Power is always reduced by running mismatched in either direction, but the tone change is different. Mismatching low (8-ohm amp into a 4-ohm cab) produces a flubbier, thicker, smoother tone; mismatching high (8-ohm into 16-ohm cab) produces a flatter, more complex, more midrange tone. Matching always produces the most power & most even frequency response.
If you are going to mismatch VALVE amps it is actually less bad, (for the amp or more specifically, the Output Transformer), to mismatch to a lower impedance, (16 ohm into 8 ohm cab), rather then higher. Unlike solid-state amps, valve amps are basically self limiting current wise (the valves!) into a lower impedance, though the valves will take more wear. Into a too high impedance the risk is different – potential very high flyback voltages can fry the OT. But most valve amps with strong output transformers will take a 1/2 to 2x mismatch without complaining.
Although not recommended, technically you can short circuit a valve amps output (0 ohms) without frying it – it is trying to put a signal into an open circuit (virtually infinite ohms) that is a real killer. Total opposite from solid-state amps of course – short circuits will kill the power transistors pronto, but they’ll sit happily all day with no speaker load applied.
Solid State amps: Safe with rated load or with any higher impedance speaker load (up to & including an open circuit ie infinite ohms.) Develop less power as impedance increases. Do not short circuit (0 ohms) as this is sudden death for the output transistors.
Valve amps: Match impedance if possible. If mismatching, it is safer for the amp to mismatch low. This will wear the valves but the amp shouldn’t suffer. A short circuit (0 ohms) is normally survivable.
If mismatching high there is a risk to the OT, which increases with the severity of the mismatch, & with how hard you are pushing the valve output section. The ultimate ‘high’ mismatch is no speaker load ie infinite ohms. If trying to pass a signal into this then there is a severe risk to the OT from high flashback voltages which can arc through the insulation layers & burn out the tranny. Mismatching between 1/2 & 2x the impedance the amp ‘expects to see’ is normally problem free for most amps with healthy OTs. It is never guaranteed safe though, & being manufacturer specific Marshalls fail much more often when doing this then Fenders do. There is a lot of misunderstanding about impedance & mismatching issues, but I repeat that the advice I’m giving is correct for valve amps. (The exact opposite is true for solid state amps, but they work very differently, usually having no output transformers for starters!)
For why Marshalls are extra sensitive, could be the transformer design, could be that selector switch. I personally would not worry too much about a 2:1 mismatch too low, but I might not do a mismatch high on Marshalls with the observed data that they are not all that sturdy under that load. In that light, pulling two tubes & leaving the impedance switch alone might not be too bad, as the remaining tubes are running into a too-low rather than too-high load.
Yes running 8 ohm amp –> 16 ohm cab is probably within normal safe limits (within 1/2 to 2x impedance range) for most valve amps. In fact you’ll often get away with playing Russian Roulette by running a 4 ohm amp to a 16 ohm cab, but the risk to transformers is definitely greater when going into a higher then intended impedance. You really are always better to have the speaker impedance lower if mismatching! If you value your amp that is. Going lower strains the valves more then normal, but they are disposable in a way that the output transformer isn’t. (Which is after all why valves are removable from their sockets.)
The thing you CAN do to hurt a tube output transformer is to put too high an ohmage load on it. If you open the outputs, the energy that gets stored in the magnetic core has nowhere to go if there is a sudden discontinuity in the drive, & acts like a discharging inductor. This can generate voltage spikes that can punch through the insulation inside the transformer & short the windings. I would not go above double the rated load on any tap. & NEVER open circuit the output of a tube amp – it can fry the transformer in a couple of ways.
It’s almost never low impedance that kills an OT, it’s too high an impedance. The power tubes simply refuse to put out all that much more current with a lower-impedance load, so death by overheating with a too-low load is all but impossible – not totally out of the question but extremely unlikely. The power tubes simply get into a loading range where their output power goes down from the mismatched load. At 2:1 lower-than-matched load is not unreasonable at all. If you do too high a load, the power tubes still limit what they put out, but a second order effect becomes important.
There is magnetic leakage from primary to secondary & between both half- primaries to each other. When the current in the primary is driven to be discontinuous, you get inductive kickback from the leakage inductances in the form of a voltage spike. This voltage spike can punch through insulation or flash over sockets, & the spike is sitting on top of B+, so it’s got a head start for a flashover to ground. If the punchthrough was one time, it wouldn’t be a problem, but the burning residues inside the transformer make punchthrough easier at the same point on the next cycle, & eventually erode the insulation to make a conductive path between layers. The sound goes south, & with an intermittent short you can get a permanent short, or the wire can burn though to give you an open there, & now you have a dead transformer.
So how much loading is too high? For a well designed (equals interleaved, tightly coupled, low leakage inductances, like a fine, high quality hifi) OT, you can easily withstand a 2:1 mismatch high. For a poorly designed (high leakage, poor coupling, not well insulated or potted) transformer, 2:1 may well be marginal. Worse, if you have an intermittent contact in the path to the speaker, you will introduce transients that are sharper & hence cause higher voltages. In that light, the speaker impedance selector switch could kill OT’s if two ways – if it’s a break before make, the transients cause punch through; if it’s a make before break, the OT is intermittently shorted & the higher currents cause burns on the switch that eventually make it into a break before make. Turning the speaker impedance selector with an amp running is something I would not chance, not once.
Too high impedances on the speaker outputs are much more dangerous for valve amps then too low impedances are. You can short circuit the + & – speaker output connections by connecting them together. (This gives very low impedance, close to zero ohms.) Valve amps can survive this. But if you don’t believe me then try this: power up your own valve amp, unplug your speaker, take it off standby & crank that baby into an open circuit very high impedance load. (Fresh air) You stand a very good chance of frying your amps OT!
Speaker load impedances and reflected loads to the output tubes are all “nominal”. An 8-ohm speaker may actually look like anything from 6-ohms to 100-ohms, depending on the frequency, since the reactive impedance changes with frequency. This means that the reflected load to the tubes is varying widely over the frequency range.
A nominal 8-ohm load may reflect 4k to the plates of the output tubes with a given transformer. The amp might be designed to produce its maximum power into this load, with a designed frequency response. This is the “power bandwidth”. If we change the load to 16-ohms, the reflected load doubles and the frequency response shifts upward. We lose bass but have a brighter sound, and also lose power. If we change to a 4-ohm load, the reflected impedance drops to 2k, into which the tubes produce less power, and the bandwidth is again narrowed.
The reason for the confusion, I believe, is that people think tubes will try to behave the same way transistors do. Into half the load impedance, a transistor will try to deliver twice as much current. The device may overheat and destroy itself in the process. Tubes, however, simply don’t behave like transistors.
The design issue for impedance matching comes into play when a designer takes the approach that “everything is critical”. In some circuits, this may be the case. Tubes don’t really care. There is no optimum load for a tube unless you are going for minimum THD, and this then depends upon the other operating conditions. For guitar, criticality is purely aesthetic. The designer says “this is good”, “this is bad” and in that decree believes it to be so. He is correct in his subjective impression, but should not confuse the subjective and objective.