So I came up with a way of doing this using a curious network of resistors. I do understand that this would most certainly be done more effectively/reliably using a MOSFET or some such, but as a sheer n00b I'd just like to know if this harebrained idea can actually be improved upon to be used as a practical circuit construct. Anyway, the main drawbacks and limitations I can see from my not-so-lofty position:
- Proper operation of the circuit seems to demand that the load resistances not fluctuate too much. Otherwise, things get out of balance and charges start flowing in undesirable ways. I suspect that isn't easily remedied and as such the usefulness of this sort of circuit may be limited to small current loads, and moreover that both loads may be required to have roughly the same internal resistance. True?
- While ensuring that the branch connected to the lower LED is completely de-energised seems a trivial task, doing the same for the other branch appears quite tricky (notice that the upper LED is slightly reverse-biased). Also not quite sure how to calculate the appropriate combination of resistor values in order to supply a chosen voltage to each load. Any thoughts on how that might be done?
Last edited: