Balanced Audio Technology VK-50SE line preamplifier Measurements
The VK-50SE didn't invert signal polarity unless its polarity button was pressed, meaning that the XLR jacks are wired with pin 2 hot. Into a 100k ohm load, the maximum voltage gain was a little higher than specified at 18.55dB, which is inconsequential. The volume-control setting for unity gain was "106." The tape outputs didn't appear to be actively buffered, just isolated with series resistors, so tape recorders should be left on or disconnected when not in use.
The BAT's input impedance measured a high 220k ohms, and its output impedance at 1kHz was 414 ohms from both sets of output jacks. This figure rose slightly to 422 ohms at 20kHz, which is a negligible change. At 20Hz, however, the source impedance was a very high 4.6k ohms, which suggests undersized output coupling capacitors, even with the Six-Pak option, which I assume was fitted to the review sample. Care should be taken when mating the VK-500SE with power amplifiers that have an input impedance of less than 10k ohms. The voltage-divider action of its high source impedance at low frequencies and the power amp's input impedance will roll off the low bass. Into the Linn Klimax 500, for example, which has a low load impedance of 7.7k ohms, the BAT's response at 20Hz will be down 1.8dB compared with the level at 1kHz. That Jonathan still found the combination to sound fine is presumably due to this being outweighed by other, positive aspects of the BAT's sound.
This rolloff at low frequencies can be seen in fig.1, where the lower pair of traces was made into a 600 ohm load, which is typical of some professional amplifiers and is a switchable option on the Jeff Rowland designs. Into a more typical 100k load (top traces), the LF response is flat down to the 10Hz lower limit of the graph. At the other end of the audio band, the output is -0.3dB at 20kHz, with slightly more audio-band rolloff into the low load. Channel separation was better than 80dB below the mid-treble, but with a rise in crosstalk to -67dB (L-R) and -76dB (R-L), due to capacitive coupling between channels. This will be subjectively inconsequential.
Fig.1 Balanced Audio Technology VK-50SE, frequency response at 1V and unity gain into 100k ohms (top below 1kHz) and 600 ohms (0.5dB/vertical div., right channel dashed).
The preamplifier's active circuitry is very linear. Fig.2 shows the spectrum of the output while the VK-50SE drives a 100Hz tone at 1V into 100k ohms. The second harmonic is the highest in level, but this is still at a very low -96dB (0.0015%). Even when the load was dropped to the punishing 600 ohms, the second harmonic rose to only -93dB, with the third harmonic now making an appearance at -100dB. This excellent linearity can also be seen in fig.3, which plots the THD+noise percentage into both 100k ohms and 600 ohms. Other than the equalization effect of its high source impedance at low frequencies, this preamp is not bothered by having to deliver output current.
Fig.2 Balanced Audio Technology VK-50SE, spectrum of 100Hz sinewave, DC-1kHz, at 1V into 100k ohms (linear frequency scale).
Fig.3 Balanced Audio Technology VK-50SE, THD+noise (%) vs frequency at 1V into 100k ohms (bottom traces) and 600 ohms (top traces) (right channel dashed).
This is apparent in fig.4, which plots the BAT's THD+noise percentage against the output voltage, again into 100k ohms and 600 ohms. Into the higher load, an astonishing 40V RMS is available at 1kHz at the 1% THD clipping point, which is way higher than the VK-50SE will ever be asked to deliver into a power amplifier. This means that, at the more usual output level of between 1V and 3V, the preamplifier is operating at its optimal linearity. And even into the atypical 600 ohm load, the preamp doesn't actually clip until it is delivering nearly 14V—equivalent to an output current of 23mA!—John Atkinson
Fig.4 Balanced Audio Technology VK-50SE, distortion (%) vs output voltage into 100k ohms (bottom, at 3V) and 600 ohms (top).