2

I am hooking a wall wart to a plug on a dimmer switch which is hooked to some LEDs and a resistor to create a LED "lamp." What will happen when I dim the switch? Will it reduce the voltage? The current? Increase resistance? Generate more/less heat? What will it do? Are there typically voltage regulators that won't affect it?

I haven't picked out an adapter yet, but I want it to be fairly cheap ($3-8). Do different types of adapters have different effects? (I don't care if it dims exponentially. I just need it to dim.)

Anonymous Penguin
  • 3,750
  • 8
  • 35
  • 57
  • There are dimmers sold as "LED/CFL" dimmers, that ostensibly work differently. They are designed to work with LEDs and CFLs that are also designed to be dimmed. I'm not sure exactly how they are different, but it may be a direction for research. [How do the different kind of lamp dimmers work?](http://electronics.stackexchange.com/questions/34833/how-do-the-different-kind-of-lamp-dimmers-work) – Phil Frost Jul 01 '13 at 01:07

1 Answers1

4

Lamp dimmers change the brightness of light bulbs by delaying the turn-on of the output till later on the AC cycle as shown below.

enter image description here

This scheme works pretty good for resistive loads like incandescent light bulbs. It does not work very well with the cheap transformers found in the AC transformer type wall warts. If the transformer ends up being able to pass the chopped waveform then there will be a some reduction in the output voltage for non-regulated output type wall warts.

Virtually all of the newer type of universal type AC to DC switcher type wall warts will not be too upset with being fed a chopped type of AC input waveform. That said though most of these switcher types are designed with a regulator inside the wall wart that produces an certain output voltage. The switching duty cycle or frequency varies over a nominal range in relation to the ratio of the internal rectified DC voltage off the line and the desired output voltage. If you reduce the internal rectified voltage by using the dimmer technique this will drive the compliance of the switcher circuit outside its normal range. This could simply result in the output regulator just shutting off. On the other hand a different type of circuit may very well go unstable under these conditions and result in unpredictable output and possible overheating.

Michael Karas
  • 56,889
  • 3
  • 70
  • 138
  • So... bad idea? Are there any that would work with this idea? Capacitors somehow? I want it to be controllable from the wall. I know it can be done (look at this: http://www.amazon.com/Pharox-III-DIMMABLE-Watt-Light/dp/B002BZTG4I/ref=pd_sxp_f_pt/175-9658648-4393703), I just don't know how. I could just buy one, but who wants to pay $50 for an adapter, a resistor, a few LEDs, and a case? It seems to be marked up a ton. I guess some people will buy it still. – Anonymous Penguin Jun 30 '13 at 23:19
  • 1
    **-1.** Majority of wall warts today have a rectifier->flyback topology. The flyback works at few hundred kHz. The output voltage is controlled through feedback. More moving parts than the 60Hz transformer. Because of the feedback, the flyback wall-wart will maintain the output voltage as the input is being dimmed. When the RMS input voltage drops below the UVLO (under-voltage lockout) threshold, the flyback converter will shut off completely. – Nick Alexeev Jul 01 '13 at 00:34
  • @NickAlexeev that's exactly what the last paragraph says, as I read it. – Phil Frost Jul 01 '13 at 01:04
  • @PhilFrost Oups. Speed reading artifact. Now I need to figure out how to undo the -1. – Nick Alexeev Jul 01 '13 at 01:22
  • @NickAlexeev - Speed reading will keep you from knowing the rest of the story. :^) – Michael Karas Jul 01 '13 at 02:56
  • @Nick Alexeev One "hack" that used to wok is to make an edit, undo the vote and then roll back (or not, if the edit is actually useful or trivial). – AndrejaKo Jul 01 '13 at 05:49
  • I disagree that most switch mode supplies won't be upset about dimming. An SMPS has a rectifier, and current into a capacitive load is proportional to du/dt. What's du/dt when the TRIAC triggers? Experimental evidence suggests something around 5–10A even for small <20W supplies. RMS current won't be that much, but who knows what happens when you push 2300W spikes through a 20W design. Additionally, the dimmer loses the ground reference when feeding a non-linear load, which may contribute to other interesting effects. – ntoskrnl Jul 12 '14 at 21:02