- Written by John E. Johnson, Jr.
- Published on 13 September 2010
On the Bench
Distortion measurements were made within an 80 kHz bandwidth and into a Carver Mark IV ribbon speaker. Power output and distortion vs. frequency were made using 8 or 4 ohm power resistors.
At 1 kHz, THD+N was very low, just like the 14B. Totally within specification. The distortion was primarily 2nd order, and that is something I love to see in distortion spectra. The peaks in the 18 kHz - 21 kHz range are electrical noise in my AC lines. They are not a product of the amplifier and are below audibility. However, they do make an interesting variable to take note of in terms of noise rejection. It shows up in the 3B, but in the 14B, which is fully balanced throughout, the peaks did not show up, because of common mode rejection, which is one of the principal advantages of a balanced circuit.
Using 19 kHz and 20 kHz sine waves as the input signal, the B-A peak at 1 kHz was 96 dB below the fundamentals.
IMD measured only 0.007%. Man, that is low ! ! !
THD+N vs. Frequency showed a flat distortion spectrum out to 5 kHz, then rising to 0.02% - 0.03% from 10 kHz to 50 kHz.
Here is an impedance plot of the Carver speaker that I used as the load.
Power output at 8 Ohms was 150 watts at the sharp knee, and clipped (1% THD+N) at 170 watts output.
At 4 Ohms, the sharp knee was at 200 watts, and clipping occurred at 250 watts.
The measured frequency response was 10 Hz - 100 kHz, - 3 dB at 8 Ohms, and - 5 dB at 4 Ohms.