The 3B-ST Power Amplifier

A company that takes a rather low-key approach to product introductions or updates, Bryston has recently brought its ST series of amplifiers into the market with little fanfare. The ST designation is tribute to the man who refined the basic circuitry of these Chris-Russell designed amps, Stuart Taylor. Having identified a number of potential areas of improvement while developing the multichannel 8B model, Stuart has subsequently incorporated these into the full amplifier line. The one exception is the 2B, which by its modular nature is less amenable to such changes, and as a lower-powered (50 wpc) model with fewer power transistors has been regarded as having less need (nor physical space) for such changes.

According to a new brochure, the "ST Series amplifiers incorporate the use of an innovative gain stage topology which yields a substantially lower noise floor through the use of low- impedance pathways within the amplifier. This same approach also has strong advantages in reducing the overall distortion performance. Bryston's computer controlled distortion analyzers confirm lntermodulation distortion levels less than 10 parts per million (Below 0.0009%) are typical for this new circuitry."

Of course, the 3B ST is a completely dual-mono design, with two large toroidal transformers stacked so that their magnetic fields cancel to minimize noise generation, and "multiple filter capacitors per channel, rather than single pairs of larger filter cans. This allows improved high frequency response and reduces overall losses in the supply."

Other ST changes include a rationalization of input levels, so that the balanced input does not produce 6 dB higher output level; as well, when an amplifier is used in bridged mode, its gain is identical to that when operated in stereo. These changes have been made so that the various models are more easily compatible when used in home theatre systems and professional installations.

Rated at 120 watts per channel, it comes with an individual measurement sheet listing distortion, power and noise figures. Having sets of these for three generations of 3B, I was able to make some interesting comparisons, particularly between the ST and NRB models. These should, of course, be taken in the light of an understanding that steady-state distortion measurements do not tell all, and can be made lower at the expense of increasing dynamic distortions which are very difficult to quantify.

That said, it is worth noting that none of the 3Bs reaches the figure quoted above from the brochure, 0.0009%. The ST boasts 0.00149% lM (which is close), while the NRB manages 0.00293%, while the previous 3B is a hair higher at 0.003%. Frankly I don't think these (nor also the THD figures) are all that significant, though very impressive. If there are significant progressive improvements in sound quality, it would be hard to attribute them to increments of PM or THD.

Of more importance to triple-zero he ear are power and noise measurements. Here we see a progression from 137 watts at clipping to 142.1 (NRB) to the STs 151 watts. But that has less effect on dynamic range (less than 1 dB, in fact) than the noise measurements. Over three generations the 3B S/N ratio has gone from 100 dB below its rated output to 115 dB (the NRB was -105), and the significance of these 15 dB cannot be overemphasized. To put it in perspective, a 10 dB level increase is an apparent doubling of sound level, and a tripling of power output. Not even the best current digital systems can achieve even 100 dB (see our feature, Squeezing The Lemon Again). A noise floor of 115 dB means you can put your ear to the speaker (and I did) and hear nothing, no hiss, hum, buzz or burble, if the rest of the electronic chain is similarly quiet.

It also means that the kind of noise and distortion (and at these low levels they are almost the same, the main difference being that noise obscures low-level detail while distortion rides on it dynamically) we expect in electronic music reproduction is absent to an astonishingly greater degree than found in other amplifiers. Other components can achieve close to 100 dB, and noise in a system tends to be cumulative, amplified with the signal at each gain stage. The buck stops at the amplifier stage with its own inherent noise, and it will faithfully amplify and reproduce all the noise that gets to it. This is also a good reason to seek the benefit of the common mode noise reduction provided by balanced cables.

Inefficient speakers will push the noise floor of an inferior amplifier down, while very sensitive ones will show up its failings. Though it is current orthodoxy to pair very efficient speakers with single-ended tube designs of low wattage, which usually have mediocre noise figures, it makes more sense to employ a very quiet amplifier, so that your 100dB-sensitivity speakers won't keep hissing at you. Of course, the best way to realize 115 dB of SN/Ratio as dynamic range is to use speakers that have this kind of sensitivity and can play at 115 dB.

And we haven't talked about sources yet. It could be argued that with a cumulative noise floor of only 80-to-90 dB at the amplifier input, any greater S/N ratio in the amplifier is irrelevant, but I think that the low-level distortion components inherent in an amplifier that has a poorer-than1 00-dB noise floor become a factor.

Certainly this is the most logical explanation for the quite unsubtle sonic improvement when the ST amp was substituted for the NRB. Though the noise floor in my listening room is quite low, just as one can hear below the noise floor, I could hear into the music more with the 3B-ST and layers of detail became apparent as veils of distortion/noise disappeared. The soundstage also moved back behind the speakers and had more sense of graduated depth, suggesting there was more than the elimination of noise involved. My guess is that it's transient distortions which ride the envelope of low-level complex musical signals, and which can't easily be measured as THD or IM, but do appear in noise figures. These distortion components can be power-supply leakage that modulates the audio, as well as noise generated in the circuitry addressed in the ST modifications.

Certainly the sonic improvements can be described as much in terms of absence than in relation to additive improvements:
less noise, less distortion, less veiling. I could hear into the musical performance in a way that wasn't possible with most other amplifiers, including previous Brystons.

The other noticeable change in the 3B ST was in the bass. There seemed to be a little less of it until I realized that it was there, but better controlled and clearly more specific in pitch. This effect was very similar to (but more subtle than) the comparisons I did (through the ST) between our digital and analog master tapes of the upcoming Ian Sadler organ CD. The analog bass sounded a little fuller and richer, but this was largely due to an increase in harmonic distortion, this becoming apparent when the digital's more focused and timbrally accurate bass was heard. Though THD and IM figures are far too low in both NRB and ST amps to attribute this heard effect to these again, transient distortion may well be a factor.

The other perceived part of the ST bass sound was its control, the woofers in my speakers being started and stopped in a way reminiscent of the powerhouse 7B monoblocks' control. Clearly the damping factor delivered to the speaker is much higher than that of the NRB, something which should be most clearly heard by owners of ported speakers.

In the weeks I've been listening to this amplifier I've talked to others who've heard it in various circumstances, simply because I felt uneasy at the magnitude of the improvements I'd been hearing. It's not too often that I get to directly compare an amplifier with its predecessor, and am inclined to think such refinements are usually incremental. Without offering a judgement, I asked a Bryston dealer, a sales rep (for other competing products), some of our writers, and various friends who knew my system fairly well and might notice a change for better or for worse, "what do you hear?"

I was surprised at the near unanimity. Some were surprised when the ST amps were compared with audiophile products at several times the price, but all commented on the aspects of soundstaging and bass control, as well as greater detail and openness that I've described above. One of our writers (and I'll let him reveal himself) immediately (well, not quite immediately, but after an hour of intense listening) had to have a 3B ST No previous amplifier, including the NRB series, and none other more exotic amplification devices, had seduced him over a decade. But as we listened to those audiophile icons we use to make such judgements (mostly LP, I might note), he kept grinning and saying things like, 'Gee, I never heard that harmony back there..." or, "that viola entry was late." A little Joni Mitchell, maybe some Gus Mahler, some Audio Ideas outdoor sounds and Debussy Preludes, and he was a goner.

Me too, I'm afraid. My recordings start via ST (Stuart Taylor's BMP-2 microphone preamplifier) and will end with the 3B-ST. Listening to the masters for our organ CD on the same day as I recorded them and heard to the organ itself in the church convinced me that Bryston has provided a new level of accuracy in audio reproduction. As our tube amplifier reviews in this issue suggest, this may not be what everyone wants, but it's sure nice to know that you can have it, especially when the price is so reasonable.

We invite you to experience the Bryston SST² Series amplifiers

20 Year Warranty - A Generation of Music