Classé Audio Fifteen power amplifier Measurements

Sidebar 3: Measurements

A full set of measurements of the Classé Fifteen was made in its balanced mode; selected measurements were repeated for the unbalanced configuration. Unless otherwise noted, the results presented are for the balanced configuration.

Following the 1/3-power, one-hour preconditioning test, the Fifteen's heatsinks were hot, but not too hot to touch comfortably. The Fifteen's input impedance measured under 76k ohms, balanced and unbalanced. Its output impedance was under 0.017 ohms at either 1kHz or 20Hz, increasing to (or under) 0.045 ohms at 20kHz. Gain into 8 ohms measured 23.3dB balanced (interestingly, the unbalanced gain measured 6dB higher, at 29.2dB—opposite the normal condition). The Fifteen is noninverting in the unbalanced mode; in the balanced, pin 2 is configured as the positive leg, pin 3 the negative. DC offset, though fluctuating with time, measured 37mV in the left channel, and a slightly high (but not disturbing) 154mV in the right.

Signal/Noise ratio (unweighted at 1W into 8 ohms) measured between 81.3 and 82.3dB, left or right, balanced or unbalanced. However, when we used the Fifteen for our last set of loudspeaker listening tests, the right channel occasionally developed an intermittent sputtering noise, which could be cleared by shutting the amp down, then powering it up again. (The problem never occurred during a specific listening session, only between sessions when the amp had to be briefly shut down to change cables.) I had the same problem on the bench; the sputtering would degrade the S/N to about 35dB, but could be cleared before I made the final measurements. This was clearly a sample defect. Once the problem was cleared, I had no further problems with the measurements—which are good enough that I feel confident that they're representative of typical Fifteens.

Fig.1 shows the frequency response of the Fifteen driven from its balanced inputs at 2W into 4 ohms. The response into 8 ohms, and from the unbalanced inputs (not shown), is virtually the same. The shape of a small-signal 10kHz squarewave in the balanced mode is shown in fig.2. The rounding on the leading edge is a result of the HF rolloff visible in fig.1 in the top audible octave and above. The response is down just under 1dB by 20kHz, which should have little audible effect. The 1kHz squarewave response (not shown) was virtually perfect.

Fig.1 Classé Fifteen, frequency response at 2W into 4 ohms (right channel dashed, 0.5dB/vertical div.).

Fig.2 Classé Fifteen, 10kHz squarewave at 1W into 8 ohms.

The crosstalk shown in fig.3 indicates nearly identical performance between channels. The increased crosstalk at high frequencies, which is typical of many two-channel products, is a result of capacitive coupling between the channels. The 50dB channel separation at 20kHz is adequate, but it is usual to see much better performance than this.

Fig.3 Classé Fifteen, crosstalk (from top to bottom): L–R; R–L (10dB/vertical div.).

The Fifteen's THD+noise vs frequency curves are plotted in fig.4. The distortion increases at higher frequencies, more so at lower load impedances—this is typical of an amplifier that doesn't make use of large amounts of feedback. (The distortion from the unbalanced inputs, not shown, was very slightly higher.) The 1kHz distortion waveform (fig.5) shows primarily a second-harmonic component combined with some higher-order harmonics. The waveforms into other impedances were very similar to the 4 ohm result shown.

Fig.4 Classé Fifteen, THD+noise vs frequency at (from top to bottom at 1kHz): 4W into 2 ohms, 2W into 4 ohms, and 1W into 8 ohms (right channel dashed).

Fig.5 Classé Fifteen, 1kHz waveform at 5W into 4 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

The spectrum of the Fifteen's output, driving 50Hz at 222W into 4 ohms (2/3 the measured clipping level with that load), is shown in fig.6. All of the distortion artifacts are extremely low; the largest is the second harmonic at –75dB, or about 0.02%. Fig.7 shows the spectrum of a combined 19+20kHz signal—the intermodulation products resulting from an input signal consisting of an equal combination of these two frequencies—at 167W into 4 ohms (clipping was visible with this signal just above this power level). The largest artifacts here are at 18 and 21kHz (about –68dB, or 0.04%), with the next largest at 2kHz (–77dB, or about 0.015%). The 19+20kHz artifacts into 8 ohms (at 86W) were lower in level than this, and are not shown.

Fig.6 Classé Fifteen, spectrum of 50Hz sinewave, DC–1kHz, at 222W into 4 ohms (linear frequency scale). Note that the second harmonic at 100Hz is the highest in level at –75dB (0.02%).

Fig.7 Classé Fifteen, HF intermodulation spectrum, DC–22kHz, 19+20kHz at 167W into 4 ohms (linear frequency scale).

The 1kHz THD+N vs level curves are shown in fig.8. The distortion remains very low up to the knees of the curves. The Fifteen's discrete clipping levels (at 1% THD+N) are shown in Table 1. The Fifteen meets its power rating into 8 ohms handily, but is just a bit short of doubling its output as the load impedance halves—not enough to be audibly significant, however.

Fig.8 Classé Fifteen, distortion (%) vs output power into (from bottom to top at 100W): 8 ohms, 4 ohms, and 2 ohms.

Table 1: Classé Clipping Power (1% THD+N at 1kHz)

Both Channels DrivenOne Channel Driven
LoadW (dBW)W (dBW)
8178.3 (22.5)178.7 (22.5)184.3 (22.7)
4301.6 (21.8)301.7 (21.8)332.6 (22.2)
2529.5 (21.2)

The bench tests of the Classé Fifteen point to an excellent amplifier—which was Larry Greenhill's subjective impression as well.—Thomas J. Norton

Classé Audio
5070 François-Cusson
Lachine, Quebec H8T 1B3
(514) 636-6384