darTZeel CTH-8550 integrated amplifier Measurements
I examined the darTZeel CTH-8550's measured behavior using mainly Stereophile's loaner sample of the top-of-the-line Audio Precision SYS2722 system (see the January 2008 "As We See It" and www.ap.com); for some tests, I also used my vintage Audio Precision System One Dual Domain.
Before performing any tests on an amplifier, I run it for 60 minutes at one-third its specified power into 8 ohms, which is thermally the worst case for an amplifier with a class-B or -AB output stage. Cold out of the box, the darTZeel offered 0.083% distortion at one-third power into 8 ohms. After 45 minutes of running with both channels driven at this level, the THD+noise had increased slightly, to 0.093%. More importantly, I cut short the preconditioning at that point, as the top panel was too hot to touch and the side-panels covering the heatsinks were so hot that I was concerned about the amplifier going into thermal runaway. Even with no signal, the CTH-8550's chassis gets very warm after a whilecontrary to the manufacturer's claim, this is definitely not a cool-running amplifier.
The MM phono input of our review sample"Built for T.H.E. Show," according to the front-panel displaywasn't activated, but it did have an MC phono stage. As supplied, this offered 54.7dB of voltage gain at 1kHz, measured at the Rec Out jacks, and the input impedance was to specification, at 1000 ohms across the audioband. The RIAA error was commendably low in the audioband, with superb channel matching, but increased with increasing frequency above 10kHz, due to the manufacturer including the so-called "Neumann 4th Pole" (fig.1, footnote 1). The darTZeel's MC input was quiet, with the wideband, unweighted S/N ratio measuring 51.6dB ref. an input of 500µV at 1kHz. It improved to 69dB when A-weighted. The phono stage's distortion was very low, at typically 0.025%, and the overload margin at low and middle frequencies was superb, at 30dB. The margin dropped to just 5.2dB at 20kHz, however, which is disappointing.
Fig.1 darTZeel CTH-8550, MC input RIAA error at 1mV input at 1kHz (1dB/vertical div., right channel dashed)
Turning to the line stage, I didn't examine the behavior of the proprietary Zeel and darT inputs and outputs. The conventional unbalanced jacks had an input impedance of 42k ohms at low and middle frequencies, dropping to a still-high 30k ohms at 20kHz. The balanced XLR jacks had an input impedance of 19k ohms across the audioband. All the inputs preserved absolute polarity, ie, were non-inverting, the XLRs being wired with pin 2 hot.
Measured at the speaker terminals, the maximum line-input voltage gain into 8 ohms, with the volume control set to "+12," was 38.4dB for unbalanced inputs and, unusually, 6dB lower at 32.4dB for the balanced input. The volume control operated in accurate 0.5dB steps; at "0," the gain at the Preamp Out jacks was 0.27dB and at "+12," it was 12.27dB. These jacks offered a usefully low output impedance of 80 ohms and again preserved absolute polarity.
The CTH-8550's output impedance was quite high for a solid-state design, at 0.4 ohms in the bass, rising to 0.5 ohms at 20kHz. As a result, the modification of its frequency response by the usual Ohm's Law interaction between this impedance and the load impedance of our simulated loudspeaker reached ±0.25dB (fig.2, gray trace), which will be audible. The small-signal response into 8 ohms (fig.2, blue and red traces) was flat within the audioband and didn't reach 3dB until 125kHz. This wide bandwidth correlated with a well-shaped 10kHz squarewave (fig.3), and didn't vary significantly with the volume control setting. Into increasingly low load impedances, however, the upper 3dB frequency progressively dropped and the beginning of an infrasonic rise in response became evident (fig.2, green trace).
Fig.2 darTZeel CTH-8550, frequency response at 2.83V into: simulated loudspeaker load (gray), 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (green). (1dB/vertical div.)
Fig.3 darTZeel CTH-8550, small-signal 10kHz squarewave into 8 ohms.
The CTH-8550's channel separation was good rather than great, at 78dB in both directions below 1kHz, reducing to 65dB at 20kHz. This is rather less than the specified 90dB, however. The background noise level, measured with an unbalanced input short-circuited and the volume control at its maximum setting was 63.6dB ref. 2.83V into 8 ohms (wideband, unweighted measurement). The A-weighted S/N ratio was 87.2dBA, ref. the same level, which is a little lower than the manufacturer's specification of >115dBA ref. full power.
The darTZeel comfortably exceeded its maximum power specification. This can be seen in fig.4, which plots the THD+noise percentage against output power into 8 and 4 ohms with both channels driven, and into 2 ohms with one channel driven. Defining clipping as 1% THD+N, the CTH-8550 clipped at 230W into 8 ohms (23.6dBW), 375W into 4 ohms (22.7dBW), and 450W into 2 ohms (20.5dBW). The distortion starts to rise above the noisefloor at levels of a few hundred milliwatts, but was not particularly low. The traces in fig.5 were taken at a constant 10V into 8, 4, and 2 ohms, and only really show acceptably low distortion into the higher impedances. Ameliorating this behavior, to some extent at least, will be the fact that the THD+N percentage doesn't increase with frequency.
Fig.4 darTZeel CTH-8550, distortion (%)vs 1kHz continuous output power into (from bottom to top): 8, 4, 2 ohms.
Fig.5 darTZeel CTH-8550, THD+N (%)vs frequency at 10V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta), 2 ohms (green).
The bottom trace in fig.6 shows the waveform of the spuriae; the peaks in that waveform occur suspiciously close to the zero crossing points of the sinewave signal (upper trace), which suggest the presence of crossover distortion. Given how hot the amplifier runs, which suggests a high output bias current for the output stage, this surprised me. FFT-derived spectral analysis of the amplifier's output at low levels indicates that the dominant harmonics present are the second and third harmonics (fig.7), which are relatively innocuous, subjectively speaking. At higher powers, however, a regular series of higher-order harmonics makes an appearance (fig.8) and the amplifier's performance on the demanding high-frequency intermodulation test is disappointing (fig.9).
Fig.6 darTZeel CTH-8550, 1kHz waveform at 20.4W into 4 ohms (top), 0.206% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.7 darTZeel CTH-8550, spectrum of 1kHz sinewave, DC10kHz, at 1W into 8 ohms (left channel blue, right red; linear frequency scale).
Fig.8 darTZeel CTH-8550, spectrum of 1kHz sinewave, DC10kHz, at 96W into 8 ohms (left channel blue, right red; linear frequency scale).
Fig.9 darTZeel CTH-8550, HF intermodulation spectrum, DC24kHz, 19+20kHz at 135W peak into 4 ohms (linear frequency scale).
"Disappointing" is the word I would use overall to describe the darTZeel's measurements. When I measured the first product from this Swiss company, the NHB-108 amplifier that Wes Phillips reviewed in April 2005, which had a very similar measured performance, I wrote that "The NHB-108's measured performance indicates that, with the exception of its slightly bent transfer function, designer Hervé Delétraz has avoided the problem most commonly associated with the absence of the usual overall negative-feedback loop." But like the earlier design, the CTH-8550 hasn't managed to reduce the measured aberrations to the point where I could confidently describe them as inaudible under all circumstances. Which is a shame, given the amplifier's beautiful appearance and versatile user interface.John Atkinson
Footnote 1: See Keith Howard's article on the RIAA equalization in general, and this subject in particular, in the March 2009 issue.