@ThePhoton Q3 and q4 are supposed to work Schmitt trigger like to have a somewhat nicer curve of the LED lighting. I could probably use it as a low side switch, wanted to do it that way because the existing game has the light directly at the batteries negative

@PlasmaHH But... they're all emitter followers?

(not looking at any pictures atm)

@PlasmaHH min/max supply?

@Asmyldof was planning for LiPo like or three AAs, did the simulation with 4V

with the FET it probably works on either side

So 2.7V to 5V

yeah, thereabouts

Hmn, that FET forms a source follower, won't be awesome at 3V or below

gimme few minutes

no need for awesomeness, its a kids toy, haven't even selected the LED yet...

But with the same components, except a P-chan you can get actual hysteresis and suchness

yeah, with Ps I could probably also get the quiescent current down and components lower etc. but that wouldn't be hard to get, and I tried to go for the aesthetics of having all BJTs the same, and ideally all the resisotrs the same... maybe I will also try to get rid of the fet, dunno yet

@PlasmaHH Just the one P-mos

Maybe if I find one in the box that looks the same like those I will be using (BC547B probably, with gray on the flat side)

but for just now I have to figure out why ltspice gets abysmal slow when I introduce a base resistor for Q3

Alright, BC547

Starting point for further component equalisation

Plot at 5V ^^

Plot at 2.2V ^^

And no, I'm not going to figure out how to export from LT in colour, because I'm busy

isn't that colour? ^^

I will breadboard it at the weekend probably and have a look, since the fet does look different anyways it probably doesn't matter if it is P or N

@PlasmaHH Yes it does, but maybe not by that much

It'll muck with timing and hyst

Ad brightness

And*

with a pmos I wonder if one could make it soft power latch button like and put it into the front of everything to make it save power when off, but 60µA isn't too bad ^^