EQUIPMENT PROFILE
EDWARD J. FOSTER

Authorized RePrint of Audio Magazine

BRYSTON 4B ST AMPLIFIER

It's been almost two years since a Bryston product crossed my bench. That one (the BP 20 preamp, Audio, July 1994) was such a pleasure that I looked forward to getting my hands on the 4B ST power amplifier. Bryston's "ST" line of power amps strikes me as quite sensible. The 4B ST is a bridgeable stereo amp delivering 250 watts per channel into 8 ohms in stereo or 800 watts, bridged. The 3B ST is a similar amp, with an 8-ohm stereo rating of 120 watts per channel. The 7B ST has two amp modules but is designed for monophonic use; the two modules can be "bridged" to double the output voltage into relatively high-impedance loads (3 to 8 ohms) or operated in parallel to double the current into relatively low-impedance loads (1 to 3 ohms). Interesting concept. The 8B ST is a four-channel amp that is otherwise similar to the 3B ST; it can be configured to drive four, three, or two speakers.

Technically, Bryston seems to have done its homework. The 4B ST's topology is classic, with each amplifier module built around eight custom output devices powered by separate positive and negative 85 volt supplies. There are two large toroidal power transformers, one for each channel.

Both unbalanced and balanced inputs are provided, the former via gold-plated RCA phono jacks and the latter through gold-plated connectors that can accept both XLR and 1/4-inch stereo phone plugs. All inputs sport fully discrete active buffers. A slide switch centered between the jack pairs selects balanced or unbalanced input. The output terminals are gold-plated multiway binding posts set on standard 3/4-inch centers and outfitted with wire holes large enough for heavy-gauge wire. Another slide switch, between the RCA jacks and the output terminals, selects bridged or stereo operation. In bridged mode, either channel 1 input is used, and the load is connected between the red output terminals. (The wiring details are clearly indicated on the back panel.) Between the output terminals and the combination IEC line-cord jack and fuse holder is a toggle switch for lifting the signal-ground connections to the chassis. Lifting the grounds breaks the hum-pickup loop that might occur through a common power ground when you're using multiple amplifiers, yet it still permits the chassis to be grounded for safety. Good thinking!

The front panel is straightforward: solid handles near each end, a centrally located power button, and two multicolor LEDs above the power switch. The LEDs indicate normal operation (green), the approach of clipping in each channel (yellow), and clipping (red). They also flash red momentarily on power-up and may glow red if signal is present as the supply voltages collapse when powering down; neither condition is cause for concern.

The 1/4-inch black-anodized aluminum panel is drilled for rack mounting. When not rack-mounted, the 4B-ST sits on feet that provide a bit more than 1/2 inch of clearance above the supporting shelf. Black anodized heat sinks along each side provide adequate cooling without a noisy fan. The chassis is designed so that you can temporarily rest this hefty amp on its backside without having to disconnect the wiring. Nice thought!

Measurements

With continuous signals, the Bryston 4B ST ran warm (but not dangerously hot) during my bench tests but remained a good bit cooler in normal operation. Perhaps the heat was due to the high supply voltages. Theoretically, an amplifier whose power supply delivers +85 and -85 volts could deliver 450 watts into 8-ohm loads, almost 2.6 dB above Bryston's 250-watt rating. What the 4B ST actually delivered on the bench was 300 watts (24.8 dBW) of continuous power into 8 ohms at clipping and 0.1 dB more (310 watts) with the IHF dynamic-headroom tone burst-more than it's rated for but still less than theory would suggest. That's not a complaint; the data simply suggests that Bryston uses a rather stiff power supply, which many people consider to be an advantage.
I measured total harmonic distortion plus noise (THD + N) as a function of output for stereo operation with 8-ohm loads, stereo operation with 4-ohm loads, and bridged mono operation with an 8-ohm load (Fig. 1). The data was taken at 20 Hz, 1 kHz, and 20 kHz. I used the unbalanced inputs for these measurements, since that is how most people will use the 4B ST. With 8-ohm loads, both channels driven, I could keep my AC line at the standard 120 volts. With 4-ohm loads, I was unable to hold the line at 120 volts when both channels were driven to rated output. I therefore made the 4-ohm stereo test with only one channel driven, justified by the fact that the 4B ST has separate supplies for each channel. Bridged operation into 8 ohms, while technically similar to 4-ohm stereo operation, makes it impossible to drive only one channel at a time. Thus, as the amplifier approached clipping in this mode, the line dropped to approximately 110 volts.

The results for dynamic output power and headroom listed in "Measured Data" were obtained with both channels driven and with my meter reading an average of 120 volts across the AC line (the meter fluctuates a bit while an amplifier is being pulsed). Since the 4B ST was being driven to full power only part of the time in this test, I could maintain 120 volts at lower output settings of my variable-voltage transformer than I had been able to use in the sine-wave tests. Therefore, I also took data using the same transformer settings that had been needed to maintain 120 volts at the sine-wave clipping point (or the maximum I could manage when 120 volts couldn't be maintained). At these settings, my line meter naturally read more than 120 volts, and the numbers came out 0.5 to 1 dB better than those listed in "Measured Data."
Measurements of THD + N versus frequency (Fig. 2) were made at output levels of 1 watt, 10 watts, 100 watts, and rated power. The last, of course, varied with conditions: 250 watts per channel in stereo with 8-ohm loads, 400 watts per channel in stereo with 4-ohm loads, and 800 watts for bridged operation into 8 ohms. Driving the amplifier in bridged mode to 800 watts with an 8-ohm load dropped the AC line well below the 120-volt standard. I therefore repeated the measurement, this time at 700 watts; calculations showed that the amplifier's distortion at this power level with the available line voltage would be the same as its distortion at 800 watts with a true 120-volt line. The worst-case THD + N is listed in "Measured Data" for each operating configuration at 10 watts and at rated output. I have also reported the bridged 700-watt data, since it is probably more typical of what the distortion at rated power would have been had I been able to maintain a 120-volt line.

If you compare these distortion figures with those of many similarly rated amplifiers, you'll see that Bryston has placed prime emphasis on performance. Even the worst-case THD + N is vanishingly low under all conditions, and the 1-watt curves in Fig. 2 (which stand above the others over most of the frequency range) reflect residual noise rather than distortion.
The noise is unusually low for an amplifier that combines such high power with reasonably high gain. On an A-weighted basis, worst-case output noise was a minuscule -93.6 dBW when the unbalanced inputs were used and negligibly worse with the balanced input. When referenced to rated output power, A-weighted signal-to-noise ratios ran from 116.3 dB (balanced input, stereo) to a whopping 123.5 dB (unbalanced input, bridged).

Figure 3 shows third-octave noise power (in dBW) versus frequency for the worstcase operating mode, 8-ohm stereo operation with balanced input. Obviously, the 4B ST is outfitted with excellent power-supply filters: Note the relative absence of components at 120 Hz and its harmonics. The components that do appear are based on 60 Hz and its odd harmonics, suggesting that the components are ~ caused by minute amounts of magnetic flux leakage from the power transformers (even toroids aren't perfect) or from the power-line wiring. In any event, since the very worst noise level (channel 1 at 60 Hz) is a mere -94.3 dBW, hum should certainly be inaudible. Because my setup in the lab is designed to avoid ground loops, it made little difference whether I made this measurement with the signal ground connected to the chassis or lifted from it. In normal use, you may well find the ability to separate the signal and chassis grounds quite beneficial. Figure 4 shows frequency response with 8-ohm loads. In stereo mode, only the channel 1 data is shown, since the results for channel 2 were identical; channel balance was as near perfect as it is possible to document. It's clear that the Bryston 4B ST is a wideband design: Its high-frequency -3 dB points approach 200 kHz in stereo and 130 kHz in bridged operation, where the rolloffs of both channels combine. At the low end, the -3 dB point is far below 10 Hz, the limit of the measurement; in the most important range, 20 Hz to 20 kHz, response is dead fiat.

Channel separation was a perfectly adequate 70 dB, in both directions, from 20 Hz to 1 kHz; it was 65 dB or better at 4 kHz and better than 60 dB at 8 kHz. Input/output phase error was less than 5-degrees at 10 kHz and less than 10-degrees at 20 kHz. Input impedance was relatively high for a power amp, almost 50 kilohms at the unbalanced input; this should present no problems for any preamp. Damping factor was a solid 600 at 50 Hz, and output impedance remained relatively low across the frequency band (180 milliohms at 20 kHz).

Use and Listening Tests

With only one 4B ST at my disposal, I confined my auditioning to the stereo mode, using the unbalanced inputs. My program sources were a variety of CDs and a few DATs that I had recorded live. Again and again, I heard an effortless quality at all listening levels and a clean, tight bass (especially noticeable in piano recordings) coupled with a bright high end that had nary a trace of frizziness. This is definitely a golden-ear amplifier at a silver-ear price. It's so solidly built, it carries a 20-year warranty! It's technically impressive, sonically superb, and a top value. I find it easy to recommend the Bryston 4B ST with enthusiasm and would love to keep it hard at work in my reference system. Pair it with a BP 20 preamp, and you'll have a system to drool over.

We invite you to experience the Bryston SST² Series amplifiers

20 Year Warranty - A Generation of Music