Step 1: The circuit
I wanted one. A huge lumbering monster of a robot which will come to life when a red illuminated button is pressed and start smashing everything in sight.
So I decided to build one. The button to control the robot, I mean. Since there was no robot to control it will have to control something profoundly significant and what better use than to use it to initiate the end of the world?
OK. For now, it only makes a noise, but I guess you will get the general idea.
It is very difficult to design a circuit like this. Almost impossible. Doctor Johnson was forced one evening to endure a piano recital by the hostess. This was before dinner and he was hungry, so he could not just walk out. At the end of it he was rudely awakened by the polite applause of those guests who had remained awake.
"You know, Doctor," the hostess gently rebuked him, "that was a most difficult piece."
"Difficult, Madam?" was his reply. "I wish it had been IMPOSSIBLE."
Now this is that sort of impossibility. But it was made possible by the advanced circuit simulation and design software, (SP)ICE 7.3.01 Release IV (gamma) which instantly freezes every computer on the network and blows the main fuse, so that further work has to be done on paper by candlelight. Which explains the general scarcity of components in the circuit.
Switch S1 controls the power to the circuit. In the 'off' position any charge on capacitors C1 and C2 are removed via diodes D1 and D2.
When S1 is switched on, if S2 is in the resting position (shown) C1 charges to the supply voltage of 5 volts. When S2 is operated, if S3 is in the resting position (also as shown) its charge is redistributed with C2, and since their capacitances are equal they both get charged to half that voltage, around 2.5 volts. Then if S3 is changed over, one contact of the pushbutton gets this voltage. Now pushing this button will apply a current to the gate of the SCR via R1 and it turns on. R2 prevents it from turning on due to transient pickup on the gate lead. Even after the gate drive is removed due to capacitor C2 running out of charge the SCR will remain 'on' and it will only turn off when switch S1 is turned off.
I did (could) not design a PCB for this circuit because for one, the computers had all frozen, and second, any potential collaborator I approached was scared stiff. This world was dear to them, and they did not want anything to do with the process of initiating the end of it. Even remotely.
So I had to go it solo. I am very proud of the achievement, prouder than the injun who has just outwitted a couple of cowboys.