Preamplifiers
Pass Labs XP-20 Stereo Preamplifier
- Written by John E. Johnson, Jr.
- Published on 01 August 2011
On the Bench
Distortion measurements were taken using the XLR inputs and outputs, and were made within an 80 kHz bandwidth.
A 1 kHz sine wave test signal resulted in only 0.001% THD+N at 2 volts output.
Using a combination of 19 kHz and 20 kHz sine waves, the B-A peak at 1 kHz was 118 dB below the fundamentals.
There was only 0.0009% IMD at 2 volts output, which is the lowest IMD we have ever measured on a preamplifier. This was responsible for the incredible amount of detail that I heard in the recordings.
IMD is an insidious infection of the music. It produces more damage than simple harmonic distortion. Here's why. Let's say you have a trumpet playing a single note at middle C and an oboe playing an E at the same time. Each instrument will have a fundamental frequency and harmonics that create the sound of the instrument. In the audio system, each of those fundamentals and harmonics produced by the instrument will have harmonics produced by the preamplifier, amplifier, and speakers. Those harmonic distortion peaks could be counted if you like. But, with IMD, what happens is all the harmonic distortion peaks interact with the other distortion peaks, creating additional distortion peaks that represent the sum of two peaks, and the difference between the two peaks (the lower one subtracted from the larger one). And, those interaction peaks, then interact again, with each other, and with other harmonic peaks. So, theoretically, you can have an infinite number of IMD peaks. The results? Mushy sound where instruments seem to be blending with one another rather than remaining distinct. The IMD measurement in the Pass XP-20 was a revelation for me, as I have heard preamps and power amps with 0.001% THD+N, but never a preamp or power amp with an IMD less than 0.001%. I conclude a direct correlation with the stunning detail I heard in the recordings with that low IMD.
THD+N was less than 0.002% all the way out to 50 kHz. Superb!
With a 100 kOhm load, the sharp knee (the practical output limit) was at 4 volts RMS. Output at clipping (1% THD+N) was at 34 volts.
The frequency response was down 3 dB at 60 kHz as per the specification. In the audible band, the frequency response was 20 Hz - 20 kHz, - 0.2 dB.
