emerged in response to a latest local blackout event + a realization that the 77% of the energy would go "smoking in to the air" on the current limiting resistor e.g. the light would be ON some 3.3x less than otherwise . . . (but in the case of no (any) light even ←that would do ...)
1-st ) -- the step-down¹ with the separate regulation of voltage and current -- a case study ::
- ↑ yields the 39% (and worse ...) efficiency for given setup ↑
- ↓ has an unwanted startup overload spike ↓
(might be removed by delaying/modifying the 555 startup)
2-nd ) -- the step-down² using the proven (Hi-Power) switcher variant (@ Lo-Power) -- gives a slight step in efficiency but requires adding the output load or redesigning controls (such as using the lower power mos-Fet or Bjt ) --and/or-- adjusting the inductor ~ switching-frequency -- to get the efficiency higher
3-rd ) -- the step-up.. (← in my experience has
statistically shown to have
greater efficiency than the step-down conversion) ..using a commercial switching controller -- has a most promising(/preferred) result here . . . considering i didn't spent much time on fine tuning the following setup -- it does however have an unwanted startup over-spike (that may damage the LED-s on a "wrong day" or in case of too frequent power cycling)
SUM ) -- two last variants compared
PS ! ) -- as the title says -- !!! all these designs are functionally incomplete !!! -- e.g. :: not set to handle various specific situations , such as : too high input , too low input , output short , reverse polarity , switch/inductor failure , etc. ...
+Plus ) -- Couple of things to compare the prev. against . . .
1 ) -- an experimental converter (is lacking the control module -or- a concept for)
2 ) -- basic , simple -- linear regulator
3 ) -- basic , simple -- Op-Amp v. PNP versus NPN input stage
4 ) -- basic , simple -- Op-Amp v. PNP input stage (target LM324 or other Quadruple Op-Amp)
5 ) -- basic , simple -- linear regulator (for target input of 8x AAA to aged or not full 7A·h 12V PbSO₄)
[Eop]
