Wavelength Audio Cardinal XS monoblock amplifier Measurements

Sidebar 1: Measurements

Except where noted below, the measurements of the Wavelength Cardinal were made using the Sylvania 6188 input tube, Mullard/Sylvania 5AR4 rectifier, and VAIC VV30B Type I output tube—the tube complement recommended by the manufacturer for best sound. Eight-ohm measurements were made with the 8 ohm setting of the output taps, 2 and 4 ohm readings with the 4 ohm setting.

The Wavelength Cardinal was warmed up for one hour at a third of its rated maximum power (8W). It did not run warmer than expected for this type of design. Its input impedance measured 99k ohms, and its output impedance was a very high 3.4 ohms at 1kHz, increasing to 5.9 ohms at 20Hz and 5.3 ohms at 20kHz. This amplifier's frequency response will be extremely sensitive to varying loudspeaker loads. Its voltage gain (into 8 ohms) was a very low 17.3dB—most amplifiers produce about 10dB more gain. The use of this amplifier with sources having a typical output, combined with a passive (zero-gain) preamplifier, might result in insufficient sound pressure levels. S/N measured 78.5dB over a 22Hz-22kHz bandwidth (ref. 1W into 8 ohms unweighted), 91dB A-weighted, and 76dB (unweighted) over a wider 10Hz-500kHz bandwidth. DC offset measured an inconsequential 0.2mV.

Fig.1 shows the frequency response of the Wavelength Cardinal XS. Note the variation with the simulated real-world load; the scale has been expanded here to ±4dB. The response irregularity into the simulated load, while it will not be precisely the same with all real loads, will be audible with most loudspeakers. Fig.2 show the small-signal 10kHz squarewave response. It indicates a fairly slow risetime with a small degree of well-damped oscillation (visible at the top and bottom of the waveform). The 1kHz response, not shown, is reasonably clean, with a small downward tilt indicative of the low-frequency rolloff visible in fig.1.

Fig.1 Wavelength Cardinal XS, frequency response at (from top to bottom): 2W into 4 ohms, 4 ohm tap; 1W into 8 ohms, 8 ohm tap; 2W into simulated speaker load, 4 ohm tap (1dB/vertical div.).

Fig.2 Wavelength Cardinal XS, small-signal 10kHz squarewave into 8 ohms.

The Wavelength's THD+noise percentage (at 1kHz) is plotted against frequency in fig.3. While this performance is not particularly impressive, this is to be expected from this type of amplifier design. It is also clear from this graph that the Wavelength, as configured, cannot be recommended for use into 2 ohm loads, producing massive amounts of distortion. [Note, however, that the high impedance peaks of the simulated speaker load result in much lower distortion.—Ed.] The waveform of the distortion when the Cardinal is driving 1kHz into a 4 ohm load from its 4 ohm transformer tap is shown in fig.4. It is very heavily second-harmonic, with only a trace of irregularity indicating higher-order components at a very low level. The distortion components into 8 ohms and 2 ohms, not shown, indicate an almost pure second harmonic into 8 ohms and a noticeable—but still relatively small—increase in higher-order components into 2 ohms.

Fig.3 Wavelength Cardinal XS, THD (%) vs frequency at 2.83V into (from bottom to top): simulated speaker, 8, 4, and 2 ohms.

Fig.4 Wavelength Cardinal XS, 1kHz waveform at 2W into 4 ohms (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.5 shows the spectrum of the Cardinal XS's output in response to a 50Hz input, taken at a 3.3W output into a 4 ohm load. The distortion here is fairly high (especially considering the wattage involved), but low-order artifacts predominate: -32dB (about 2.5%) at 100Hz and -44dB (0.6%) at 150Hz. [Note the decreasing level of the harmonic with increasing order, a musically consonant phenomenon.—Ed.] A similar measurement made into a simulated real load, not shown, indicates the same trend and marginally lower distortion.

Fig.5 Wavelength Cardinal XS, spectrum of 50Hz sinewave, DC-1kHz, at 3.3W into 4 ohms, 4 ohm tap (linear frequency scale). Note that the second harmonic at 100Hz is the highest in level, at -32dB (about 2.5%).

Feeding a combined 19kHz+20kHz signal into the Cardinal XS results in the spectrum shown in fig.6. This was at a level of 3.3W into 4 ohms (somewhat lower distortion was obtained into 8 ohms, not shown). This is a mediocre result: -32dB (2.5%) at 1kHz and -38dB (1.2%) at 18kHz and 21kHz.

Fig.6 Wavelength Cardinal XS, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 3.3W into 4 ohms, 4 ohm tap (linear frequency scale).

The THD+noise vs output power measurements in fig.7 show the very low power output obtained from this amplifier, typical of single-ended tube designs. Here I also measured the result using the Golden Dragon 300B output tube; note that there is slightly lower distortion with the latter, but a similar maximum output. The actual clipping levels are shown in Table 1. Note that we present the values here at the 3% THD+noise level, instead of the usual 1%.

Fig.7 Wavelength Cardinal XS, distortion (%) vs output power into (from bottom to top at 10W): 4 ohms (4 ohm tap) and 8 ohms (8 ohm tap), both with Golden Dragon 300B output tube; 4 ohms (4 ohm tap), 8 ohms (8 ohm tap), and 2 ohms (4 ohm tap), all with VAIC VV30B output tube.

Since the Wavelength is specified to produce 8W maximum power, I measured the distortion at this output (1kHz). Distortion measured 8.7% with the VV30B output tube, 9.9% with the Golden Dragon 300B. Viewed on a 'scope, however, the clipping was quite soft; while the sinewave was squashed a bit, it was not chopped off. This amplifier is unlikely to sound harsh even if pushed into relatively high levels of distortion. I also noted that the clipping was slightly asymmetrical, perhaps indicative of slightly misadjusted bias (there are no user-accessible bias adjustments on the amplifier).

The test bench results for the Wavelength Cardinal XS could not be rated as anything but poor to mediocre for a new amplifier, though it might have been a formidable design in 1940. Still, given the chosen single-ended configuration, the measured results are about what you would expect. Measurements, however, do not seem to be the point with this sort of design; its passionate advocates consider them irrelevant. Nevertheless, can these sorts of static measurements be defined as "high fidelity"? Can such levels of distortion at low power—despite the known "musical" nature of low-order THD—be acceptable in a product costing this much? I leave that to each reader to decide.

As I've stated before about such low-power amplifiers, you should go into such a purchase with your eyes open. It's true that low-power amplifiers can sound more powerful than their ratings might indicate, usually because they clip gracefully—as does the Cardinal XS. But there's a limit to this, given anything but the most sensitive loudspeakers. And you should definitely—because of the high output impedance and low power output and gain of the Cardinal—consider the entire proposed system when choosing such an amplifier. System matching is always important; with this amplifier, it's critical.—Thomas J. Norton

Table 1

Wavelength Cardinal XS Clipping
(3% THD+noise at 1kHz)
VV30B Output
W (dBW)
300B Output
W (dBW)
83.9 (5.9)4.3 (6.25)
45.1 (4.0)5.3 (4.28)
21.5 (-4.3)
Wavelength Audio
3703 Petoskey Avenue
Cincinnati, OH 45227
(513) 271-4186