|
|
frequencies. A regulator can have low output impedance down to DC. Excessive current loading can cause a drop-out effect similar to drop-out with low line voltages.
Disturbances injected from other circuits
Signal interference from other parts of the circuit can enter into an amplifying stage. If it is from another channel, this creates cross-talk. If this is from an entirely different source, such as a tape output signal interfering with a tape recording signal, annoying interference results. The most common case is the signal interfering with itself, typically an output stage that feeds signal currents back through the power supply rails to the input or driver stages. In severe cases this results in low frequency oscillation ("motor-boating"), but mild cases can result in odd phase shifts and frequency response changes. Because regulators can be effective down to DC, they are helpful with these problems.
Disturbances injected into the power supply that could disturb other circuits
This is where some regulators cause more problems than they solve. If a series-type regulator is supplying a circuit with changing current load, these current changes are magnified "upstream" (i.e. ahead of ) the regulator. This can aggravate interference with other circuits. In severe cases, especially with computer-type switching power supplies, the regulator presents a negative resistance to what is driving it. Using the wrong filtering can then cause oscillations. Shunt-type regulators, on the other hand, actually smooth the current upstream from the regulator.
|
|
|
|
|
|
Changes with time and temperature
Slow changes in supply voltage with time or temperature can cause finely-tweaked circuits to go out of adjustment. In audio applications, this is less of a problem than in, say, instrumentation, but it can still lead to subtle sonic changes that are often attributed to "warm-up" or "break-in."
Sonic Tradeoffs in Regulator Design
Regulators definitely have an effect on the sound of the amplifying stage they feed. In many cases there is an improvement due to the steady supply voltage and low supply impedance. But beyond this, people have noticed tonal changes to the sound based on the type of regulator used. Since often the power supply is part of the signal path, serving as the stiff counterpoise as the brick wall in the example above, the characteristic of the regulator can affect the sound of the circuit. This is a notoriously fuzzy area to analyze, though, and very dependent on subjective evaluation. However, several things seem to come up among many different experimenters:
1. High-gain solid-state regulators, typically using
Op Amps, impart a dry, hard sound, typical of
high-feedback designs or solid-state amplification.
2. Low-gain or "no-gain" regulators sound better,
but must be used carefully, since their imperfect
regulation can cause problems of stages interacting
with each other.
3. Non-regulated, well-designed power supplies using
good quality components can sound very good, but
are subject to shifts in tube operating points with
line changes.
|
|
|
