Ideas for brief one-shot connection-making circuit needed

[SearchAndRescue]

I recently came across a See-n-say Story Maker (I'll get a pic up soon). Was bending it with the old spitty-finger method and found a crash point. What I think was happening was a very brief short circuit between the opposite sides of the clock crystal which threw the logic circuits off and made the thing crash in cool ways ("yickity-yackity-gribble-gromp HISSSS!" and 't...h..e..c.ow ... mmmmmooooooooOOOOOOOOO!!!!!!!! (pitch and volume rising eerily gradually)" being some of the better examples) . I'm wondering if anybody had any ideas about how to make a one-shot, extremely brief connection between the two leads. Also, I'm wondering if my spit was conducting the electricity differently than wires would? What is the capacitance of human saliva, anyway?

I've been circuit bending for a while, but i've never gotten into more theory based, solid state electronics. I've been more comfortable in the camp of "drill a ton of holes in the front, wire up a ghetto bolt patch bay, go party". My idea for getting crash control is pretty crude, and I'm wondering if it would even be effective. But here it is:

I'm thinking I could hack open a toy camera for a big electrolytic cap ( I know it's dangerous), which I would connect to the toy's power supply through a N.O. pushbutton. When the button is depressed, the charge would build in the capacitor, until it discharged through a relay which would briefly connect the leads of the crystal (perhaps thru a small-value resistor or cap to prevent full voltage contact?) When the cap discharged, the relay would switch back to open, and the toy would resume at normal clock speed and abnormal operation. The discharge from the cap to the relay would then go through a high-value resistor to the toy's ground. Would this voltage across the common ground cause problems for the circuit? I'm also thinking the high-voltage discharge unshielded from the audio circuit could be noisy (just amplify the whine of the charging cap and then POP! when it discharged ?), although perhaps it would be cool, too (?).

If anybody has any ideas about how to make this more elegant (and less dangerous), i'd appreciate it. Thanks!

[phantompowers]

Can't you just use a momentary on switch?

[Circuitbenders]

i agree, if i'm understanding you correctly that does seem like an exceptionally technical way of going about doing something that you could do just as easily with a non latching push to make button. Just connect it across the points and give it a tap. If the effect is too severe try sticking some resistance in line.

[SearchAndRescue]

I'll give these suggestions a try, but I'm thinking that the amount of time that my finger would be depressing a regular N.O. button would still be too long. I'm thinking that the clock signal has to be interrupted for only milli/micro-seconds, since it's going at such a high frequency compared to my finger's ability to hit the switch. I was only able to get the thing to crash the right way a few times out of a hundred. Most of the time, it just lost functionality with a little speaker-pop, and I had to disconnect the battery to get it opperational again. Anybody have ideas about an IC chip that would give me the ability to make very brief connections? The real problem for me is that i am sampling these bends with an Akai sampler (an old one, too, cumbersome and hard to edit using a 16X2 character lcd screen), and i don't want to waste my time making hundreds of attempts to get one good sample.

[Bogus Noise]

I'll give these suggestions a try, but I'm thinking that the amount of time that my finger would be depressing a regular N.O. button would still be too long. I'm thinking that the clock signal has to be interrupted for only milli/micro-seconds, since it's going at such a high frequency compared to my finger's ability to hit the switch. I was only able to get the thing to crash the right way a few times out of a hundred. Most of the time, it just lost functionality with a little speaker-pop, and I had to disconnect the battery to get it opperational again. Anybody have ideas about an IC chip that would give me the ability to make very brief connections? The real problem for me is that i am sampling these bends with an Akai sampler (an old one, too, cumbersome and hard to edit using a 16X2 character lcd screen), and i don't want to waste my time making hundreds of attempts to get one good sample.

Have you tried putting some resistance in between the button and the glitch points? Test around with some pots, sometimes you'll find that it'll be more stable when it's not a full-on short circuit. Sometimes this'll be variable as well, so you can use the pot for a 'Glitch Amount' knob and the button to trigger it. Also, they're always worth testing in conjunction with a voltage drop knob to see if there's an interplay between the two. (Casio SAx style, basically )

Another tip - solder a push to break button in between the red or black battery wire to act as a reset, saves you from taking the batteries out each time it goes too far!



And while I'm at it, for getting the noises into your Akai - how about recording into a computer wave editor a full session playing about with it, slice out the best bits, and THEN use then sample them in? Best of both worlds then.

[Circuitbenders]

i don't want to waste my time making hundreds of attempts to get one good sample.

You're probably in the wrong place then