I'm embarrassed to say I don't know how to do this. I guess I've been away from transistors too long. Well, I know how to do it with a relay, but would prefer to stay away from mechanical stuff and pitted contacts if I can.
The project is a personal project - a button pusher for the remote controller of a camcorder (which also takes still pictures). I'll use it to do time-lapse still photography, with an interval of from
5 seconds to one minute or more..The button pusher itself is a 12V solenoid, with a DC resistance of about 100 ohms. The timing circuit is powered by +6VDC provided by four C cells. Since the solenoid really does need a full 12V to pull in properly against the resistance of the button, I've concluded that a typical voltage doubler won't quite work - because of the diode drops, etc.
But as luck would have it, I happen to have a MAX232 chip in the junque box, and it produces +/-10V outputs from the 6V supply. And I've found that if I connect a 220 ufd capacitor across the +10 and
-10 outputs, the MAX232 will charge it up to near the full 20V in about 3 seconds, which is quick enough. Then, I connect the solenoid across the capacitor, which then discharges suddenly, and that's enough get the solenoid to pull in and push the button.
But now I need to get the timing circuit to make that connection, driven from the Output pin (pin 3) of a 555 timer, which switches between nearly +6VDC and 0V. To complicate things further, as the capacitor discharges, the voltages on its leads with respect ot gound rapidly collapse from +/-12V, and end up, oddly enough, at about -4.5VDC.
I tried using two transistors to do this. The 555 Output drives the gate of an NPN, with the emitter grounded. The collector of the NPN sinks the base of a PNP, with the emitter connected to +12V, and the collector connected through the solenoid to the -12V lead of the capacitor. And that kinda works. Of course it only lets the cap discharge to ground (because the emitter of the NPN is at ground). So the button push is really short, and I'm not sure it's going to be long enough to be reliable.
I had thought about triggering an SCR gate through a capacitor, but the current never actually stops - the MAX232 still produces juice, so turning the SCR off might be a problem.
I had also though about driving an NPN base through a capacitor, but am concerned about the rapidly rising voltage on the emitter as the big cap discharges. I think this would cut off the current flow early, and possibly damage the 555.
Well, it may just be that a relay is the way to go. I even have one in the junque box. But if anyone knows of an elegant-but-cheap, non-relay solution for this, please post.
Thanks very much for any help.