I have several (Cree LED bulbs) that I would like to dim via microcontroller. According to Cree the bulbs are dimmable with compatible dimmers. A local hardware store here in Canada has one of the listed dimmers in a convenient form factor (Lutron TTCL-100LH-WH-C). The dimming control in the particular product is a slider. The goal is to try to spoof the slider signal so it can be controlled remotely via microcontroller.

Does anyone have an experience how the dial controls of LED dimmers interface with the control circuitry? I'm hoping that the slider acts like a potentiometer and outputs a linear voltage to the control circuitry. This would mean that I can easily drive it with PWM + low pass filter or external DAC.

Hoping to get some feedback before cutting into one these modules. Has anyone done a teardown on these devices? Thanks.

  • \$\begingroup\$ The circuit in this kind of dimmer is just a few parts, is non-isolated from the mains and has high voltage across the rheostat. It is non-trivial to "spoof" that functionality- it's easier to start from scratch (also non-trivial, but easier). \$\endgroup\$ – Spehro Pefhany Jul 19 '15 at 3:53
  • \$\begingroup\$ Are you sure it's a rheostat? According to Cree's description of the part it uses a "leading phase cut edge" which to me indicates some sort of semi-intelligent or supervised control. \$\endgroup\$ – user25955 Jul 19 '15 at 3:56
  • \$\begingroup\$ That's fancytalk for cutting the leading edge of the sine wave, which is what the pot through the DIAC does. \$\endgroup\$ – Ignacio Vazquez-Abrams Jul 19 '15 at 4:03
  • \$\begingroup\$ Your question is about the Lutron product you saw at Lowes, no? That one has a rheostat (slide pot), dollars to doughnuts. It controls the firing angle of a triac phase control via a diac. \$\endgroup\$ – Spehro Pefhany Jul 19 '15 at 4:04
  • \$\begingroup\$ Instead of trying to modify the Lutron product, could I instead go DIY similiar to how Zerocross Tail does it? The schematic halfway down this page. I guess I'm wondering if I could use a generic incandescent approach to dimming LED bulbs. \$\endgroup\$ – user25955 Jul 19 '15 at 4:15

The drive circuitry in dimmable CFL/LED bulbs detects the "truncated" sine waveform in the supply line generated by classic DIAC-TRIAC dimmers and converts it to the appropriate mechanism for dimming the given light generation technology. So yes, generation of this modified sine waveform by a MCU in the supply will also cause these bulbs to dim.

| improve this answer | |
  • \$\begingroup\$ I prefer not to manipulate the output sine waveform but rather manipulate the control dial (in my case the slider) electronically. \$\endgroup\$ – user25955 Jul 19 '15 at 3:40
  • \$\begingroup\$ You don't generate the waveform directly, but rather trigger the triac at the appropriate point in the current half-period. See AVR182 for a (DANGEROUS) method to perform zero crossing detection to determine when to perform the triggering. \$\endgroup\$ – Ignacio Vazquez-Abrams Jul 19 '15 at 3:44
  • \$\begingroup\$ Not going to claim to be an expert on the subject, but I suspect that this works for incandescent bulbs but perhaps there's additional challenges with LEDs. There's dimmers that work for incandescent but not CFL or LED, so there must be a reason for that. I figure it would be easier just changing the control signal by intercepting the slider signal rather than trying to figure out the rest - not lazy, just don't want to damage my expensive LED bulbs or burn down my house :) \$\endgroup\$ – user25955 Jul 19 '15 at 3:48
  • \$\begingroup\$ The slider doesn't have a signal; it's a line-referenced pot connected to a DIAC and then a triac. \$\endgroup\$ – Ignacio Vazquez-Abrams Jul 19 '15 at 3:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy