# Time Delay Relay: output = HIGH until triggered, then return HIGH after interval?

My goal is for a button press to temporarily disrupt the input voltage (120VAC) for a specified interval (around 10 seconds). Ideally, I'd like the interruption of voltage to coincide with the leading edge of the button press; I'm flexible about whether the delay would start at the leading or trailing edge.

I'm getting a headache trying to determine the proper configuration of components to achieve this. I imagine I will need some type of time delay relay in tandem with some type of momentary push button (normally on? normally off?). I've tracked down a few resources that summarize the various timing modes available with TDRs, but nothing seems to offer the functionality I describe above.

I'm a novice tinkerer, and it would not surprise me to learn I'm missing something quite obvious. Any help greatly appreciated!

Edit: If at all possible, I would like to avoid dependence on a second DC source, such as is required in the circuits Samuel and miceuz have proposed. Perhaps there's no getting around this? In lieu of introducing an entirely separate voltage source, would anybody condone splicing into the 120 VAC supply, and then using a transformer to step down to the ~12VDC involved in the 555 / RC timer circuits?

• Do you have a dc power supply that can be used as the power for the basic switch latch and timeout circuit? Commented Aug 29, 2013 at 22:55
• "Monostable multivibrator" Commented Aug 29, 2013 at 22:59
• Good question, Andy. See edit :) Commented Aug 30, 2013 at 14:57
• well, stepping voltage down via transformer, rectifying and regulating - that's what is a DC power supply - a transformer, 4 diodes, a cap, LM7812 and anoter cap. Commented Aug 30, 2013 at 15:56

Use a 555 timer in monostable mode. This can be triggered with a push button on the leading edge. The output of the 555 should control a mosfet which supplies current to a normally closed relay (the 555 will not be able to drive the relay as it is). When the relay is activated it will open the contacts on the relay, disrupting the connected 120VAC. Note that the 555 timer, mosfet, and coil side of the relay will need to be powered by a separate DC source.

simulate this circuit – Schematic created using CircuitLab

Your relay doesn't need to be a SPDT, you can find a SPST that's normally closed. Also depending on the relay you can find, you may need to include a flyback diode. Use one of the many calculators online to find proper values for R1, R2 and C1.

• I thought somebody might suggest using a 555. Could I, perhaps, power this circuit using a transformer in conjunction with an offshoot of the AC line? Commented Aug 30, 2013 at 14:54
• That's exactly how I'd do it. Use a wall-wart to power the low voltage electronics. You can imagine this is how many in-line switches work, for instance, electronically controlled holiday-light timers. Commented Aug 30, 2013 at 15:57

You can get away without using a 555 timer - a capacitor and a resistor is enough.

When you press a button, C1 is charged to supply voltage, this makes voltage on the gate of the mosfet Q1 to go over threshold and the mosfet turns on. Then charge from C1 leaks thru R1 until voltage on Q1 gate goes below threshold.

The exact values for C1 and R1 will depend on your supply voltage, mosfet turn on threshold voltage and minimum relay holding current. You can experiment starting with 22uF capacitor and 100k resistor. To increase the interval increase either or both of the values.

If you want to do some math, go to this tutorial. You would have to integrate the discharge curve from your supply voltage to mosfet threshold voltage to get a delay time.

This circuit will not be as precise as 555 one, but depending on your requirements it might be good enough quick and dirty solution.

simulate this circuit – Schematic created using CircuitLab

• Precise timing is a non-issue, so this would work for me just as well as a 555 circuit. Please see the above edit concerning the issue of introducing a separate voltage source :) Commented Aug 30, 2013 at 15:02
• if you are comfortable to work with HIGH VOLTAGES, you could use a transformerless power supply - by dropping voltage over a series capacitor, rectifying it and regulating over a zener. but this raises more requirements for the system - proper grounding and a button providing sufficient isolation should be the first on the list. check this: ww1.microchip.com/downloads/en/appnotes/00954a.pdf‎ Commented Aug 30, 2013 at 15:51