Conrad-Johnson PV10 preamplifier Measurements

Sidebar 2: Measurements

The output impedance of the PV10A at its line output measured about 690 ohms, give or take less than 10 ohms depending on the channel, the setting of the level control, and whether the balance control was in or out of the circuit. The line-level input impedance varied between 39k ohms and 49.5k ohms, again depending on the above conditions. Phono input impedance measured 48k ohms (L) and 52k ohms (R). The output impedance at the tape output was 495 ohms with a 25 ohm source impedance, 1051 ohms with a 600 ohm source impedance; there is apparently a fixed resistance in the tape output circuitry, but no active tape-out buffering.

The DC offset at the PV10A's outputs measured a low 1.6 to 1.7mV on both channels. The preamp inverts from its inputs to its main outputs (the phono stage by itself, measured at the tape outputs, does not invert). Line-stage gain (aux input to line output) measured 16dB with the balance control in, 17.1dB with it switched out. The gain of the phono stage measured 48.5dB, higher than usual for a moving-magnet input.

The frequency response of the PV10A is shown in fig.1. This measurement was taken at the aux input with the gain control set at maximum. At lower settings, the response was virtually unchanged; it was also the same with the balance control in or out. The tracking of the level control was excellent, at its worst (3dB down from maximum level) varying no more than 0.27dB channel to channel (measured with the balance control out of circuit). The RIAA response of the phono stage is also shown in fig.1; while it is within ±0.25dB across the frequency band (and even better in the right channel), it suggests a very slight warmth to the sound.

Fig.1 Conrad-Johnson PV10A, line input frequency response (top) and phono input RIAA error (bottom) (right channel dashed, 0.5dB/vertical div.).

The line-stage crosstalk (balance in) of the PV10A is shown in fig.2 (full gain). While less good than that typically found in a solid-state preamplifier, the channel separation is still acceptable and very consistent in both directions (left to right and right to left). The result is slightly better with the balance control out (lower curve, right-to-left crosstalk shown). In both cases the typical high-frequency decrease in separation is likely due to capacitive coupling between channels. Via the phono inputs, there was some asymmetry in the crosstalk measurements (fig.3, taken at 15mV input at 1kHz). Though the separation in one direction is notably better than in the other, the crosstalk on the worst channel is only a few dB lower than that from a high-quality MC phono cartridge. There should be no audible imaging problems, however.

Fig.2 Conrad-Johnson PV10A, line input crosstalk (from top to bottom): L-R, R-L (Balance control in circuit); R-L (Balance control bypassed) (5dB/vertical div.).

Fig.3 Conrad-Johnson PV10A, phono input crosstalk at 15mV input at 1kHz (from top to bottom): L-R, R-L (5dB/vertical div.).

Distortion through the '10A's line stage (the lower pair of traces in fig.4, taken with 100mV input, full gain, balance control in) is very low across the audio band, and was unchanged with the balance control switched in—after compensating for the slight resulting change in gain (not shown). The PV10A would accept a signal level of 2.47V RMS at its aux input before reaching 1% THD+noise. This gave an output of 17.8V, more than enough to drive a typical power amplifier with a sensitivity of 1V well into clipping. (Turning down the gain control, increases the overload margin and decreases the output voltage, of course.)

Fig.4 Conrad-Johnson PV10A, THD+noise (%) vs frequency for line input (top, 100mV input, volume control max, Balance control in) and phono input (bottom, 15mV input at 1kHz) (right channel dashed).

The phono stage's distortion is shown by the upper pair of traces in fig.4, taken with a 15mV input at 1kHz. (This input level gave the lowest THD+noise reading; at lower input levels, noise predominated.) The result is low across most of the range, with a slight increase above 10kHz.

As is often the case with tube circuitry, the PV10A's phono overload margins were good: 1% THD+noise was reached at an input of 91.5mV at 1kHz, 731mV at 20kHz, and 6.4mV at 20Hz. The differences are due to the boost and cut characteristics of the RIAA curve and equate to overload margins of 25.2dB, 23.3dB, and 22.1dB, respectively, referenced to 5mV input at 1kHz.

There were no surprises in the measurements of the PV10A; its measured performance is solid across the board.—Thomas J. Norton

2733 Merrilee Dr.
Fairfax, VA 22031
(703) 698-8581