Transistor latching circuit

[Tony Mahon]

Hi Folks. Thanks for including me.
My first time here so please be patient and gentle  . I'm trying to create a latching circuit using transistors and its driving me nuts! The idea of the circuit is to replace the electronics in a kettle. This latch is intended to supply power to a thermo switch using the original thermistor to turn off a comparator latch using a 555 timer. The kettle needs power after the 555 turns off as I'll be using pin 7 to drive a beeper to indicate when the kettle has boiled and turned off. The latch will drop out as soon as the kettle is lifted off its base and loses power. The 555 will power the original circuit's relay via an existing transistor (the led is just for development purposes. Driving the led, the 555 draws 22mA). I've got the 555 part sorted and works well, it's just this latch that's giving me strife, I'm this close to saying 'damn it', and using a dpdt relay to make the latch. As per the attached drawing, it's a really simple circuit, but the led lights as soon as power is applied and stays lit until it is removed. I even tried using a cap to ensure that the base of BD140 is initially high, insuring the tyranny is turned off. On my first try, I used 2 x 557's with no luck, so I tried using the 140 , thinking that a tranny with less gain might fix the issue. I've seen variations of this circuit and they seem to work. Can someone please point me in the right direction, and help me regain my sanity.
Many thanks.

P.S. The resistor I'm driving the led with is actually 1k.

[kkontak]

BC557 is a PNP transistor, not NPN.
Use a BC547 instead.

[Zero999]

Yes, you need an NPN transistor, as mentioned above.

I'm interested how you've implemented the rest of the circuit. Please post a complete schematic.

It's possible the whole thing can be done with comparators or a single microcontroller IC.

[Terry Bites]

Look up thyristor therory and how you can emulate one with a pair of transistors. It latches when a gate current is applied and unlatches when the load or power supply is disconnected.
How easy is that? https://www.youtube.com/watch?app=desktop&v=lrPiu0quods

[Tony Mahon]

Thanks Terry, I do know how a thyristor works, but this circuit needs to be triggered by a negative going pulse whilst switching the positive supply. Thanks for your time and effort. Trying to emulate a thyristor with these requirements is what got me in this situation. Cheers mate.

[Tony Mahon]

Thanks kkontac, I'll try using a 547 instead of a pnp
 Btw, this circuit is basically a ripoff of one of Dave's soft start single button on/off sw published a few years ago. He used 2 pnp's and got away with it. Thanks for your reply. Cheers.

[Tony Mahon]

Thanks zero999, I'll try using an non instead. This circuit is ripped off from the front end of one of Dave's soft start on/off circuit, so I thought I'd be able to use it too. I'll scribble out the rest of the circuit in a presentable state and post it on here. Thanks for your time and effort
 Cheers.

[Tony Mahon]

Please check out Dave's video here. This is where I got my idea from.

https://youtu.be/Foc9R0dC2iI?si=HrH2y0OU9r8tywcP

[pqass]

The circuit works but the PNP transistor base resistor is likely too high in value.
A BD140 has low hFE of just 25.  So (with 10K) it can only support at most 28mA ((12V-0.65V)/10K*25) load; you did say there was a relay in the mix.
I'd also lower the base to emitter pull-up resistor as 100K may not be strong enough to keep the PNP off on startup.

See this simulation. By lowering the base resistor to 1K it can support up to 280mA load. 
See for yourself by using your mouse scroll wheel over the LED resistor to lower it thus increasing the load beyond 250mA and see the voltage collapse.

It may not matter to your application but a P-MOSFET instead of the PNP would drop less Vce and consume less power in the base resistor (10K).

PS. Looking at Dave's circuit... using 100K resistors work there because he's using a P-MOSFET (vs your PNP) which need orders of magnitude less current to keep off or turn on.

[Tony Mahon]

Thanks very much pqass, this kind of guidance is exactly what I was looking for. I was certain that the circuit should work, but the devil is in the details. BTW, the relay is driven by the original onboard transistor, so the 555 will only be supplying the base current for that. I'll get stuck into the circuit following your advice as soon as I get home. Greatly appreciated.
Cheers 👍