A properly designed meter will not suffer damage from changing modes while connected to voltage.
So, I can connect to 480VAC on a 600V scale and switch to Rx1 with no damage? You are placing a lot of faith in those fuses and not every meter has fuses. Early Simpson 260s didn't and I have seen the results of attaching to 480V with the switch in the wrong position. The meter literally exploded. That's the fun thing about working on 480V, arcs turn to plasma at low current but substantial heating. Entire swithboards burn to the ground from a simple phase to phase fault. The plasma current is too low to trip the protective devices but high enough to melt aluminum or copper.
Not only
can you do that, it won’t even pop a fuse (they’re across the current inputs, not volts), and the meter will still be in-spec afterwards. Remember my qualifier:
in a properly designed meter. And this isn’t a hypothetical — people who test meters do just this to them, and expect them to survive without damage.
Note that I did
not say that the current jacks could be used; that would pop the fuse. I only talked about switching from V to any other range, not moving the inputs.
Comparing an old analog meter is kinda pointless, since it wouldn’t meet today’s safety standards. A modern industrial meter like a Fluke 87V is expected to not only not explode when within spec but in the wrong mode, but to handle transient spikes many, many times higher than its maximum range. (Literally 8000V for Cat IV.) Now, the meter will not survive that degree of abuse, but it will remain safe. I don’t know whether it’s feasible to design an analog multimeter that can tolerate that kind of abuse, but for sure it’s expected of any halfway decent DMM.
The Zotek meters are great value for money for electronics use, but they wouldn’t survive the abuse we are talking about. Wrong mode on the volts jacks, maaaybe. (I don’t remember the test results for them.) But an 8KV transient? Not a chance.