Preamp suggestions for Atma-Sphere Class D Monoblocks


As I impatiently await the arrival of my Atma-Sphere class-d monos, I am writing to solicit opinions from experienced members on this site. Balanced inputs and outputs as well as the ability to connect a subwoofer are important. Headphone amp would be nice but not critical. Ditto for built-in phono. These are some of the tube and solid state units I am considering (not necessarily exhaustive):

  1. Atma-Sphere MP-3 mk3.3
  2. Cary SLP-05 (ultimate edition, if I can find a pristine used pair at a good price)
  3. Modwright LS-100
  4. PrimaLuna EVO 400
  5. Supratek - Champagne or 6SN7 Cabernet - I have no idea what the difference between these two are and the website is not helpful at all
  6. Anthem STR
  7. Benchmark LA-4/HPA-4
  8. Bryston BR-20 (or one of their purely analog variants - BP-19?) - I am hesitant to bundle streaming and dac into a preamp
  9. NAD M66
  10. ?

I will initially use the Bel Canto DAC2.8 as a dac and preamp with an Auralic Aries G1.1 as the streamer. My speakers are Totem Acoustic Forest Signatures, sub is a REL s/510 that I seldom use. I am unsure whether I can use the speaker level connections with the A-S monos (my current Bel Canto Ref510S class d amp has a chassis grounding screw to which I connect the REL’s ground wire).

Philosophically speaking, I have owned a wide range of equipment over the years - everything from "measurements above all" type gear (a full Benchmark stack), a few tube pieces (PrimaLuna, Tavish Design, Rogue), Class A amplification (Pass Labs XA-30.8) and a completely active Genelec 8351b based system. I seem to be slowly but surely gravitating towards subjectively great sounding gear, regardless of how they measure - the Totems are a case in point. However, I don’t want to find that I have moved too far in any one direction. At the end of the day, I value realism and emotional connection more than the ability to deconstruct the music but I don’t want to listen to a sloppy system that homogenizes everything. I’m sure this is all clear as mud! It is however, a reflection of where I am at the end of 28 years of being on this hamster wheel. All thoughts and suggestions welcome. Thanks in advance.

Ag insider logo xs@2xsrama

Srama,

Between the LS-100 and the MP-3 I thought soundstage and realism were roughly equal, but I definitely preferred the tonality of the MP-3, which is likely due to the reasons Ralph outlines above. One other variable is that the LS-100 is tube rectified and the rectifier has a definite effect on the sonics. I ended up liking the 5U4GB much more than the stock 5AR4. The MP-3 is solid state rectified and eliminates that variable.

I failed to mention before that I also owned a Don Sachs DS2, sort of the forerunner to the Raven. It was terrific in its own right but more romantic than the LS100. I owned them at the same time and slightly preferred the LS-100, but if the Raven is as much better than the DS2 as most (including Don Sachs himself, who should know) are saying, that probably changes the equation.

I understand if you would rather not do so in a public forum, but it would be great to hear your engineer’s perspective on the Spatial Raven and the Backert Labs Rhumba, and how the MP3 differs from these designs (or provide some links if you have already discussed this).

@srama There are two main differences. The first you already know about- the MP-1 and MP-3 employ a balanced direct-coupled tube output, which allows it to be very transparent. Usually in most preamps the output coupling cap defines the low frequency cutoff and also causes the output impedance to increase as the frequency goes down. When you direct-couple, the low frequency cutoff is defined by coupling caps prior to the output and so are much smaller in value (typically by 2 orders of magnitude) which allows them to be more transparent and the output impedance with respect to frequency is a ruler flat line across the audio band, something you can't do with coupling caps and very hard to do with output transformers.

To my understanding the Spatial uses an output transformer. Personally I think if you can eliminate the transformer the circuit will be lower distortion and wider bandwidth. Transformers unavoidably introduce distortion due to the hysteresis loop that is always present (although minimized in good designs).

As best I can make out, the other difference is that in addition to being fully balanced, the MP-1 and MP-3 are also fully differential. This allows a lower parts count, lower noise (in theory up to 6dB per stage of gain) and lower distortion although the distortion difference as opposed to a non-differential balanced circuit is fairly small.

Not having heard the Spatial or Rhumba I can’t comment on the sound (which by all accounts is excellent), but what I would expect (as this has played out in other comparisons over the last 35 years) is the use of the direct coupled output is immediately obvious in terms of transparency and bandwidth (the latter best heard in the bass region- the deeper your speakers go, the more obvious this becomes).

Not having an output transformer also means the preamp is less critical of the interconnect cable used between it and the amplifier. Transformers are inductors and cables have capacitance, put the two together and there will be a resonance at some (usually ultrasonic) frequency. Not having the numbers available (such as the inductance of the transformer presented to the cable) its hard to know what frequency, but I noticed that the designer recommends avoiding higher capacitance interconnect, so I’m sure he’s aware of this issue.

+1 Aric Audio Motherlode XL. I have used this combination in my refernce system and it sounded superb.

@atmasphere, thank you so much for your detailed response. I cannot claim to understand all the details, but I think I'm following in broad strokes. One thing I noticed in comparison to the other two amps under discussion is the significantly larger number of tubes in the MP-3 - is this because it is the only truly balanced preamplifier is this set? Does this have any implications on maintenance, require more frequent changes to an entire complement of tubes, etc.?

Just to clarify, I don't intend to  jump into the rabbit hole of constant tube rolling - my limited experience with doing this in the past caused more stress than anything else! I'm only concerned with long-term reliability and cost of ownership. 

@srama

The atma sphere monos have unusually low gain for an amplifier (22db)  compared to most that typically have 29 or 30 db gain.  

much like my Pass Labs XA 25 amp that has 20 db gain, they really match well with preamps that have higher than typical gain.  
My Cary SLP 05 has that with a 24 db low noise gain stage with balanced connections.  It matched amazingly well with the Pass amp and should do just as well with the atma spheres.  The SLP 05 does create a smooth holographic atmo sphere of its own in addition to be excitingly dynamic.  Get the 05.