Class A Amplification

Class A Amplification expresses the Musical Communication, Low-Power Amplification, and Engineering-Discipline-as-Voicing schools at their intersection. The topology has a strong Japanese-Artisan affinity (Leben, Shindo, Audio Note UK adjacent), an American-Designer affinity (Pass Labs / First Watt / Threshold), and a British affinity (Sugden, Audio Note UK). It does not belong to any one school exclusively — like the brands that argue for it, it sits at the intersections.

What It Is

In a Class A amplifier, the active device — output tube, output transistor, or output MOSFET / SIT / JFET — is biased so that its quiescent current is at least half the peak signal current the device will deliver. The device therefore conducts current through the entire audio signal cycle (positive and negative excursion) with no portion of the waveform delivered by a switched-off device. Class AB by contrast biases the active devices to switch off across a small region around zero crossing, with a complementary pair handling the positive and negative halves separately and the small switching region managed by feedback. Class D switches the output device at high frequency (hundreds of kilohertz) between fully on and fully off, with the audio signal recovered by a reconstruction filter; the device is never operating in its linear region. The Class A decision is independent of topology (single-ended or push-pull), device family (tube or solid-state), and power rating. A 25 W push-pull Class A solid-state amplifier (Sugden A21, Accuphase E-380) operates in Class A throughout its range. A 200 W push-pull Class AB solid-state amplifier biases Class A through the first ~10 W and transitions to Class AB above that point — listeners call this "Class A up to X watts" and the spec is meaningful because the bias point determines small-signal behavior. The consequence of the Class A decision is high heat dissipation (most input power is converted to heat in the output devices, not to audio), large heatsinks (or in tubes, glowing plates), substantial power supply (the device must be biased to conduct the full envelope), and a chassis-and-thermal scale that the Class AB / Class D market does not have to absorb.

Why It Matters

Class A matters because it is one of the clearest cases in audio where the *posture* decision — what the active device does across the signal cycle — produces a different sonic result than a topology decision or a power decision. SET argues at the topology level (single-ended). High-efficiency loudspeakers argue at the sensitivity level. NOS argues at the filtering level. Class A argues at the bias level — the question of whether the active device is on through the entire signal or only through part of it. The structural claim is that some properties listeners value (tonal continuity across transients, absence of crossover-region small-signal artifacts, harmonic structure that reads as natural rather than added) are decided at this junction, and that they are decided by the bias commitment rather than by feedback, topology, or device choice. The page exists so the editorial choice can be understood at the architectural-posture level rather than as Pass Labs nostalgia or "vintage Class A romanticism." If amplification is judged by efficiency, heat envelope, and power-per-dollar, Class AB and Class D are better engineering. If amplification is judged by what the active device does across the cycle and how that propagates to the output, Class A makes a defensible case validated by seven decades of repeated commercial production and continuous listener demand.

What It Gives

• No crossover-region small-signal artifacts — the active device conducts through the entire signal cycle, so the small-signal region around zero crossing is not a transition between active and inactive devices; listeners experience this as tonal continuity through quiet passages and absence of switching-character on transients
• Harmonic structure as a structural property of linear-region operation — Class A devices operate in their most linear region throughout the cycle; the harmonic content the listener hears is a property of the device biased into linearity, not a residue of feedback correcting non-linearity
• Posture-independent of topology — the bias decision applies equally to single-ended (SET, First Watt SIT) and push-pull (Sugden, Leben, Accuphase), to tube and solid-state; this lets the Class A cluster span device families and topologies that would otherwise read as unrelated
• Engineering discipline made visible — Class A chassis publish the trade in their physical form: exposed heat sinks (Sugden), large mass-thermal assemblies (Pass, Accuphase), or glowing output tubes (Audio Note, Shindo); the cost of the bias decision is not hidden, and the brand's editorial position is visible at the chassis level
• A durable architectural commitment across seven decades — Sugden A21 in continuous production since 1967; Accuphase Class A integrateds since the 1970s; the Pass Labs / Threshold / First Watt continuity from the 1970s; the Audio Note UK / Shindo / Leben tube tradition; the position is persistent where fashion-driven topologies fade

What It Costs

• Heat dissipation is real and continuous — Class A converts most input power to heat at the output device, and the chassis runs hot even at idle; this is a thermal commitment the listener accepts rather than an inefficiency to be engineered away, and rooms without ventilation or with heat-sensitive equipment nearby will feel the cost
• Power efficiency is poor — Class A converts about 25% (push-pull, theoretical max) or ~50% (single-ended, theoretical max) of input power to output audio; the rest is heat; for any given output rating, a Class A amplifier draws meaningfully more wall power than a comparable Class AB or Class D design
• Cabinet size, mass, and power supply requirements drive cost — the active devices must be biased for full-envelope conduction, the supply must deliver that bias current continuously, and the heatsinks must dissipate the heat; the price floor for a Class A amplifier at a given output rating is higher than for Class AB at the same rating
• Power-per-dollar and power-per-pound trail Class AB and Class D — a 200 W Class A amplifier is exotic; a 200 W Class AB or Class D amplifier is mainstream; listeners who anchor on watts-per-dollar will see Class A as inefficient engineering, and the trade is invisible to that framework
• Class AB and Class D legitimately win on efficiency, heat, weight, cost, and power-per-dollar — listeners who want high SPL in difficult rooms, who run multiple low-sensitivity speakers, who want compact integrated amplification, or who are at price tiers where the Class A trade dominates the rest of the chain are correctly served by well-designed Class AB or Class D; the Class A argument is principled, not universally better

Why Listeners Still Choose This

Class A amplification is the architectural commitment that the active device — tube, transistor, or FET — conducts through the entire audio signal cycle rather than switching off and on across portions of the waveform. The decision is upstream of topology (single-ended or push-pull), upstream of device family (tube or solid-state), and upstream of power rating. Class A persists because listeners committed to tonal continuity, harmonic naturalness, and freedom from crossover-region artifacts have chosen the bias decision for seven decades, across SET tubes (Audio Note UK, Shindo), Class A push-pull tubes (Leben CS600X), Class A solid-state push-pull (Sugden A21, Accuphase E-series), and single-ended Class A solid-state (Pass First Watt SIT designs). Class AB and Class D are coherent engineering — they win on efficiency, heat, cost, and dynamic envelope. The Class A argument is not that they are wrong; it is that operating the active device through the full cycle produces a different result than operating it through part of the cycle, and that the difference is what a chain voiced for tonal continuity is built around.

The received view of Class A is that it is a vintage engineering position preserved by audiophile nostalgia — kept alive by tube enthusiasts and a small cluster of Class A solid-state holdouts (Sugden, Accuphase, Pass / First Watt) who refuse to move to more modern, more efficient designs. That reading misreads the architectural fact: Class A and Class AB do different things at the device. Class AB biases the active device to switch off across a portion of the signal cycle (typically the small-signal region around zero crossing), with feedback and crossover-region compensation managing the switching artifact. Class A biases the active device to conduct through the entire cycle, with no switching at all — the active device is in its linear operating region the whole time. The two routes are not equivalent at the transistor or tube; they produce different small-signal behavior, different harmonic structure, and different transient characteristics. The result listeners report most consistently is tonal continuity through transients and absence of "switching-character" artifacts in the small-signal region — and the consistency of the report across implementations from very different topologies (Sugden Class A push-pull solid-state, Pass First Watt single-ended SIT, Audio Note UK Class A SET tube, Leben Class A push-pull tube) suggests an architectural cause rather than per-implementation variance. The persistence is durable. Sugden's A21 (1967) has been in continuous production for ~60 years; Nelson Pass's career across Threshold, Pass Labs, and First Watt has been a 50-year argument for low-power Class A solid-state; Accuphase's A-series Class A integrateds have been a Japanese reference for ~40 years; the tube tradition (Audio Note UK, Shindo, Leben) carries the same posture from the other end of device physics. After multiple generations of Class AB and Class D commercial success, Class A has not declined — it has stabilised as a permanent minority position by listeners who value the result enough to absorb the heat, the size, and the price floor.

System Fit

Class A amplification pairs naturally with brands and components whose ideas align on tonal continuity through transients and freedom from switching-character artifacts: voiced sources (analog front end with low-output MC + SUT chain, or NOS / R2R digital), moderate-to-high-sensitivity loudspeakers (typically 87 dB+; high-efficiency 93 dB+ for the SET / First Watt low-power expression), and listening rooms small-to-medium enough that the Class A power envelope is adequate. The canonical chains span the full device-family range: Audio Note UK Class A SET (300B Ongaku family) driving Audio Note AN-E speakers; Shindo Cortese 300B SET driving DeVore Orangutan or Altec horns; Leben CS600X Class A push-pull tube driving DeVore O/96 or Harbeth Super HL5; Sugden A21SE Class A push-pull solid-state driving Harbeth, Graham, ProAc, or DeVore O/93; Pass First Watt SIT (J2, F8) single-ended solid-state driving high-sensitivity speakers; Accuphase E-380 / E-650 Class A integrateds driving moderate-sensitivity floorstanders in the 88–92 dB range. The chain decision propagates: the Class A bias commitment presupposes a speaker partner whose sensitivity makes the modest Class A power envelope concert-level rather than insufficient. Anti-pairings reveal the trade-off: large rooms requiring 100+ W average SPL (Class A topology is inefficient at scale; Class AB or Class D delivers the same SPL at substantially lower wall power and chassis size); low-sensitivity loudspeakers (below 84 dB) where the Class A power envelope is insufficient; listeners with ventilation constraints or heat-sensitive equipment co-located; cost-floor-constrained builds where the Class A premium dominates the rest of the chain budget; and listeners who anchor amplifier evaluation on power-per-dollar or efficiency, who will read Class A as inefficient engineering and miss the editorial argument.

Understanding Class A Amplification

Related Brands

Related Technologies

Further Reading

• J. E. Sugden — A21 lineage (manufacturer page)
• Pass Labs — Class A amplification (manufacturer page)