Hopefully Al **will** chime in. But in the interim, much has been written about this subject. Before I go off the techno-babble deep-end here, let me say that I am dubious that trying different output taps will harm a tube amp. In most cases the conventional wisdom is to use the taps that sound best ... to you.
Al ... I started to explain my understanding of nominal tap output impedance, back impedance matching, output voltage regulation, and so forth, but started writing a magnum opus. So please ... chime in.
Let me just say that if a tube amp is able to produce clean power off the 4 ohm taps while it is driving a speaker whose impedance likely varies, possibly quite considerably as a function of frequency, then I would expect that bass will be tighter and extended, and higher frequencies less augmented.
The reason is that in most cases the "apparent" output impedance of the tube amp will be lower off the 4 ohm taps, thus resulting in less output voltage variation. I want to emphasize that this assumes the amp is being operated within its design parameters, i.e., not being over-driven.
In other words, the tube amp will perform somewhat like a Voltage Paradigm amp (almost always a SS amp). Most SS amps have extremely low output impedance and present a constant voltage source to the speakers if operated within their operating parameters.
Now ... what I just wrote assumes that the speakers in question were designed and voiced to be driven by a SS amp. In such cases, tube amps having lower output impedances and therefor tighter source output voltage regulation will possibly do a better job driving such speakers off the 4 ohm taps ... **IF** the amp is able to deliver clean power.
By contrast, if the amp is overdriven, the impedance mismatch between the speakers and the amp will result is attenuated power delivery and increased distortion which will likely sound bad. Unless, one likes to listen to tube amps while they clips and produce lots of funky distortion. :)
