- Written by John E. Johnson, Jr.
- Published on 29 April 2013
On the Bench
All distortion measurements were within an 80 kHz bandwidth. I used XLR connections for the input signal and output. A load of 100 kOhms was used. The preamp did not produce output with a 600 ohm load, so you should use a power amplifier with an input impedance of at least 10 kOhms.
At 1 kHz and 2 volts output, distortion was well below 0.001%.
At 5 volts output, distortion was still astonishing at only 0.0003%.
Using a combination of 19 kHz and 20 kHz test signals, the B-A peak was more than 120 dB below the fundamentals, i.e., inaudible, not to mention barely visible.
And at 5 volts output, the B-A peak was 105 dB below the fundamentals.
IMD at 2 volts output was 0.001%, a great result.
At 5 volts output, IMD measured even lower, but not significantly so.
THD+N vs. Frequency at 2 and 5 volts output produced one of the best set of spectra I have ever seen in a preamplifier. Amazing, at any price.
The measured frequency response was essentially flat out to 20 kHz, where it then began a long slow roll-off, ending at - 2 dB at 200 kHz.
I measured two volume settings, one with the input equal to the output (2 volts) and the other with the input at 2 volts and the output at 5 volts. The results are shown below. The best part of the range is between 100 mV and 10 V. Fifteen volts was the maximum output, regardless of the input and volume control settings.
I tested the RIAA accuracy by measuring the frequency response using a pre-emphasis EQ curve compensation file. For the MM setting ("Moving Magnet"), the adherence to the RIAA spec was within ± 0.2 dB.
For the MC setting ("Moving Coil"), the curve showed a bit more deviation, which is to be expected, since there is an additional gain stage in the path. Nevertheless, the response was within ± 0.5 dB of the RIAA spec.
For those of you who want to use the High-Pass/Low-Pass capability of the XSP-1, here are the curves for settings at 50 Hz and 250 Hz. The curves indicate a second-order (12 dB/octave) Linkwitz-Riley crossover design, which is performed in a passive analog circuit (there is no A/D - D/A conversion involved).