I'm using a Winbond ISD chipcorder with the goal of creating the sort of circuits you see in greeting cards that emit a noise upon opening by using some sort of switch mechanism. The problem is, to begin playback with looping you need to make a transition from low to high on one of the pins, and I'm not sure how to accomplish this with just a switch.
Note that this circuit does not include an arduino, so I can't program the delay. Also, I can't use a temporary switch (the kind that switches back after an interval), since there are other parts of the circuit I need to power.
What I've come up so far is using a pulldown resistor on the pin, and then using the switch to connect the power line to the pin, with possibly a capacitor in between. My theory is that if I can delay the power signal via a capacitor, it will register as a high transition once it reaches the pin.
I'm not sure if I'm going in the right direction with the capacitor or if there's a better way. I also don't know what the appropriate capacitance value should be to delay the signal so it doesn't just override the low signal of the pulldown resistor from the beginning.
The chip has built-in debouncing on that pin, so that shouldn't be a problem. What do you recommend for a capacitance value, and how do you figure that out?
This means if you multiply the resistance in ohms and the capacitance in Farads you get time in seconds. One time constant it the time it takes to charge up roughly to 70% of the supply. It is exponential so by three time constants it is prity much all charged up.
The chip has built-in debouncing on that pin, so that shouldn't be a problem. What do you recommend for a capacitance value, and how do you figure that out?
Do you know at what voltage level, the chip will trigger? If you are trying to calculate capacitor/resistor then you will need to know that. Or you can just experiment with a few different capacitor/resistor combinations.
Thank you all for the suggestions. The RC constant seems like a good fit; however, I was actually able to get it to work by connecting power to the pin using a resistor and capacitor in series. For reference, the resistor is 200 Ohms and the capacitor is .1uF. Does anyone know how this worked? The pull-down resistor I attached to the pin ensured that it was low as soon as the circuit was connected. Since capacitors don't transmit DC charge, I'm not sure where the transition to high came from.
I was actually able to get it to work by connecting power to the pin using a resistor and capacitor in series.
That is the circuit you can use when you calculate your time constant.
Since capacitors don't transmit DC charge, I'm not sure where the transition to high came from.
However capacitors do transmit change, so you change the voltage on one end of the capacitor and the voltage on the other end of the capacitor instantly changes to keep the previous voltage difference between the to leads of the capacitor the same. After that the capacitor will charge or discharge depending on what DC voltages it sees.
My teacher at Uni said, that if someone puts an RC on a digital signal, one should be executed immediately. He is right. If you need to delay it a few nanosec, than put there inverter pairs, if more, put there pre-made ICs. Dont put RC.
I am not sure he had it right, as to who should be executed immediately. Just kidding, LOL.
If you have constructive suggestions, they are more than welcome.
I don't see the problem with using RC. There's no "right way" or "wrong way" -- there are basic electronics principles that, when observed, either achieve the desired effect, or don't.
It's possible some particular method has side effects that are undesirable, but an engineer's sole task is to pick a suitable solution from the available compromises...
(BTW, the 555 timer uses an RC to do the timing. Along with an input with hysteresis, so the Schmidt trigger was a good suggestion too. But we're talking about a greeting card thingie here... Good enough is.)
The circuit I built did not exploit the RC constant, since the capacitor was in series with Vout, instead of connected to ground in parallel. The Schmidt Trigger is good idea, but as SirNickity said, it's not totally necessary given the limited scope of the project. I'm content with how it currently works, albeit I'm not 100% sure how it works. Thanks for all your help