dead-man circuit to protect against computer controller freezes and relay staying on

Status
Not open for further replies.

quantboy

Newbie level 5
Joined
Oct 10, 2012
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,380
For a project, I need to design a circuit that will make sure that a relay (connected to a load) will not get stuck "on" if the computer that controls it freezes (with the input to the circuit being either high or low). I think I recall this being called a dead man switch.

COMPUTER ----> DEAD MAN CIRCUIT W RELAY ------> LOAD

The best thing I could come up with is a pulse generated by the computer that must be present for the deadman circuit to remain active (and the relay turning on the load, closed). If the driving computer freezes with the output high or the output lo, the circuit opens the relay and shuts down after a few seconds.

The responsiveness could be very slow -- Ideally a pulse of once per second would be sufficient and the response time for relay to open could even be as long as 30 seconds.

Attached is a rough "guess" for what is needed... Anyone have any ideas? Any better way of doing this? Do the values look kind of correct in the attached circuit?

ANY help would be greatly appreciated.

Ted
 

Attachments

  • 2012_10_09_17_28_57.png
    275.5 KB · Views: 76

Way back in 1984, the TV transmitters supplied to my organisation was controlled by micro processor based controller. On site some one thought, what would happen if the controller or the comms link went down, so a "pulse control" module was invented, which powered the transmitters, though the software had to be modified to make the "on" and "off" signals square waves. You could perhaps think of using your controller to keep switching the relay on,the default being off.
Frank
 

Do a search here, and on google, for "missing pulse detector". Very similar to what you have...
 
The application is to turn on and off the heating element in a tank.
A number of temperature sensors read out the temperature every second or so the the computer.
In a few test runs, every so often the computer would freeze, once with the io pin high (keeping the heating element on) which luckily someone was there to catch it -- otherwise there would have been a steam buildup

Turning on and off a pin to make a square wave from the software is easy -- I'm confused on how to make the circuit that makes sure the square wave is still there and not just ON or OFF... otherwise, turns off the output relay.

Any ideas on how the to make the circuit that has to get an on - off - on -off pulse in order to close a relay??

OR do you know of anything commercially available like this?

Thanks!

 

Thanks to all! "inverted missing pulse detector" is exactly what I needed.

Ted
 

Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…