Electronic dimmer as Variac

It is very doubtful. A dimmer needs a minimum current load to keep it conducting the remainder of each half cycle after it is fired. Resistive loads are ideal, and average the resultant waveform quite well. SMPS usually rectify the line waveform and apply that directly to an energy storage capacitor that charges up to the peak waveform voltage each half cycle (and does that during a small part of the cycle, near the peak).

When your dimmer fires, the rectifier and cap will look like a near short circuit to a voltage that is something less than the line voltage at that instant (whatever voltage remains on the cap from the previous half cycle). And the moment the line voltage starts to go down, the rectifier will turn off, isolating the cap from the dimmer, causing the dimmer current to fall below its holding current, so it will turn off. In other words, you will probably not get the smooth control you are needing (any setting that includes the peak of each half cycle will produce the same result) and and may damage the dimmer and/or the rectifier and/or the capacitor with the big pulses of current that occur as the dimmer switches on.

Sometimes you need the right tool.

My immediate response is don't do this. The SMPS will pull large amounts of current as the voltage approaches peak (rectifier from power line charging main capacitor), and it may exceed the peak current rating of the dimmer, even accounting for the fact that light filaments pull a large amount of current when cold (at least they heat up rather quickly and pull less current). My advice is (with all the caps dischharged) double-check all the rectifiers and other 'main' semiconductors (such as the main flyback transistor), then stand back and give it full power. But then I'm posting from SED and don't do much repair. If you don't want me steering you wrong, don't crosspost outside the *.repair group.

I tried to use an SSR once to power a SMPS. I think I went through about 3 SSRs before I figured out that you can't do that. And that was just on/off!

so as

the

as a

double-duty

feasible?

Read the instructions that came with the dimmer. It probably says that you're only supposed to use it with a resistive load such as a light bulb. It'll probably burnout if used on a PS.

"DaveC" skrev i en meddelelse news: snipped-for-privacy@news.individual.net...

Likely not, use the light bulb trick instead - I prefer a Toaster in series mysel just to save the fuses.

A SMPS will, if it is a decent design, have a threshold where the supply will start - so the first 1/3 of the input voltage range it will be off anyway and ramping up the voltage does not buy you anything other than saving a fuse: It will start suddently by itself.

A SMPS is usually a capacitive load, the dimmer is designed for an Ohmic load i.e. a light bulb, and the TRIAC in the dimmer will not appreciate piring directly into a capacitor that will look like a dead short. They you can stick an inductor in series - and you might get at resonant overvoltage blowing the supply/dimmer away.

IOW: It creates more trouble than the problem you started with.

And if the OP uses the dimmer to control a light bulb, and connect the SMPS in parallel with the bulb? Will it work?

OK, OK, I think I get the idea. I had no plans to try it unless others here thought it might work. Guess not...

So it's a light bulb (or toaster) in series.

Thanks for lending me your minds,

A divider made of two light sockets wired in series may be helpful, also. Connect the supply across the bulb on the neutral side of the divider. If the two bulbs are the same wattage, the voltage is limited to no more than half the line even if the supply does not draw current. It takes little time to screw in various bulb combinations.

A lot of the nastiness can be reduced by putting an appropriate resistor in series with the dimmer. It's got to be high peak wattage so that it can take series connection between the (effectively) peak mains voltage, and zero of an uncharged cap.

If you pick the R so that the RC filter formed is some 5ms or so, then that'll probably be around the right number.

kHx.

It might just work if you use one resistive load (Say, a 25 to 40 watt light bulb) as a minimum load and put a second resistor (say, a 40 to

double-duty

kHx.

I assume that the first one is to keep the triac conducting. Why? If it's going to a rectifier/cap, then how does it matter? When the current goes to zero, the rectifier turns off anywya.

The phase timing circuit in many dimmers only functions properly if there is a resistive load on the dimmer. The timing current passes through the load.

The simple answer is that if the dimmer sees mostly a resistive load, it will work reasonably well and survive. This *can* include things like shaded pole motor-based fans (but probably not other types of induction motors or universal motors). However, if it's driving a lot of capacitance or inducatance, at the very least it will behave strangely and more likely will self destruct or damage the equipment. There's also usually a minimum load below which it won't do anything predictable. In short, get a proper Variac. You know my motto: "You can never have too many Variacs!". :)

--- sam | Sci.Electronics.Repair FAQ Home Page:

Important: Anything sent to the email address in the message header is ignored. To contact me, please use the feedback form on the S.E.R FAQ Web sites.

as

double-duty

kHx.

This won't work, a light dimmer doesn't vary the voltage, it varies the duty cycle, so if you plug a switching supply into one you'll fry it.