Tube Preamp Reviews

I measured the Rogue Audio RP-1's electrical performance with my Audio Precision SYS2722 system (see the January 2008 "As We See It").

The RP-1's phono input appears to use ultra-low-noise LT1115 op-amp chips. I left the internal settings as used by Herb Reichert for his auditioning— high gain and a 300 ohm load —which resulted in a voltage gain of 65.1dB at the main outputs with the volume control set to its maximum, and 55.2dB at the fixed output jacks. This is a little lower than the specified 60dB for moving-coil cartridges, but the shortfall will be due in part to the Ohm's law interaction between the Audio Precision's 20 ohm output impedance and the RP-1's phono-stage input impedance of 298 ohms, this figure consistent across the audioband. I examined the phono stage's performance at the fixed jacks. It preserved absolute polarity and offered relatively high signal/noise ratios: 58dB (unweighted, wideband), 60dB (22Hz–22kHz), and 69.8dB (A-weighted), all ref. 1kHz, 500µV, and measured with the inputs shorted to ground.

The RIAA error is shown in fig.1. While the left channel (blue trace) offers an essentially perfect response, other than the slight rolloff in the very low bass (–3dB at 14Hz), the right channel (red trace) shelves down in the treble by 0.5dB. This is probably due to a slightly out-of-spec passive component. Channel separation (not shown) was excellent, at >70dB below 1.5kHz, R–L, and >80dB 20Hz–20kHz, L–R. The phono-stage overload margin was superb, at >31dB from 20Hz to 20kHz, ref. 1kHz 500µV; harmonic distortion was commensurately low, with just a trace of third harmonic visible, even at 10.5mV input voltage (fig.2). Intermodulation distortion (fig.3) was similarly negligible.

Turning to the line inputs, the volume control operated in 0.95dB steps, with the unity-gain setting between "49" and "50." The gain with the control set to its maximum, "60," was 9.9dB, left channel, 9.6dB, right, measured at the main out jacks, and 12.8dB at the headphone output jack. There was no gain at the fixed output jacks. The line inputs inverted absolute polarity and offered a usefully high input impedance of 64k ohms at low and middle frequencies, dropping inconsequentially to 54k ohms at 20kHz. The output impedance at 20Hz was a fairly high 1600 ohms, dropping to a still high 726 ohms at 1kHz and 712 ohms at 20kHz. As a result, the frequency response into the demanding 600 ohm load (fig.4, cyan and magenta traces) rolled off in the bass, reaching –3dB at 25Hz. Into the high 100k ohms load, however, the low-frequency response was flat down to 10Hz (blue, red traces). At the other end of the spectrum, the ultrasonic response was curtailed, reaching –3dB at 80kHz. Note, also, the 0.3dB imbalance between the channels in this graph. (I made sure the balance control was centered before performing the measurements.)

This graph was taken from the main output jacks with the volume control set to its maximum; there was no appreciable change in the response at lower volume-control settings. To my surprise, however, the response from the headphone jack featured a rolloff in the high treble that reached –5dB at 20kHz (fig.5). But, as HR wrote, the RP-1's headphone output did lack "the richness, sparkle, and dynamic impact of its overachieving phono and line stages." The headphone stage's output impedance measured close to 1 ohm across the audio band.

Channel separation via the line inputs was good rather than great, at 70dB at 1kHz, decreasing to 45dB at the top of the audioband. The unweighted, wideband S/N ratio (ref.1V and measured with the inputs shorted to ground but the volume control set to "60," the worst-case situation) was 75.6dB. This improved slightly, to 80dB, when the measurement bandwidth was restricted to the audioband, and to 84dB when A-weighted. As shown by fig.6, the primary source of noise was spuriae at the AC line frequency of 60Hz and its odd-order harmonics, these most likely due to magnetic interference from the toroidal power transformer.

Fig.7 plots how the percentage of THD+noise changed with output voltage into 100k ohms. The distortion starts to rise from the noise floor above 500mV, but the preamp doesn't actually clip (defined as when the THD+N reaches 1%) until 23V! I then tested the RP-1 into 10k ohms, which is more typical of the load impedance it will see in practice. Although it clipped at a lower level, 19V, the distortion at lower levels was, to my surprise, one-third of the 0.03% into 100k ohms, 0.01% (fig.8). Into 600 ohms, however, the amplifier clipped at just 300mV.

Measured into 100k ohms, the RP-1's percentage of THD+N didn't change with frequency (fig.9). However, into 10k ohms, though the absolute level was lower, as expected from figs. 6 and 7, the THD+N did rise in the treble, reaching 0.09% at 20kHz. The distortion signature into 100k ohms was primarily second-harmonic in nature (fig.10), though this dropped almost to the level of the third harmonic into 10k ohms (fig.11). Intermodulation was also low (fig.12).

Other than the rolled-off highs from its headphone output, the Rogue RP-1 measures well. Its phono stage is simply superb. Just don't use the RP-1 with power amplifiers having input impedances lower than 10k ohms.—John Atkinson