Audio Research Reference 150 power amplifier
It's always good to have a reference. No matter the endeavor, references help guide us and set standards for all we do. For many hours of every day, I'm lucky to enjoy the reference of live, unamplified music. Right now, I average over 20 hours a week of rehearsals and performances of various ensembles, and four to five hours of listening to recorded music on my hi-fi. Clearly, for me, my musical reference is not the sound of my audio system, but the sound of live music created in various venues and acoustics.
And the more time I spend making real music, the less I'm impressed with home audio reproduction. No audio system I've heard has been able to create the sound of 120 voices singing Beethoven's Symphony 9, or the physical sensation I feel through my entire body during a performance of Orff's Carmina burana, or the total envelopment I sense when I listen toor singrenaissance motets in an acoustically lush space. To my ear, all recordings and all audio systems sound fake. Not only do audio systems impose on music colorations, grain, and an unnatural electromechanical feel, they also omit the dynamic nuances, spatial cues, and sense of scale that make music come alive. Fortunately or unfortunately, my reference standard of musical reality is too high, too strong, too real to allow me to be fooled into thinking a recording is actually live music.
And yet, I find myself constantly inspired to make reproduced music sound as good as it possibly can. I think it comes down to the fact that there are so many recordings I love, of so many musical moments that will never again occur, that I want to have as deep a relationship with this music as possible. I find the quest of being an audiophile thrilling, challenging, and fun, and the journey toward audio excellence often more rewarding than the destination itself. My ultimate goal is that of most audiophiles: to hear my favorite music and performances with as little as possible coming between me and the recorded event.
But because technical perfection in reproduced audio is, in my opinion, impossible, I want my stereo to also serve as an emotional conduit between me and my music. I want gear that is accurate, but when accuracy is not possible, I want a component that can overcome its shortcomings with soul. My reference gear, therefore, has always been equipment that simultaneously sounds as close to the real thing as possible and makes possible in me the strongest emotional connection to my music.
So when Audio Research Corporation introduced, as part of their new Reference line, the Reference 150 balanced tube amplifier, designed specifically for my favorite output tube, the KT120, I knew I wanted to hear if it could come close to re-creating my own reference: real live music.
A New Reference
The Audio Research Reference 150 ($12,995) is a member of the venerable Minnesota company's new Reference line. Based on what ARC learned in creating the Anniversary Edition preamplifier, they've introduced the Reference 250 monoblock and the Reference 150 stereo amplifier, the latter replacing the Reference 110, which Robert J. Reina reviewed in August 2007. ARC has come up with new, larger, higher-bandwidth output transformers to better match the impedances of the KT120 output tubes used in the Ref150. According to ARC's Dave Gordon, "These transformers were designed in-house by our engineering department, and they are manufactured for us nearby in the Midwest. They are not sourced from Asia."
The Ref110's power-supply energy storage has been doubled in the Ref150, to 1040 joules. "Our final proprietary capacitor designs were chosen empirically, after many, many hours evaluating and comparing each different design in prototype amplifiers," Gordon said. "We ended up using a mixture of Teflon, copper, and hybrid materials in the capacitors in the Ref150." The Ref150 is claimed to output 150Wpc into 8 ohms. This power comes by way of eight KT120 output tubes in matched pairs, each pair driven by a 6H30 driver tube. The output-stage coupling is a combination of the familiar ultralinear configuration and ARC's patented "partially cathode-coupled" topology. The Ref150 has 16, 8, and 4 ohm taps.
Biasing of the Ref150's output tubes is done in pairs, requiring matched sets of Sovtek KT120 tubes. According to Gordon, "Individual tube bias requires individual coupling caps for each tube. Regardless of how tightly measured and matched the coupling caps may be, there are slight amounts of phase shift introduced when multiple coupling capacitors are used for multiple tubes. There is a certain purity when one coupling cap is utilized in one bias test point for each tube set vs each tube."
The downside of using matched pairs is that when an output tube goes bad, you'll have to buy a matched pair instead of just a single new tube. However, the word on the street is that the KT120's longevity and reliability are pretty darned good; this shouldn't be a constant problem. Hours of tube use can be tracked via a handy LCD display inside the Ref150's chassis. The Ref150's circuit is fully balanced; it has only XLR inputs. Two low-noise fans at the rear cool the tubes inside. The amp is connected to AC via a 20-amp IEC connection.
The Ref150's faceplate is made of nicely brushed aluminum in silver or black, designed in classic Audio Research form: nothing too fancy, but well made. Inside, the circuit layout is all anyone could ask for: clean, clear, concise. The parts quality looks especially high, and the solder points very well done. The two aluminum handles dominating the front panel are extremely useful when trying to lift and position this beast's 75 lbs.
The Audio Research Reference 150 was a piece of cake to set up. Before lugging the amp to its new home under my 250-lb butcher-block audio table, I removed its aluminum top plate to access its innards. I installed all eight KT120 tubes, each labeled to ensure its being paired with its proper mate, hooked up my speakers (at first my Revel Performa F30s, and later a pair of Mårten Djangos) to the 4 ohm taps, and switched on the Ref150. After a few hours of burn-in, I checked the bias of each pair of output tubes and found them all within 1mV of ARC's target setting of 65mV. During the first month of listening I checked the tube biases every few days but never noted any drift. I let the amp get at least 200 hours of playing time before I did any critical listening.
The Ref150 sat downstream of a Bel Canto DAC 3.5 VB Mk.II D/A converter fed digital data by a Bel Canto CD2 CD player and analog signals from a Clearaudio Concept turntable with Nano phono stage, or a Clearaudio Ovation turntable with Basic+ phono stage.