If you read the Stereophile review of the Momentum, JA was only able to achieve 80dB SNR (ref 1w), so I would expect that the classic is similar.
The slightly higher than average noise floor is really nothing to be overly concerned about, particularly these days with the dearth of dynamic recordings.
The technical reason that it's a bit higher is because the input stage gain is quite low on Dan's amps. Typically, input stage gain is high, so the input stage is able to dominate the overall noise figure performance, and of course small signal input stage devices are usually low noise transistors. Reducing the gain of the input stage allows the usually noisier transimpedance stage to dominate the noise figure performance, and efforts to degenerate the input stage also add thermal noise inputs.
So you might ask, why design it that way?
I can only guess but one advantage is derived from needing to use less internal HF compensation, which translates to high slew rate and nearer to desired unconditional stability.
A second reason is that distortion is more likely to be more even across the frequency spectrum, and since input tails usually generate odd order distortion products, a lower input pair gain will reduce the LTP's overall contribution to the harmonic signature.
Talking to Dan at a show a few years ago, he said that he deliberately designed the momentum to be slightly to the warm side of neutral, which is the preference for most audiophiles.
|