EEWeb_XformerlessPS02 UNLISTED

It might help if you didn't think of this as a production ready circuit. It's merely to illustrate a concept for a different forum (EEWeb). The output terminals are the two load resistors (Rload12 and Rload5). But, one is not always interested in "outputs" when running a simulation. I was more interested in what was going on at the 32VMax point. Regarding the number of ground points, I'm very much a beginner, too -- so my question is, what is the significance of the number of ground points?

Also, notice that this is Circuit is unlisted. The reason this is unlisted and not public is I was only interested in comments regarding my failure to achieve a meaningful simulation. In other words, I would only have made it public if it was a completed design.

And, and even other way of putting this is: this offering was framed in an intended context: the context of wanting help with the simulation of an uncompleted circuit -- not the context of inviting critique of the design.

@peterlonz: I took a look at your profile and as of this writing I see one circuit -- a transformerless LED driver that is coincidentally very similar to a circuit presented by someone else at the EEWeb forum I designed my TPS for. Whacky synchronicity! In fact, the circuit on this page is a refinement of that circuit to conform, more, to the specifications given by the forum poster.

BTW: your circuit excludes comment, but in case you didn't know, you need at least one ground in your circuit in order for it to simulate correctly. Also, it's clever little circuit -- I learned something from it! At first I thought the 1N4148 diodes would fail, because their PRV is 100V. But I failed to consider that the combination of the series capacitance and the current drawn by the LEDs, will limit the reverse voltage to below 100V. I realized that, after running a Time Domain simulation on your circuit (following the addition of a ground ;)

But, I'm wondering why R1 is so low in value. It's purpose is to discharge C1 when power is removed -- not to contribute significant current to drive the LEDs, right? With it at 470Ω, the peak current through the LEDs is 400ma. Did you mean for that to be 470K (more in the range that I would use)?

One reason TPS's don't work (at least in terms of what I assume you mean by "doesn't work") is because the current through the series capacitor is not uniformly bi-directional. In other words, the current through the series capacitor must alternate and be the same in both directions, otherwise the cap will charge more in one direction than the other and thus the balance of the current will be less than required.