Krell KRC-HR preamplifier & Audio Standard power amplifier Measurements

Sidebar 3: Measurements

Unless otherwise noted, the measurements presented are for balanced operation. Unless otherwise stated, the line-level gain switch was set at normal.

Line-stage voltage gain measured a sensible 11dB, balanced, and 4.9dB, unbalanced (16.9dB and 10.8dB, respectively, with the high gain switch engaged). The output impedance of the Krell KRC-HR measured an extremely low 3.2 ohms balanced, 1.7 ohms unbalanced in the left channel (3.5 ohms and 2.0 ohms, respectively, in the right). The input impedance measured 101.6kohms, left and 97.4kohms, right, balanced (47.8kohms and 48kohms, unbalanced). Both input and output impedances were little affected by the level control setting. The output impedance at the tape output was just under 45 ohms regardless of source impedance, indicating full active buffering of the tape outputs. The KRC-HR's polarity is pin 2 positive in the balanced mode, and non-inverting in the unbalanced mode.

The DC offset at the Krell's balanced outputs was suitably low at a maximum of 0.7mV across either leg of the balanced pair. Unbalanced DC offset measured 1.6mV in either channel. S/N measured 84.9dB (unweighted) over a bandwidth of 22Hz-22kHz, 75.9dB (unweighted) from 10Hz-500kHz, and 87.4dB, A-weighted (all ref. 1V output) for the line stage (the corresponding unbalanced readings were somewhat better at 91dB, 80.8dB, and 93.7dB).

The frequency response of the Krell is shown in fig.1. There was no difference between the balanced result (shown) and the unbalanced (not shown). The low level of crosstalk (better than -90dB below 2kHz, not shown) is unlikely to affect the audible performance of the preamp in any way.

Fig.1 Krell KRC-HR, frequency response at 1V output into 100k ohms (right channel dashed, 0.5dB/vertical div.).

Regarding the very low levels of THD+noise (fig.2), it is interesting that the distortion through the unbalanced line input is lower than the balanced, though the difference is trivial and may be caused by residual levels of noise. This conjecture was confirmed by fig.3, which shows the THD+noise plotted against output voltage at 1kHz. (As with our other preamp tests, the minimum points in the curves, just before the distortion increases rapidly, were the values chosen to run the crosstalk and distortion vs frequency measurements.) The output spectrum of the Krell reproducing a 50Hz input at a very high output level of 10V is shown in fig.4. The highest level artifact—the third harmonic at 150Hz—lies at about -60dB (0.1%). Reducing the output level to 5V dropped the level of this harmonic to -74dB (0.02%).

Fig.2 Krell KRC-HR, THD+noise vs frequency at 2.5V into 100k ohms, balanced (bottom), and at 1.25V into 100k ohms, balanced (top, right channel dashed).

Fig.3 Krell KRC-HR, distortion (%) vs output voltage into 100k ohms (unbalanced bottom at 1V).

Fig.4 Krell KRC-HR, spectrum of 50Hz sinewave, DC-1kHz, at an output level of 10V (linear frequency scale). Note that the third harmonic at 150Hz is the highest in level, at -60dB (about 0.1%).

The test bench performance of the Krell KRC-HR was very good to excellent in most respects.—Thomas J. Norton

45 Connair Rd.
P.O. Box 533
Orange, CT 06477-0533
(203) 799-9954