Theta Digital Enterprise monoblock power amplifier Measurements

Sidebar 3: Measurements

I measured the Enterprise with both balanced and unbalanced signals; there was no appreciable difference between the two modes. Neither input inverted signal polarity, and the input impedance at 1kHz was a high 52k ohms unbalanced, 82k ohms balanced. The unbalanced voltage gain into 8 ohms was 25.3dB, a little lower than average. Surprisingly, the balanced gain was even lower at 25dB, 6dB lower than specified. However, it's fair to point out that this will not be a problem for any good preamp, especially Theta's own digital processors, which have very high output levels. (See our review of the Generation VIII in February.) The unweighted, wideband signal/noise ratio (ref. 1W into 8 ohms) was a good 84dB, though this figure fluctuated somewhat due to the presence of some very-low-frequency noise. Switching in an A-weighting filter improved the figure to an excellent 96dB. As MF said, the Enterprise is indeed a "quiet amplifier."

The Theta's output impedance was a little higher than usual, at 0.5 ohm (20Hz and 1kHz) and 0.55 ohm (20kHz), which means that there was slightly more variation of the amplifier's frequency response than is normally seen with a solid-state design—±0.4dB (fig.1, top trace)—and slightly more reduction in gain with each halving of the load impedance (fig.1, bottom two traces). The amplifier's small-signal bandwidth is very wide, however, with a -3dB frequency of 121kHz, which gives a perfect 1kHz squarewave response (fig.2). Commendably for a design with minimal negative feedback, the bandwidth doesn't change with load impedance.

Fig.1 Theta Enterprise, balanced, frequency response at (from top to bottom at 2kHz): 2.83V into simulated loudspeaker load, 1W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms (0.5dB/vertical div.).

Fig.2 Theta Enterprise, small-signal 1kHz squarewave into 8 ohms.

However, I did notice that whenever I changed the operating conditions, the Enterprise took several seconds to settle down to its new steady-state performance. I noted this phenomenon in our review of the Ayre AX-7 amplifier last October (Vol.26 No.10). The amplifier's low amount of loop negative feedback means that its operating parameters are continually changing according to the demands of the music. Admittedly this is to a very small degree, but it is not something I am used to seeing. (The underlying phenomenon still happens in high-feedback designs, but is compensated for to a much greater extent by the action of the feedback.)

Before taking any measurements, I ran the Enterprise for an hour at 100W into 8 ohms. At the end of this period, the chassis was hot, with the black metal-mesh panels inset into the top of the enclosure too hot to touch. During this time, however, the measured THD percentage dropped from 0.2% to 0.13%. Fig.3 shows how the small-signal distortion changes with both frequency and load impedance. It is respectably low, though the amplifier is obviously working hard into 2 ohms. Fig.4 shows that the spectrum of that harmonic content is predominantly third-order, though other harmonics make an appearance at high power levels (fig.5).

Fig.3 Theta Enterprise, balanced, THD+N (%) vs frequency (from bottom to top): 2.83V into 8 ohms, 4 ohms, 2 ohms.

Fig.4 Theta Enterprise, 1kHz waveform at 1W into 8 ohms (top), 0.034% THD+N; distortion and noise waveform with fundamental notched out (bottom, not to scale).

Fig.5 Theta Enterprise, spectrum of 50Hz sinewave, DC-1kHz, at 100W into 8 ohms (linear frequency scale).

The Enterprise's behavior when tested for intermodulation distortion was unusual in that, while the difference component resulting from an equal mix of 19 and 20kHz tones was respectably low (fig.6), higher-order products made an appearance. Yes, this graph was taken close to visual clipping into 4 ohms with this demanding signal, but I was nevertheless a little perturbed to see this behavior.

Fig.6 Theta Enterprise, HF intermodulation spectrum, DC-24kHz, 19+20kHz at 100W into 8 ohms (linear frequency scale).

Finally, fig.7 shows how the measured THD+noise percentage changes with increasing output power into 8, 4, and 2 ohms. In common with some other amplifiers I have tested that feature low loop negative feedback, the Enterprise's circuit actually gets more linear at moderate output currents, before the distortion starts its inevitable rise toward clipping. This graph also shows that the Theta doesn't quite meet its specified output power at the usual 1% THD clipping point: 280W (24.5dBW) instead of 300W (24.8dBW). (I note that the Enterprise's power output appears to be specified at 1.75% THD.) The 0.3dB shortfall will not be audibly significant, but relaxing the clipping point to 3% THD does allow the Enterprise to give out 452W into 8 ohms (26.6dBW) and 580W into 4 ohms (24.6dBW). But only 215W are available into 2 ohms at 1% THD (17.3dBW), meaning that the amplifier is really best used into speakers of 4 ohms and above. However, it did drive 2 ohms at its clipping power without blowing fuses or otherwise misbehaving, which is what you'd expect from a true high-end amplifier.

Fig.7 Theta Enterprise, distortion (%) vs 1kHz continuous output power into (from bottom to top at 50W): 8 ohms, 4 ohms, 2 ohms.

Overall, the Theta Enterprise offers a somewhat mixed bag of measurements.—John Atkinson

Theta Digital
5330 Derry Ave., Suite R
Agoura Hills, CA 91301
(818) 597-9195