Classé Omega monoblock power amplifier Measurements

Sidebar 3: Measurements

Following the usual IEC preconditioning of running an amplifier at one-third power into 8 ohms for an hour, the Classé Omega's chassis was hot. However, other than the side-mounted heatsink, which was above 60 degrees C, it was not too hot to keep my hand on. An interesting fact emerged from this preconditioning: The Omega's measured THD percentage dropped from an admittedly low 0.005% when the amplifier was cold, to 0.0018% when fully warmed-up.

At 29dB from either input, the Omega's voltage gain into 8 ohms was to specification but slightly higher than average. The input impedance at 1kHz was to specification at 100k ohms unbalanced, 200k ohms balanced. The XLR is wired with pin 2 hot, and the amplifier doesn't invert signal polarity. The output impedance measured 0.13 ohm over most of the audioband (including 6' speaker cables), this figure rising slightly, to 0.15 ohm, at 20kHz.

As a result of this low source impedance, there was very little modification of the Omega's frequency response by the Ohm's Law interaction between it and the manner in which our simulated speaker load's impedance varied with frequency (fig.1, top trace). The response into an 8 ohm resistive load was flat from below the audioband to 20kHz, with then a smooth rolloff reaching -3dB at a sensible 148kHz. This was the same into all impedances, balanced or unbalanced. A 10kHz squarewave was reproduced with a very short risetime, with no hint of ringing or overshoot (fig.2). The shape of a 1kHz squarewave (not shown) was essentially perfect.

Fig.1 Classé Omega, frequency response at (from top to bottom at 2kHz): 2.83V into simulated loudspeaker load, 1W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms (0.5dB/vertical div., right channel dashed).

Fig.2 Classé Omega, small-signal 10kHz squarewave into 8 ohms.

The Omega was extremely quiet, its wideband, unweighted signal/noise ratio measuring 86.4dB (ref. 1W into 8 ohms). The A-weighted figure was almost 10dB better. As well as low noise, the Omega featured very low levels of harmonic distortion. To lift the true distortion out of the noise floor, I plotted the percentage of THD+noise against frequency at 9V rather than the usual 2.83V: 10W into 8 ohms compared with 1W. Nevertheless, as can be seen from fig.3, taken via the unbalanced input, the THD figure remained below 0.01% over most of the audioband, rising only at high frequencies and with reduced load impedance. Even so, this is a superbly linear amplifier. It was even more so through its balanced input, which more than halved the measured THD percentage (not shown).

Fig.3 Classé Omega, THD+N (%) vs frequency (from bottom to top at 40Hz): 9V into 8 and 4 ohms, 2.83V into simulated loudspeaker load, 9V into 2 ohms.

The spectral content of the distortion is heavily third-harmonic (fig.4). Though some higher-order harmonics are present (fig.5), these all lie at the -100dB (0.0001%) level. Intermodulation distortion was missing in action at low powers. With an equal mix of 19+20kHz tones (fig.6) at 350W into 4 ohms (!), some 1kHz difference component is apparent, though this is still almost entirely due to the D/A converter used as a source. The spectral components at 18kHz and 21kHz in this graph are due to the Omega, but, at -80dB (0.01%), won't bother anyone.

Fig.4 Classé Omega, 1kHz waveform at 200W into 4 ohms (top), 0.01% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.5 Classé Omega, spectrum of 50Hz sinewave, DC-1kHz, at 500W into 4 ohms (linear frequency scale).

Fig.6 Classé Omega, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 350W into 4 ohms (linear frequency scale).

always leave the clipping power tests to last, because that is where amplifiers tend to break. (I place the amplifier being tested in the corridor outside my lab for this test.) Not the Omega. I clipped it with impunity into 8, 4, and 2 ohms with continuous drive. The manner in which the THD+N percentage changes with output power into these loads is shown in fig.7. Below 10W, noise dominates the reading. Between 10W or so and the "knee" in each trace, the distortion starts to increase slightly, more so into the lower impedances, but the Omega basically remains extremely linear right up to the point where it runs out of headroom.

Fig.7 Classé Omega, distortion (%) vs 1kHz continuous output power into (from bottom to top): 8 ohms, 4 ohms, 2 ohms.

The Omega more than exceeded its specified power. Defining clipping as 1% THD, the actual clipping powers available were: 590W into 8 ohms (27.7dBW), 1400W into 4 ohms (28.45dBW), and 2kW into 2 ohms (27dBW). Unusually, the 4 ohm delivery was slightly more than twice the 8 ohm figure, which means that its maximum output voltage is slightly higher into the lower impedance. I was suspicious of this—perhaps the dummy load had got so hot that its resistance was higher than 4 ohms, so I repeated the test with a different, cold dummy load. The result was both the same and repeatable, so it appears to be real. Perhas it has something to do with the amplifier's output regulation?

During these last tests, by the way, I inadvertently knocked the RCA connector out of my Audio Precision System One's output jack while the Omega was driving 500W into 8 ohms. The amplifier immediately turned it itself off, the front-panel logo glowing red. After a hard reboot—disconnecting the AC cable—the Omega performed flawlessly, a tribute to its protection circuitry.

When you purchase a very expensive high-end amplifier, I think you should expect enormous power to be delivered into all normal impedances, with vanishingly low distortion and without the amplifier breaking. This the Classé Omega can do with aplomb and style.—John Atkinson

5070 Francois-Cusson
Lachine, Quebec H8T 1B3
(514) 636-6384