I was trying to learn something about dimmers for light bulbs. After some reading, I found that most trailing edge dimmers contain some variant of this:

enter image description here

Found for example in this diy-project or this datasheet.

I do understand the general principle: No matter in which half wave we are, when the gate is pulled high, one of the MOSFETs lets current through to the other one which in turn lets the current flow because of its body diode.

But there are a couple of things I actually do not understand:

  • The signal could come from a µC, so there's no galvanic seperation between low DC and high AC voltage. Or is the insulation of the gate sufficient?

  • Even more confusing to me is the grounding between the MOSFETs. When the gates get pulled high, why does this not fry everything connected to that.

  • Why does nobody use an AC-optocoupler instead of this?


Both turn on when the gate voltage is high, one is shunting the body diode so neither body diode conducts (therefore the drop is proportional to Rds(on)*2). When blocking, one body diode cannot block so almost all the voltage appears across the other MOSFET.

The circuit driving the gates does not see much voltage relative to the “ground” at the junction, the trick here is to make a power supply referenced to that point and isolate signals to that circuit.

MOSFET SSRs could be used if they were fast enough, but they are typically very lethargic in comparison to the speed required- because they have to work with a small photovoltaic current to drive the relatively large gate charge.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.