The Great Debate: Linear vs Switch Mode
In hi-fi audio, power supplies are often treated as though one topology is automatically superior to another. Linear supplies are frequently assumed to be quieter and more “audiophile,” while switch-mode designs are often dismissed as noisy. The engineering reality is more nuanced.
The truth is, both topologies can deliver excellent results when properly designed. The key factors are implementation quality, noise control, regulation, grounding, and how the receiving circuitry handles the power it is given, not the topology label by itself.
How Linear Power Supplies Work
A linear power supply uses a transformer to step down the AC mains voltage, followed by rectification and regulation. The process is straightforward: the transformer provides isolation and voltage conversion, rectifier diodes convert AC to pulsating DC, filter capacitors smooth the ripple, and a regulator (often a series-pass transistor) maintains stable output voltage.
The advantages of this approach include:
- No high-frequency switching stage - This avoids the switching energy inherent to SMPS conversion, although rectifier noise, ripple, transformer fields, grounding, and regulator noise still have to be controlled.
- Simple design - Fewer components and well-understood behavior.
- Excellent transient response - Can respond quickly to load changes.
- Different EMI challenges - Linear supplies avoid high-frequency switching radiation, but transformer fields, rectification currents, and layout can still create interference problems.
Key Insight: Linear power supplies can be simple, robust, and very quiet when well implemented, but they are typically larger, heavier, and less efficient than switch-mode designs. Their performance still depends heavily on transformer quality, rectification, filtering, regulation, grounding, and physical layout.
How Switch-Mode Power Supplies Work
Switch-mode supplies take a fundamentally different approach. They rectify the incoming AC directly, then use high-frequency switching (typically 50kHz-500kHz) to convert and regulate the voltage through a much smaller transformer.
The benefits include:
- High efficiency - 85-95% is typical, meaning less heat and smaller components.
- Compact size - High-frequency operation allows tiny transformers.
- Wide input range - Can often accept 100-240V without switching.
- Potentially excellent regulation - Well-designed feedback and control loops can maintain very stable output across changing conditions.
The Real Question: Noise
The primary concern with SMPS in audio applications is noise. The high-frequency switching can create both conducted and radiated interference. However, modern SMPS designs can control switching noise through careful topology selection, filtering, shielding, layout, regulation, and, in some designs, techniques such as synchronous rectification or spread-spectrum switching.
More importantly, the noise present at the supply output is only part of the story. Local regulators, filtering, grounding, power-supply rejection, and circuit layout determine how much of that noise can actually reach sensitive audio circuitry.
| Characteristic | Linear PSU | Switch-Mode PSU |
|---|---|---|
| Efficiency | 30-50% | 85-95% |
| Noise Floor | Potentially very low | Design-dependent, potentially very low |
| Size/Weight | Large & heavy | Compact & light |
| Heat Generation | Significant | Minimal |
| Cost at High Power | Expensive | Moderate |
| Transient Response | Design-dependent | Design-dependent |
What Actually Matters for Audio
Rather than focusing on topology, consider these factors when evaluating power supply quality:
- Output noise - Measure it, don't assume based on type.
- Regulation quality - How stable is voltage under varying loads?
- Current capability - Can the supply meet the equipment’s actual current requirements while maintaining the intended voltage and regulation?
- Noise and isolation - How effectively are mains noise, switching energy, leakage currents, and interference prevented from contaminating sensitive circuitry?
- Implementation quality - Component selection and layout matter enormously.
- Downstream rejection - How effectively do local regulators, filtering, grounding, and the audio circuitry itself reject whatever noise remains?
The Bottom Line: A well-designed switch-mode supply can outperform a poorly designed linear supply, and vice versa. Judge the complete implementation by its regulation, noise, grounding, isolation, layout, and interaction with the circuitry it powers rather than assuming one topology is inherently superior.
Practical Advice
When evaluating equipment or considering upgrades:
- Don't judge a component by the power-supply topology alone.
- Look for meaningful measurements and engineering information where available, rather than relying on topology-based marketing claims.
- Consider the application and the complete implementation. A switch-mode supply can be entirely appropriate in high-performance audio equipment, including Class D amplifiers.
- Expensive linear supplies don't guarantee better sound.
- Before treating the power supply as an upgrade target, establish whether it is actually limiting the performance of the component or system.
The useful question is not whether linear or switch-mode is automatically better. It is how well the complete power system is implemented, how much unwanted noise reaches the sensitive circuitry, and whether any remaining limitation is significant enough to affect the finished component.