"Tube gain doesn’t apply to headphones. The headphone amp section powers from the preamp section whilst the tubes install in the input of the output stage of the speaker amplifier. A few words about the A3's tubes. Implementing them to this extent is one of the two most unique things we've worked on for a very long time. These tubes aren't just a signal-path component but play a crucial role in our symmetrical feedback. This was achievable because the entire amplifier is structured with direct current coupling, eliminating the need for any signal-path capacitors. We use E88CC tubes from JJ which were the most suitable choice that meet the necessary parameters. They are sufficiently linear and operating voltages closely align with our requirements for this implementation. We aimed to capture and amplify the pure authentic sound of the triode, bringing its natural essence to life in the most vivid way possible. Selecting the right tubes is crucial for the proper functioning of the entire amplifier. It is strongly advised not to replace them with alternatives not specified by us. In fact, we don't recommend tube rolling the A3 at all. The tubes we ship in our products undergo a precise burn-in, testing and selection process using specialised testing devices developed by us. In the A3 we implemented a gradual ramp-up of the heater voltage with current limitation to extend the life of the tubes. This ensures that the minimum ~4'000 hours lifespan specified by JJ is secured." High and holy rollers have been warned. The hoi-polloi stock tubes have it; for more than 5 years if we religiously listen to A3 two hours each day come rain or shine. The truly addicted who double their consumption will still be good for 2½ years before a replacement pair comes due. A pair of tested and matched tubes is available directly from Canor or their dealer for €90 ex VAT. Whilst in headfi mode, we disregard them. They're not in its circuit.

Now a mini detour in seven sentences. In a SoundStage Australia interview, Siltech boss Edwin addressed the popular wisdom of soggy bass from transformer-coupled tube amps being from poor damping factor aka high output Ω. He called it popular missdom really due to impurities in the output transformers' copper windings magnetizing under signal. He'd built identical transformers wound with Siltech hyper-pure silver which lacks magnetizable impurities. Without altering output Ω, those windings eliminated poor bass control. Myth slain. It exemplifies how broadly shared observations can still be misdiagnosed. Their effect is real but not the cause. I believe that something similar applies to 2nd/4th-harmonic aka even-order THD from valve circuits explaining their gelatinizing effect. Having compared 300B preamps from Allnic, Manley and Vinnie Rossi and sundry capacitor- and direct-coupled amplifier types, I believe that the effect ties to capacitive coupling. My Kinki amps and prior LinnenberG avoid it. So does Vinnie Rossi's preamp. Ditto A3. And it shows. It'd be trite and foregone conclusion to call it a mix of tube and transistor attributes. We know it's a hybrid. Unlike most of their kind, it simply nixes signal-path caps. Et voilà, a different level of energy transmission as though a stopper had popped. It manifests as more shine and brilliance on top; an absence of drag; greater quickness; superior articulation; faster-twitch microdynamics.

Yet the rosier tone colours remain. So do enhanced connective tissue between/around images and their increased fleshiness. What's absent is the usual shadow of energetic reluctance. My preference for direct-coupled wide-bandwidth amplifiers and avoidance of active preamplifiers credits the lack of signal-path capacitors. Before I mint more popular myths, this is mere supposition. It's based on many years with different circuits noticing that those I liked best to use the most were all direct-coupled. Even were the true explanation to be different, it's still instructive to draw this connection when for all its deliberate triode tuning, A3 differs on that score like my other direct-coupled kit does. The upshot is simple. Listeners who prioritize musical electricity to fancy widebanders without energy-storing filter networks; true ribbon headphones; autoformer volume controls which transform voltage into current not burn off signal resistively like a brake pad; cardioid/dipole bass which lingers less; time-aligned speakers and other solutions for a less compromised time-domain performance… such listeners can enjoy A3's tubular spoils without incurring their usual jump-factor debt. This isn't a slow-RPM hybrid. It's a faster-revving sort. Call it a hibrid?

As I heard it, that one swapped letter neatly underlines A3's chief distinction. For someone on my slightly electrified side of the fence, it makes Canor's triode injection attractive. It omits typical small print a bit like embracing richer eating habits without putting on weight. If that actually worked, it'd be quite the viral diet. Sonically, we might call the A3 signature creamier without the fat. It's not quite accurate unless 'fat' stood in for the common effects of lardiness: a lowered state of vitality expressed in slower reflexes. Somehow John Westlake & Team's small-signal triodes embedded deeply in a transistor power stage including its feedback loop without routine coupling caps secures certain tube traits whilst sidestepping others. Those usually come as a take-it-or-leave-it package. We could justifiably call A3 a double hybrid. It not only splits the bill between tubes and transistors. It also splits the tube half. I didn't quite know this could be done though did earlier read about it at Serbia's NAT Audio. With my DAC's 6Vrms XLR feed, A3's attenuator sat at -40dB. I didn't get beyond -30dB on fare which barely flickers because it hovers some 20dB or more below digital zero. Whilst shoving 'only' 100wpc/8Ω into 82dB sealed speakers, my reserves were stacked. A3 packs sufficient voltage gain for even thirstier speaker scenarios. Whilst on the matter of what it can drive, let's consider HeadFi. Drawing ~3A of current, my Raal 1995 ribbons were out. Though on paper Susvara's inefficiency suggested likewise, with 6Vrms signal off my Cen.Grand DAC, voltage gain would multiply favourably. Could that go loud enough if perhaps still feel a bit weak at the knees? Surely it wouldn't be the bee's knees?