A little long, and a little technical... but, if you want to know exactly what you can do, what you can't do, and the underlying reasons why, read on.
(Hopefully this dispels any myths, but also prevents people from "doing something dumb" and regrettable, or doing something unsafe for those of you who like to "jump in the deep end" and get into the
electric/electronic circuits.)
The issue with "dimmers set to 100%" to
power anything other than lights is their
TRIAC (or double
SCR) triggering gets a little funky if the load has a complex
impedance. (That is, if the
voltage waveform and
current waveform do not remain in-phase.)
Incandescent light bulbs have a simple "
ohmic resistance" with no complex roots, so they're fine. Motors and "imperfect" transformers naturally present a load
impedance with an inductive component; linear and switch-mode
power supplies often present a load
impedance with a capacitive component. These complex impedances can cause the
TRIAC or SCRs in the
dimmer to chop up the
sinusoidal AC
power wave at some
point before or after the zero crossing (even if set to 0% or 100%), leaving you with "dirty
power" that's not really very
sinusoidal anymore. This can have a number of implications:
- If the "positive" half-cycles are chopped asymmetrically compared to the "negative" half-cycles, the AC power going to your device will effectively have a "DC bias". This can cause the core of your device's motor or power supply transformer to magnetically saturate, then draw excess current, and overheat. (I've never seen this happen on a dimmer at 100%, but I've seen melted transformers from being placed on dimmers at <100%.)
- If the half-cycle chopping is offset significantly far from the zero crossings, the transformer or motor back-EMF can cause high-voltage spikes to appear on the coils (which can damage the dimmer's TRIAC or SCRs), and in the case of transformers, also induce spikes on the secondary coils (which can damage your device's internal electronics). (I've never seen this happen on a dimmer at 100%, but I've seen (and repaired) dimmers that got damaged from this while running at <100%.)
- Also if the half-cycle chopping is offset significantly far from the zero crossings, the motor or transformer will start to behave like a choke and impede the current flow. For devices with linear power supplies, this can cause a "voltage sag" in the power rails inside your device (from which its internal electric/electronic components are powered). (This makes sense, because "chopping the wave far from the zero crossing" is exactly how dimmers reduce the power going to a lamp! Only in this case, the reduction is a little more pronounced.) For switch-mode power supplies (which have a feedback loop), this can cause the power supply to draw excess current in an attempt to maintain proper internal system voltage. And for motors, this can significantly drop the torque, which reduces the motor's intended "regulating" back-EMF, causing it to draw more current (even though the motor appears to have "less power").
"
Power factor corrected" devices try to minimize the complex part of the load
impedance, so they behave more like a simple resistive load. That being said, if your
bubble machine, fan, or whatever else you want to control has a
power factor of 1 (or very close to it), running it from a
dimmer that you can guarantee will stay at 100% (or 0%) is fine. Trouble is, few products list their
power factor in the specifications.
If you have access to an oscilloscope, you can clip it to the AC
power output from the
dimmer and check if the
voltage tracks a fairly "clean" sinusoid when you connect your
bubble machine, fan, whatever. (To do this safely and properly, you need an oscilloscope with at least 2 input channels, and set it to display the "difference" between the channels. Then use attenuating probes on both channels, specifically designed for measuring high voltages. Clip one probe to the AC Live, and the other probe to the AC
Neutral, and set the oscilloscope sweep oscillator to trigger either at the measured zero crossing (which always works), or synchronized to the oscilloscope's own AC
power (which works if you're powering it from an AC source at the same frequency as the
dimmer -- nearly always the case).
(There is a way to do it with single-channel oscillopes too, but it's dangerous if you don't know what you're doing and if anyone else is around who might touch things, so I'm not going to encourage it.))
Strange enough, if the
power is not nice and smooth, you might actually be able to "clean it up" and get the "chops" closer to the zero crossings by
turning down the
dimmer level slightly! (It doesn't seem intuitive -- more
power at less than 100%, but slight
power drop as you approach 100%? But depending on the
dimmer's triggering
circuit, and the load's complex
impedance, this can sometimes be the case! Especially since the trigger
system of many
dimmer circuits is effectively in series with the load, and therefore the V-I
phase offset of a complex load
impedance affects the phasing of the
dimmer's trigger!)
Interestingly, variable speed controls for ceiling fans and variable-speed
electric drills actually operate the exact same way as light dimmers do! How do they do it without damage? They limit the range of their trigger
point to keep it close to the start of the half-cycle, so the motor never
stalls. (Or as a more intuitive concept, they "only operate in a small upper range".) Their trigger
circuit is also not too-badly affected by the motor
inductance. And their
TRIAC is rated to handle voltages a fair
bit higher than nominal AC peak. Ceiling fan controls also initially turn on the
power at "100%", so the motor gets enough start-up torque before the operator will likely turn down the speed. (And in both cases, the motors are also designed so a little phase-chopping won't hurt them.)
So if you have a carpet dryer or fan or whatever which you want to run at variable speed, the answer may not necessarily be a definitive "no"! Look into its motor design, and the design of the
dimmer you intend to use, and how the two interact. If the device has separate control electronics with their own
power supply (and you don't mind voiding its warranty and modifying circuits), you can usually
disconnect that from the motor's supply and run it off a separate "clean" continuous AC
power line. (If there are heating elements though, you don't want to reduce the air flow over them when they're in use! Best bet is to
disconnect any heating elements if you're going to be monkeying around with variable air flow rates.)