This internal vs external power supply issue is one of those cases that can drives people nuts, and audio is “full of it”. We heard that some gear sounds better with a supply outside. We also heard two suggested explanations.
One explanation is electromagnetic interference “getting into” the sensitive analog circuitry. There are numerous ways to combat the problem, and reduce it to an acceptable level (with a margin). The electromagnetic coupling is MEASURABLE. Note that we are talking about linear supplies, where the supply interference is predictable (AC line frequency and its harmonics). The point is –measurements will show you the impact on the audio. There is no magic here, with standard engineering techniques. If and when a designer can make the internal supply works as well as the external, why go external?
The other “explanation” has something to do with having a supply capable of supplying more current than is needed by the circuit itself. To my understanding, this “explanation” is reached by “ear”. Such a claim is at best very fuzzy.
First, the explanation breaks down when one is keeping the internal voltage regulators part of the supply inside, and moving the transformer and rectifiers outside. With such a setup, the internal regulators set the maximum current capability, not the external supply. If it still sounds better it can not be due to access current.
Second, when one keeps the regulators outside, the long wire self inductance, interacting with the bypass capacitors may be more of a problem than a solution, for sudden current surge requirement. That also puts the access current explanation into question.
Third, while extra current capability calls for a bigger supply, it does not always explain the need for external supply. Bigger does not always call for external. It depends on the power requirements and box size. How much more current is enough? 100% margin? 200% margin? 10000%? Is it really an internal vs external question? One needs to be clear about it.
That internal vs. external supply issue seems to me to be one of those “audio generalizations”. I am sure it began when people heard the difference, and possibly even measured the difference. But as often in audio, it was not well explained and certainly not well quantified. Instead it became one of those “I can hear it and I do not know why” statements. I can see opting for an extra heavy duty bumper on a light passenger car, but not on a military tank. With so many years of audio, I would hope for more insight than such generalization. Engineering is about dealing with physics, not “vague guidelines that may or may not hold true”.
I am not against having a designer decide they get better results on specific gear with external supply. It may be a wise strategy for some designs. My problem is with the “explanations”, and the “associated marketing” that leads a whole bunch of people to make generalizations.
Regards
Dan Lavry