Fig.1 Rogue Magnum M-120, 4 ohm tap, 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 Rogue Magnum M-120, 4 ohm tap, small-signal 10kHz squarewave into 8 ohms.
Harmonic distortion levels were satisfactorily low in the midrange, provided the load impedance was not too much below the nominal transformer tap. This can be seen in figs. 3 and 4, which plot the small-signal THD+noise percentage against frequency from the 8 and 4 ohm outputs, respectively. However, these graphs also reveal that the distortion rises at low and high frequencies. The former is possibly due to the onset of saturation of the output transformer core, and is moderate in level; the latter is potentially more bothersome. Figs. 5 and 6 reveal that the distortion content in the midrange is heavily third harmonic, with some second harmonic also present. This will be subjectively less problematic than high-order harmonics. At low frequencies and high levels, the second harmonic becomes more dominant (fig.7). This is not what I expect from transformer limitations; perhaps there is some other non-linear mechanism present at low frequencies. (Unloading the amplifier, by the way, produced a slight instability in gain at very low frequencies.)
Fig.3 Rogue Magnum M-120, 4 ohm tap, THD+N (%) vs frequency at 2.83V into (from bottom to top at 4kHz): 8 ohms, 4 ohms, 2 ohms, simulated loudspeaker load.
Fig.4 Rogue Magnum M-120, 8 ohm tap, THD+N (%) vs frequency at 2.83V into (from bottom to top at 4kHz): 8 ohms, 4 ohms, 2 ohms, simulated loudspeaker load.
Fig.5 Rogue Magnum M-120, 8 ohm tap, 1kHz waveform at 10W into 8 ohms (top), 0.06% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.6 Rogue Magnum M-120, 4 ohm tap, spectrum of 1kHz sinewave, DC-20kHz, at 57W into 4 ohms (linear frequency scale).
Fig.7 Rogue Magnum M-120, 8 ohm tap, spectrum of 50Hz sinewave, DC-1kHz, at 60W into 8 ohms (linear frequency scale).
Its poor high-frequency linearity meant that the M-120 didn't do well on the demanding HF intermodulation test (fig.8). Just below visible clipping on the oscilloscope screen from the 4 ohm tap into 4 ohms (fig.8), the 1kHz difference component lay at an audible -43dB (0.65%). The "fuzz" around each of the spectral lines in this graph is due to the presence of AC supply noise, implying that the amplifier works very hard under these conditions.
Fig.8 Rogue Magnum M-120, 4 ohm tap, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 30W into 4 ohms (linear frequency scale).
Finally, I looked at the M-120's maximum output power using the Miller Audio Research Amplifier Profiler, which provides a low-duty-cycle 1kHz toneburst to simulate the behavior of transient-rich music program. Fig.9 (taken from the 4 ohm tap) reveals that the maximum power is available when the load is half the nominal transformer tap, though this is obtained at higher THD levels than when the load is exactly matched to the transformer. At our usual 1% THD clipping point, the M-120 delivered 99.75W into 8 ohms (20dBW) from the 8 ohm tap, 98.6W into 4 ohms (16.9dBW) from the 4 ohm tap. The 8 ohm tap would deliver 114W into 4 ohms, the 4 ohm tap 112W into 2 ohms, which suggests a good transformer design. The maximum peak current I could get from the amplifier on this toneburst signal was a high 10.3A into 1 ohm from the 4 ohm transformer, implying that the M-120 would be a good choice for use with difficult speaker loads. (Remember that these powers were with the M-120 set to triode operation.)
Fig.9 Rogue Magnum M-120, 4 ohm tap, distortion (%) vs 1kHz burst output power into: 16 ohms (red trace), 8 ohms (black), 4 ohms (blue), 2 ohms (green), 1 ohm (magenta).
Poor high-frequency linearity aside, the Rogue Audio Magnum M-120's measured performance is very respectable for a tube design.—John Atkinson