Momentary switch power control

Is that from here http://www.edn.com/design/power-management/4427218/Latching-power-switch-uses-momentary-pushbutton ? That douche bag listed himself as a reference lol.

Yes, that's the source, but it doesn't seem to work.

Screw it. I can get two different types of latch switches working on my breadboard, but not on this damn simulator. I'm not quite sure what to tell you, buddy. Sorry. Maybe someone else has some ideas.

Actually i make it operatable and ths thingy reqire very precise tune up so changing state mechanism not collide with another performed operation simultaneously. A tranistors need to be perfectly tuned in values such as changng turn on voltage to 2 vo,ts from initial 0.4v cz such on voltages have almost all popular mosfets. Next is widening channel to increase current conduction to at least comparable to poor grade power mos. Those device here wth such channel and other params suggestnthat it is model of micro chip transistor rather a discrete one. Means a low power mosfet model that is created in small scale integration ic . Bjt also require shaping to match typical values of medium power transistor. Proper time frame set to make program not miss essential operation whose is need to make it work ..... devil is in details ';,;'

Okay, I've made one with just BJTs, but I can't get the cap to discharge enough to restart the cycle...even though the EXACT same circuit works in real life. If any one has any ideas...

With those similarities to real circuits i will be personally cautious. Bjt model do not have here variable capacitance presnt at junctions, avalanche breakdown and reverse negative resistance breakdown region also not exist (a negistor operation is not possible) also temperature based changes are not present as also internal noise voltage sources whose help any transistor smoothly start operation of any oscillator without need to shake his lazy semiconductor ass vigorously so it wake up. Oh i forgot to say that minority carriers storage effect is also on absence list. More in parameters we can adjust only few ones wnd 90% of rest parameters is unachivwble and set with fixed values. Inn place of person not aware how those effects influence device operation i will personally never ever think to beign able make real and smulated model behave in the same way unles circuit is simple like hell.

True. I'll give you that. This is a pretty simple circuit though lol. I've built similar circuits with 2N4401/3s, the one currently on my breadboard is with 2N3904/6s. I could probably make it work with TIP31s and 32s (actually, one TIP32A wouldn't be a bad idea to use as Q4 in my circuit if I planned on using it for medium power applications.) Anyhow, I get your point Secuture, I do, but I just wouldn't have thought this circuit would be complex enough to run into issues like that.

It's a tricky circuit whose hits porogram weakest point . A time related calculus reduction. When operting circuit present various activities whose some are performed very slow and some very fast and Both of them are vital for circuit operation then we have a big problem. When we set time frame for slower ones the fast ones are in too small timeframe and program simply aproximate it,s output function doing two calculations because it will be too far away in time to make a third try at all and we have a spiked triangle output for sine function for example. When we magnify time to make those better calculated we have another problem of waiting one day unles something else happens lol. In this circuit this weak point hits discharging of cap who must switch off mosfett. It happens too fast for program so it can catch what should happens exactly and leave transistor on like nothng happens.

Thank you both for this information and advice