2

Falstad does not have an optotriac part. I've opened a feature request issue on the CircuitJS GitHub to ask for one to be added, but is there a reasonable way to emulate an optotriac given the components and building blocks that are currently available in the simulator, in the meantime?

My intuition is that I could probably use a VCVS block attached to the triac gate, to "force" a particular voltage delta between the gate and one of the main terminals, but I'm not familiar enough with (opto)triac behaviour and models to have a good handle on what a good transfer function would look like and what other considerations I need to make.

I tried replicating a couple of old optotransistor-based triac drivers, but they behave quite differently to optotriac designs, and an optotriac is what I would end up using in a physical circuit anyway.

I'm not looking for deep physical accuracy - the behaviour only needs to be approximated closely enough to play around with some simple leading/trailing edge 50Hz 230Vrms mains dimmer circuits with ZCD. I could do this in LTspice, and I will end up using LTspice to more rigorously simulate anything I intend to actually build, but I much prefer the interactivity of Falstad for quick experimentation and prototyping. So, if there's a quick-and-dirty approach to approximating an optotriac, then it'd be strongly preferable.

Is my VCVS approach workable? Is there a better way?

Polynomial
  • 10,562
  • 5
  • 47
  • 88
  • [MOC308x](https://ltwiki.org/files/LTspiceIV/examples/LtSpicePlus/Opto/Optoacop/MOC308x%20OPTO%20TRIAC/) series has been modeled by Helmut Sennewald (our late LTspice guru). Does Falstad allow you to make custom parts? If so, the models look pretty generic. – qrk Mar 02 '23 at 01:59
  • If the Opto-Triac is also a ZCS you can simulate an analog SPST with a synchronous control signal of any R value. – Hoagie Mar 02 '23 at 02:08
  • @qrk As far as I know, the only custom parts it supports are subcircuits based on parts you can already create inside the application, so I can't just load an arbitrary model. – Polynomial Mar 02 '23 at 02:18
  • @Hoagie I'd strongly prefer the optotriac approximation itself to not behave as a zero-crossing type, since a large part of what I'm investigating is leading-edge and trailing-edge dimming behaviours, so it'd be very useful to be able to implement the ZCD separately. I suppose an analog switch _is_ a very crude approximation of a triac... might be a bit funky in terms of trying to roughly quantify the power dissipation though. – Polynomial Mar 02 '23 at 02:27

0 Answers0