TLC556 Weirdness

[SirNick]

I'm prototyping a reset controller for a micro using a CMOS 556 timer IC.  The first half is a monostable pulse generator, designed to emit a short HIGH pulse when the reset button is pressed.  This resets the uC, but if the user continues holding down the reset button, the uC will resume boot-up and can then read the state of the reset button to perform some action if it's being held down.  (Typical press to reset, hold to reset to defaults behavior.)

The problem is, instead of a pulse, the output remains HIGH as long as the reset button is held down.

Attached: The schematics (from a simulator that shows proper behavior), and the PCB layout with a TI TLC556.  The left half is the monostable part.  Pin 1, 2 are Disch, Thresh, 3 is Control, 4 is Reset, 5 is Out, 6 is Trigger, 7 is Gnd, 14 is Vcc, 8-13 are the other half.

I've measured pin 1/2, and see the cap charging to 3.3v when the button is pressed.  It falls to 0v when the button is released.  Control is 2.20v.  Out is 0v until the button is pressed, then it goes to 3.3v and stays there, only returning to 0v when the button is released.  The Out pin goes to the input of a logic IC to OR with a PGood signal from the regulator -- so no appreciable load or anything.

It looks to me like everything's correct.  Am I missing something?  Is there something particular with the CMOS 555/6 that I'm not taking into account?

[mij59]

Hi,

The trigger input needs to rise above 1/3 VCC for the timer to work.
Add a capacitor and a pull up resister to the cicuit.
Why is there a 47K resistor on the rst pin ?

[SirNick]

There is a pull-up.  The switch has one pin pulled to Vcc through a 47k resistor (by the switch footprint -- and also labeled "47k Reset", which is admittedly ambiguous.)  This is the pin attached to Trigger.  The other side of the switch is grounded when pressed.

The 556 Reset pins are connected to Vcc through 47k because, during the design phase, I had a fleeting notion to disable the timer on some external event.  That is no longer the case, and they are only still there because I neglected to remove them before creating the prototype PCB.  I put them in the sim / schematic for continuity, since they're on the PCB layout.  (After all, virtual resistors are free and consume no power.)

[mij59]

There is a pull-up.  The switch has one pin pulled to Vcc through a 47k resistor (by the switch footprint -- and also labeled "47k Reset", which is admittedly ambiguous.)  This is the pin attached to Trigger.  The other side of the switch is grounded when pressed.

The 556 Reset pins are connected to Vcc through 47k because, during the design phase, I had a fleeting notion to disable the timer on some external event.  That is no longer the case, and they are only still there because I neglected to remove them before creating the prototype PCB.  I put them in the sim / schematic for continuity, since they're on the PCB layout.  (After all, virtual resistors are free and consume no power.)

Take an other pull up resistor and connect it to the switch, connect a capacitor between the switch and the trigger input.

[tautech]

Does it have to be with a 556?
I would do it with a timed one shot using a 4013
http://www.discovercircuits.com/DJ-Circuits/dflip1shots.htm

[SirNick]

Take an other pull up resistor and connect it to the switch, connect a capacitor between the switch and the trigger input.

Well, according to my breadboard, you're absolutely right.  I never realized the trigger signal had to be a pulse for it to work.  (The Falstad sim does not behave this way either.)  I was under the impression you could hold the trigger LOW indefinitely.  I guess not.  *sigh*  Bested by a 555.  Not my proudest moment.   

Well, thanks.  I guess I know what I'm going to put in the space reclaimed by removing the reset pull-ups.

[mij59]

Take an other pull up resistor and connect it to the switch, connect a capacitor between the switch and the trigger input.

Well, according to my breadboard, you're absolutely right.  I never realized the trigger signal had to be a pulse for it to work.  (The Falstad sim does not behave this way either.)  I was under the impression you could hold the trigger LOW indefinitely.  I guess not.  *sigh*  Bested by a 555.  Not my proudest moment.   

Well, thanks.  I guess I know what I'm going to put in the space reclaimed by removing the reset pull-ups.

Call it a light bulb moment, and yes the future is bright!