555 relay switch that triggers when it's not suppose to

[gore]

Howdy, folks of the engineering community! I'm having some trouble with this little circuit. It's a simple 555 based relay switch. There's a momentary SPSP switch (*JP1) to trigger the on/off mechanism. SPDT switch (*SW) to set the ON or OFF state of the relay on power on. And a power supply terminal block (*JP2) with a polarity protection diode. It's all there in the schematic bellow.

So here's the problem. Basically it works, but it often triggers with a mere touch of a screwdriver on the contacts of *JP1, for example. Or touching some of the relay outputs does the same thing. It's trigger happy! I need to tone it down a little. It's only suppose to be doing the switching when I press down the momentary switch (*JP1). Why is it so sensitive?

[w2aew]

Possibly small static charge or noise injection when touching those various nodes.  Consider the following:

Add power supply decoupling.
Consider lowering the value to the two 10K resistors on the THR/TRG line.
Consider adding a small cap 0.1uF or so to the THR/TRG line.

[gore]

Consider adding a small cap 0.1uF or so to the THR/TRG line.

Thank you for suggesting that. That seems to do the trick. I've also tried an isolated power supply (a battery), as opposed to a bench power supply. That works too.

[Gyro]

Yes, definitely a 100n capacitor to ground on the THR/TRG line, and maybe a series resistor before it too (1k?), otherwise it's effectively an open antenna for spikes.

Edit: Does that capacitor on the CON pin work reliably for defining the power-up state? It might be rather sensitive to slow power supply rise time. The more usual method would be an RC network on the Reset pin.

[gore]

Does that capacitor on the CON pin work reliably for defining the power-up state? It might be rather sensitive to slow power supply rise time. The more usual method would be an RC network on the Reset pin.

You are correct. It's not reliable. I've tested it on a bench power supply by switching the channel on and off. It won't define the power up state. Physically removing and reattaching a wire works.

[Pjotr]

Have build numerous of this circuit reliably. It's by itself quite elegant. For a normal bipolar NE555 you must decouple the power with at least a 100nF ceramic close to the pins. When switching the thing draws a large cross over current shortly. For the switch use simply screened cable with the screen grounded. Rising R4 to 220K gives a larger debounce/noise margin.

I wouldn't use this reset mechanism with relay contacts. The cont input is noise sensitive and it can also be triggered from noise from the relay then. Better hold the trh/trig input high at power on this way: