Sidebar 3: Measurements
I measured the Rogers High Fidelity 65V-1 with my Audio Precision SYS2722 system (see the January 2008 "As We See It"). The maximum gain at 1kHz into 8 ohms varied with operating mode and output tube and was mostly lower than specified, ranging from 18.7dB (KT88s, triode mode) and 21.3dB (EL34s, triode) to 23.8dB (KT88s, Ultralinear mode) and 25.85dB (EL34s, Ultralinear)—all lower than the norm for an integrated amplifier. The maximum gain at the headphone jack measured 25dB. Both the speaker and headphone outputs preserved absolute polarity (ie, were non-inverting) in all modes and with both types of output tube. I controlled the 65V-1 with the Bluetooth-connected iPad app. As downloaded, the meter on the iPad app wasn't calibrated: a measured level of 0.33W into 8 ohms was displayed as 25W!
The input impedance was a usefully high 57k ohms at all audio frequencies. The output impedance was very high, and again varied with operating mode and output tube, as well as with frequency. In triode mode with Herb Reichert's preferred EL34 tubes, the output impedance ranged from 3.7 ohms at 20Hz and 20kHz to 4.3 ohms at 1kHz. As a result, the variation in frequency response into our standard simulated loudspeaker was very high, at a very audible ±2.2dB (fig.1, gray trace), and there were large changes in output level as the load impedance changed from 8 to 2 ohms. The impedance was even higher in Ultralinear mode, reaching 6.8 ohms at 1kHz with KT88 tubes and 8.8 ohms with EL34 tubes, giving rise to response variations of ±3.3dB with the latter (fig.2, gray trace). A small peak at 120kHz can be seen in this graph, which is associated with some damped ringing with a 10kHz squarewave (fig.3).
An EL34 tube operated as a single-ended triode is not going to be able to deliver much power, and fig.5 indicates that, with our usual definition of clipping—ie, when the percentage of THD+noise reaches 1%—the Rogers clipped at just 270mW into 8 ohms. Relaxing the definition to 3% allowed the 65V-1 in triode mode with EL34s to deliver 2.275W into 8 ohms and into 4 ohms (fig.6), the amplifier clipped at 1.6W (1% THD+N), 4.15W (3%), and 8W (10%). More power was available in Ultralinear mode, both with EL34s (fig.7) and KT88s (fig.8), where a THD+N of 3% was reached at a respective 5.5W and 6W into 8 ohms. But the Rogers 65V-1 is undoubtedly a low-power, high-distortion design. Worried that our review sample may have been faulty, I checked the measurements that Rogers's Roger Gibboni had included; I am confident that my measurements characterize this unit's behavior.
Its measured performance indicates that the Rogers 65V-1 really should be used only with high-sensitivity loudspeakers, but even then, its high output impedance means that its sonic signature will be different with every speaker, and its distortion signature may well fatten the sound. But props to Roger Gibboni for the useful iPad app.—John Atkinson
Fig.1 Rogers 65V-1, EL34s, triode mode, 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.2 Rogers 65V-1, EL34s, Ultralinear mode, 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 Rogers 65V-1, EL34s, Ultralinear mode, small-signal, 10kHz squarewave into 8 ohms.
Channel separation was modest, at 60dB in both directions at 1kHz and around 45dB at the frequency extremes. I monitored the waveform of the crosstalk on an oscilloscope to ensure that I was measuring actual crosstalk and not noise. The wideband, unweighted signal/noise ratio, taken with the inputs shorted to ground but the volume control set to its maximum—the worst case—was also modest, at 60.4dB in triode mode, 56.6dB in Ultralinear, both ratios ref. 1W into 8 ohms. Switching an A-weighting filter into circuit improved these ratios to 76.6 and 72.4dB, and spectral analysis of the amplifier's low-frequency noise floor (fig.4) revealed spuriae at power-supply–related frequencies, as well as a rise in the random noise floor in the lower midrange and below.
Fig.4 Rogers 65V-1, EL34s, triode mode, spectrum of 1kHz sinewave, DC–1kHz, at 1W into 8 ohms (linear frequency scale).
Fig.5 Rogers 65V-1, EL34s, triode mode, distortion (%) vs 1kHz continuous output power into 8 ohms.
Fig.6 Rogers 65V-1, EL34s, triode mode, distortion (%) vs 1kHz continuous output power into 4 ohms.
Fig.7 Rogers 65V-1, EL34s, Ultralinear mode, distortion (%) vs 1kHz continuous output power into 8 ohms.
Fig.8 Rogers 65V-1, KT88s, Ultralinear mode, distortion (%) vs 1kHz continuous output power into 8 ohms.
Fig.9 shows how the THD+N percentage varies with frequency with EL34 tubes in triode mode at 125mW into 8 ohms (blue and red traces) and into 4 ohms (cyan, magenta). As expected from figs. 5 and 6, the THD+N is lower into 4 ohms than into 8 ohms, but is still high. It was even higher in Ultralinear mode, but fortunately, the distortion is predominantly the subjectively innocuous second harmonic (fig.10). However, some higher-order harmonics were present at low frequencies (fig.11). While high-order intermodulation distortion with an equal mix of 19 and 20kHz tones was fairly low, the second-order difference product at 1kHz reached 1% even in the lowest-distortion condition: triode mode with EL34s into 4 ohms (fig.12).
Fig.9 Rogers 65V-1, EL34s, triode mode, THD+N (%) vs frequency at 1V into: 8 ohms (left channel blue, right red), 4 ohms (left cyan, right magenta).
Fig.10 Rogers 65V-1, EL34s, triode mode, 1kHz waveform at 1V into 8 ohms, 0.75% THD+N (top); distortion and noise waveform with fundamental notched out (bottom, not to scale).
Fig.11 Rogers 65V-1, EL34s, triode mode, spectrum of 50Hz sinewave, DC–1kHz, at 1V into 8 ohms (linear frequency scale).
Fig.12 Rogers 65V-1, EL34s, triode mode, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 1V peak into 4 ohms (linear frequency scale).
