Lamm Industries M2.1 monoblock power amplifier Measurements
Before I performed any measurements, I ran the Lamm M2.1 monoblock at one-third power into 8 ohms for an hour, with its output stage set to its High impedance parameters. Although the amplifier runs hot, I could keep my hands on the heatsinks at the end of that period, which implies a temperature below 60 degrees C. As an amplifier with a class-A output stage runs hottest when it isn't under load and coolest at maximum power, the Lamm's output MOSFETs do appear to have a significant level of class-A operation—36W into 8 ohms, according to the specification.
The M2.1's unbalanced input impedance was a fairly high 38k ohms and the balanced figure twice that, as expected. The amplifier didn't invert absolute signal polarity through the top RCA jack or the XLR jack, but it did invert polarity through the bottom RCA jack, again as expected. (The unconnected jack has to be shorted to ground with the supplied jumper for unbalanced operation.) The voltage gain at 1kHz was significantly higher than usual, at 31.75dB into 8 ohms (balanced input, both output-stage settings). The unbalanced figure was very slightly lower, at 31.6dB. It takes less than 2V to drive the amplifier into clipping, meaning that it will be well suited for use with a passive control unit.
At 0.22 ohm, the output impedance was a little higher than is usually found with a solid-state design, presumably, like the high gain, due to the absence of loop negative feedback. This figure was identical with both output-stage settings, and was typical of the M2.1's performance in the midrange and below; though it rose to 0.26 ohm at 20kHz, the difference is not significant. However, the interaction between the amplifier's source impedance and the speaker load will result in response changes that are larger than with a low-impedance amplifier like the Parasound JC-1, which Michael Fremer reviewed in February.
This can be seen in fig.1, which shows the M2.1's frequency response plotted with it driving the magazine's standard simulated speaker load, as well as the responses into 8, 4, and 2 ohms. At +0.15d, -0.25dB, the response changes into the simulated speaker are still small in absolute terms, but might be just audible. The infrasonic response is restricted by an inter-stage coupling capacitor, but this does not affect audio frequencies. The ultrasonic response rolls off by 3dB at 156kHz, regardless of output-stage setting or input conditions. Correlating with that extended bandwidth, the M2.1's reproduction of a small-signal 10kHz squarewave (fig.2) was excellent, with short leading-edge risetimes and no sign of overshoot or ringing.
Fig.1 Lamm M1.2, balanced, 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 Lamm M1.2, small-signal 10kHz squarewave into 8 ohms.
The Lamm's unweighted, wideband signal/noise ratio was good, at 68.7dB ref. 1W into 8 ohms. The A-weighted figure was 10dB better, due to the elimination of some power-supply spuriae at 120Hz. (In this respect, the M2.1 does seem a bit sensitive to system grounding.) The amplifier's distortion level depended on the load impedance. Fig.3, taken at a moderately high level to lift the true distortion out of the noise floor, shows a small increase in THD when the load impedance is halved from 8 ohms to 4 ohms, but then a much larger increase into 2 ohms. A small rise in THD above the audioband is also evident. This graph was taken with the High impedance output-stage setting. The picture was not significantly different into 2 ohms with the Low impedance setting, though the 4 ohm THD was then as low as it was into 8 ohms.
Fig.3 Lamm M1.2, balanced, High impedance setting, THD+N (%) vs frequency (from bottom to top at 4kHz): 12V into 8 ohms, 4 ohms, 2 ohms.
At low output powers, such as 1W into 8 ohms, the distortion was low in level, with the second harmonic dominant, at -81.3dB (0.009%), and all other harmonics below -100dB. But as the output power increased, the third harmonic increasingly became apparent (fig.4). The balance between the second and third harmonics at high powers appeared to depend on the output stage setting. Figs. 5 and 6 were taken at the identical power—175W into 4 ohms—but with the output stage set to Low and High impedance, respectively. The odd harmonics are less dominant in the latter condition, though the former has slightly less THD overall.
Fig.4 Lamm M1.2, balanced, High impedance setting, 1kHz waveform at 187W into 8 ohms (top), 0.16% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.5 Lamm M1.2, balanced, Low impedance setting, spectrum of 50Hz sinewave, DC-1kHz, at 175W into 4 ohms (linear frequency scale).
Fig.6 Lamm M1.2, balanced, High impedance setting, spectrum of 50Hz sinewave, DC-1kHz, at 175W into 4 ohms (linear frequency scale).
The 120Hz component I mentioned earlier can be seen in these two graphs, though it is fair to point out that this component is better than 90dB down from the signal level. There was no significant change in the harmonic spectrum between balanced and unbalanced drive, suggesting that what governs the amplifier's distortion signature is the behavior of the subsequent single-ended driver and output stages, not the input buffers and differential pair.
Correlating with the small increase in THD seen at high frequencies in fig.3, the M2.1's intermodulation behavior with the demanding 1:1 mix of 19kHz and 20kHz tones was good rather than excellent, as can be seen in fig.7, which was taken just below visible waveform clipping on the oscilloscope with the low-impedance output setting. Nevertheless, the 1kHz difference component is still 70dB down from the peak level (0.03%), which is good.
Fig.7 Lamm M1.2, balanced, Low impedance setting, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 160W into 4 ohms (linear frequency scale).
The M2.1's maximum output power depends, of course, on whether the output stage has had the voltage rails and bias current optimized for low or high impedances. Fig.8 shows the M2.1's High-impedance behavior with continuous drive. At the clipping point, defined as 1% THD, the Lamm comfortably exceeded its specification, though it is fair to note that the AC wall voltage in my lab was a rather high 124V for these tests. It delivered 297W into 8 ohms (24.75dBW), 500W into 4 ohms (24dBW), and 780W into 2 ohms (22.9dBW), even though Lamm Industries is adamant that the M2.1 should not be used to drive such low loads when it is set for high impedances. Optimally set for low impedances (fig.9), the amp delivered 190W into 8 ohms (22.8dBW), 305W into 4 ohms (21.8dBW), and 510W into 2 ohms (21.1dBW).
Fig.8 Lamm M1.2, balanced, High impedance setting, distortion (%) vs 1kHz continuous output power into (from bottom to top at 1W): 8 ohms, 4 ohms, 2 ohms.
Fig.9 Lamm M1.2, balanced, Low impedance setting, distortion (%) vs 1kHz continuous output power into (from bottom to top at 1W): 8 ohms, 4 ohms, 2 ohms.
On the face of it, even with its output stage optimized for high impedances, the Lamm amplifier gives out more clipping power into low impedances than when it is set for low impedances. But if you look at how the distortion percentage into 2 ohms changes with increasing power in these two graphs, you can see that with the amplifier set to Low impedance, it offers better linearity over a wider range of powers into 2 ohms than when it is set to High impedance. This can also be seen in fig.10, which plots the summed level of the distortion harmonics against power with a low-duty-cycle 1kHz toneburst (10 cycles on, 400 cycles off). Interestingly, the clipping powers with toneburst drive vary between just 0.1dB and 0.3dB greater than with continuous drive, the small difference indicating excellent power-supply regulation.
Fig.10 Lamm M1.2, unbalanced, Low impedance setting, distortion (%) vs 1kHz burst output power into: 8 ohms (red trace), 4 ohms (black), 2 ohms (blue), 1 ohm (green).
The Lamm M2.1's measurements reveal no significant weaknesses and imply a well-engineered design. It also appeared to be very rugged. When I drove over to Lamm, a near neighbor, to pick up a pair of amplifiers to measure—this making better logistical sense than shipping Paul Bolin's pair to Brooklyn from Minnesota and back again for him to finish the review—Vladimir was casually clipping one of the amps on his test bench with a 1kHz tone into 8 ohms, and leaving it running while we talked about the design. When I expressed some concern—I always leave the clipping tests to the end of my measurement program because that's when amplifiers break—Vladimir reached over and momentarily shorted the M2.1's output to ground. There was an impressive blue spark, but the amplifier safely turned itself off. Its red front-panel LED flashed away for 14 seconds, after which it turned back on and continued to drive full power into the load.
The M2.1 may be an expensive amplifier, but, in addition to fine sound quality, you also get a bombproof design that should deliver that sound for many years.—John Atkinson