Classé CAP-100 integrated amplifier Measurements

Sidebar 4: Measurements

The CAP-100 ran cool not only while playing in my listening room, but also when run at 1/3 rated power into 8 ohms. I could easily keep my hand on the chassis top.

The line input impedance measured 33k ohms (unbalanced) and 12k ohms (balanced), exactly as specified. The phono input measured 160 ohms in the moving-coil position, more than triple the 50 ohm specification. The CAP-100 doesn't invert polarity from its line inputs to loudspeaker outputs. DC offset at the loudspeaker terminals measured 2.4mV (left channel) and 10.2mV (right). Volume-control tracking was only fair: I measured a 0.4dB channel imbalance with the volume control between 10 and 11 o'clock, a typical listening setting (this was with the balance pot at its center-position detent).

Overall voltage gain into 8 ohms measured 45dB, split at 29dB from the power amplifier and 16dB from the preamplifier. The phono board in the MC setting added 43dB of gain, not the 55dB specified. The phono stage's input overload was 22mV at 1kHz, 2.2mV at 20Hz, and 220mV at 20kHz. This is plenty of overload margin for any moving-coil cartridge. The CAP-100's S/N ratio from phono input to loudspeaker output (referenced to 1W into 8 ohms) measured 71.7dB (left) and 69.9dB (right) unweighted, figures that improved to 75.3dB (left) and 74.8dB (right) when an A-weighting filter was applied. S/N was only slightly better from the line input to speaker output, measuring 74dB (left) and 73.8dB (right) unweighted, and 76.9dB (left) and 76.7dB (right) A-weighted. Note that the S/N ratio can be measured in a variety of ways: If full output power is the reference, then the S/N specification can be stated as a much higher figure. Classé specifies the CAP-100's S/N ratio as 110dB, A-weighted.

The CAP-100's output impedance measured a low 0.03 ohms across most of the band, increasing to 0.1 ohms at 20kHz. This is low enough not to cause interaction with a loudspeaker load that could introduce frequency-response variations. This can be seen in fig.1, the CAP-100's frequency responses at 1W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms, and 1W into a reactive load that simulates a real-world loudspeaker. I offset the response into the simulated loudspeaker (it's the lowermost curve) so you could see the very slight kink in the curve at about 5kHz caused by interaction between the CAP-100 and the reactive load. Amplifiers of high output impedance will exhibit response variations of as much as 4dB when driving a real loudspeaker load.

Fig.1 Classé CAP-100, line frequency response (from top to bottom): 1W into 8 ohms; 2W into 4 ohms, and 2.83V into simulated speaker load (right channel dashed, 0.5dB/vertical div.).

Fig.2, the CAP-100's small-signal 10kHz squarewave response into 4 ohms shows a rounding of the squarewave's edges that correlate with the amplifier's band-limited response. The phono board's RIAA error (fig.3) shows a 0.75dB rolloff at 20kHz, accompanied by a rising bass response. The treble rolloff encompasses a broad bandwidth; although the rolloff's amplitude isn't severe, the wide bandwidth will produce a very slight softening of the sound—not a bad thing with most moving-coil cartridges. Although the treble response measures as being slightly rolled-off, I thought the sound through the phono section was brighter than from CD (even with the slightly bright RCD-990 CD player). The lift in the bass should definitely be audible.

Fig.2 Classé CAP-100, small-signal 10kHz squarewave into 8 ohms.

Fig.3 Classé CAP-100, phono section RIAA error (right channel dashed, 0.5dB/vertical div.).

The CAP-100's crosstalk (not shown) measured 55dB at 1kHz, 20dB worse than the published specification. The crosstalk measurement was made with a 1V input and 1W output into 8 ohms.

Looking next at the CAP-100's distortion, we can see from fig.4 that the amplifier has low THD+N in the low treble and below, but it does rise at higher frequencies, particularly when the amplifier drives low impedances. Fig.5 shows the CAP-100's distortion waveform with the fundamental notched out when driving 1W into 8 ohms. The distortion is primarily third harmonic (although at a low level), but the distortion waveform is dominated by what appears to be classic crossover distortion. The spikes in the distortion waveform seem to occur at the waveform's zero crossing points, suggesting that the amplifier's output transistor pairs aren't optimally biased. Specifically, zero-crossing distortion occurs when one or both of the output transistors in the pair is incorrectly biased, leaving a gap in the waveform where neither transistor is amplifying the signal at the zero-crossing transition (where one transistor turns off and the other turns on). I could see a slight discontinuity in the waveform when looking at the CAP-100's reproduction of sinewaves on the oscilloscope. Interestingly, nothing in the auditioning suggested that anything was wrong with the amplifier. [Crossover distortion expressed as a percentage rises in measured level as the signal amplitude drops.—Ed.]

Fig.4 Classé CAP-100, THD+noise vs frequency at (from top to bottom at 10kHz): 4W into 2 ohms; 2W into 4 ohms; 1W into 8 ohms.

Fig.5 Classé CAP-100, 1kHz waveform at 5W into 8 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.6 is the CAP-100's harmonic distortion spectrum, made by driving a 4 ohm load at 100W with a 50Hz sinewave. The crossover distortion does not introduce visible harmonics at this high power level. The distortion is primarily third harmonic. The CAP-100 also introduced very little intermodulation distortion (fig.7), though again it must be noted that this measurement was taken at a very high power level, minimizing the effects of crossover distortion.

Fig.6 Classé CAP-100, spectrum of 50Hz sinewave, DC–1kHz, at 100W into 4 ohms (linear frequency scale).

Fig.7 Classé CAP-100, HF intermodulation spectrum, DC–22kHz, 19+20kHz at 100W into 4 ohms (linear frequency scale).

The CAP-100's output power vs distortion curves when driving 8, 4, and 2 ohms are shown in fig.8. These curves were made with only one channel driven. The 2 ohm curve's unusual shape above clipping was caused by the 4A rail fuse blowing at a little over 200W output. Table 1 shows the measured clipping points (1% THD+Noise) with both channels driven.

Fig.8 Classé CAP-100, distortion (%) vs output power into (from bottom to top at 10W): 8 ohms, 4 ohms, and 2 ohms.

Table 1

LeftRightOne Channel Driven
8106W (20.25)107W (20.29)118W (20.7)
4157W (18.95)158W (18.98)178W (19.5)
2200W (17.1) (fuse blew)

The CAP-100's overall bench performance was excellent, with the exception of the apparent zero-crossing distortion seen in fig.7. However, this could be a sample fault, not necessarily indicative of a design flaw.—Robert Harley

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