I have a mission-critical device, which involves a micro and mosfet controlling a pump. I want to implement a failsafe, so that if basically anything happens to the micro, or the mosfet blows, the system will engage and turn on a backup pump. The easiest way to do this, it would seem to me, would be to have an always-engaged (NC) relay that is disengaged by a failure, energizing the backup pump. So, my question is, is it a good idea to have a relay on continuously for what may amount to years? The extra 20ma for the coil is negligible in this application, but are there any other consequences to doing this?
-
\$\begingroup\$ Been a while since I looked over the data sheet for any relay, but don't they usually publish a "duty cycle" rating? \$\endgroup\$– Solomon SlowSep 4, 2015 at 17:45
-
\$\begingroup\$ Look for Continuous Duty Relays, should be fine up to the expected lifetime hours of the device \$\endgroup\$– crasicSep 4, 2015 at 17:47
-
3\$\begingroup\$ Your description is slightly confused : a NC relay will be switched ON (Closed) with no power applied, and OFF (open) when you apply power, i.e. disengaged UNTIL a failure. Thus when power fails, the switch will close. This is good news, because relays can fail to open (contacts weld shut), but not usually fail to close (unless the coil has burnt out, which is easily detectable. If the coil only takes 20mA, chances are this is a continuous rating anyway \$\endgroup\$– user16324Sep 4, 2015 at 17:54
-
\$\begingroup\$ @Brian Sorry for the confusing wording. By "engaged" I meant that the coil would be on, but the relay would be open. \$\endgroup\$– 0xDBFB7Sep 4, 2015 at 18:45
-
\$\begingroup\$ OK. Your main system should be able to test the relay and detect it failing to open (i.e. backup pump running when it shouldn't) and warn or behave appropriately. \$\endgroup\$– user16324Sep 4, 2015 at 18:54
1 Answer
Look for a high quality relay that has a rated maximum continuous coil voltage. As long as the voltage driving the coil stays below this value and the environment (temp, humidity, etc) stays within the rated ranges then you should expect to be able to drive the coil continuously for its rated lifetime.
One failure mode to consider is that relay contacts can become "stuck" shut. In this case, the relay does not open even though the coil is deenergized. This happens usually because of sparking across the contacts when they are connected to an inductive or high voltage load when they open. It can also happen if you over heat the contacts by driving too much current though them and getting them hot.
It might be a good idea to periodically deenergize the relay both to exercise and to test it. Ideally, you'd do with without a connected load to minimize wear the contacts.