Features

Next, the more demanding, wide-range fff fragment of choir and orchestra: I had to reduce the gain of the mike amp and readjust the 'scope to 2V/div.; there is therefore no direct level relationship between figs.1-3 and figs.4-6. This is perhaps regrettable in hindsight; however, I was not trying to record absolute SPL levels, just making comparisons among the three amplifiers using two different musical signals.

Fig.4 is the 25W solid-state amp again, now generating 7.6V p-p at the mike amp output. The 10W 300B easily beats it at 11.3V p-p (fig.5), and even the 4W EL84 amp excites the microphone to just beyond the level produced with the 25W solid-state amp, at 8.1V p-p (fig.6).

Fig.4 25W transistor amplifier plus Audio Note E speakers, fff chorus and orchestra from CD 1, track 2 of La Traviata.

Fig.5 10W 300B single-ended amplifier plus Audio Note E speakers, fff chorus and orchestra from CD 1, track 2 of La Traviata.

Fig.6 4W EL84/triode push-pull amplifier plus Audio Note E speakers, fff chorus and orchestra from CD 1, track 2 of La Traviata.

Some might argue that these measurements are based on instantaneous peak levels, that the ear will apply some form of integration, and that the integration time will probably even vary with frequency. And then there are the Fletcher-Munson curves. To evaluate all that and implement it in a measurement setup would probably require the effort needed to complete a thesis, so I decided not to go into it.

The closest I could do was to place my Tandy (RadioShack) sound-level meter next to the measuring microphone. I played the fff part again and noted 100dB SPL (C-weighted, fast response) for the 300B, and 97-98dB SPL for the 25W solid-state and the 4W EL84. This seemed to correspond reasonably well with the peak-type measurements. Perhaps an expert in perception could comment as to which of the two measurements, peak-peak or averaged, is the closer approach to what the ear registers as loudness.

One way or the other, my earlier findings appear to hold true when the tambourine stroke is substituted by more difficult stimuli, such as a soprano voice or choir and orchestra playing and singing at fff. The evidence is somewhat less obvious than in the former case, perhaps, but still quite convincing. Tubes are louder.—Peter van Willenswaard