3
\$\begingroup\$

While disconnecting relay contact, it is recommended that the action should be done at or near current zero. As the contact area decreases rapidly, a spark is generated due to high current density. But what about the time while connecting the relay? The current will be zero before connecting the contacts. Should I connect the contacts at voltage zero so that the current can't increase much? Or It's unnecessary?

\$\endgroup\$
6
  • 3
    \$\begingroup\$ 50-60Hz is too fast and an electromechanical relay doesn't open fast enough to catch that. Solid state relay yes, electromechanical, no. \$\endgroup\$ – DKNguyen 2 days ago
  • 1
    \$\begingroup\$ So making a current zero detector for driving relay (electromechanical) is useless? \$\endgroup\$ – Sadat Rafi 2 days ago
  • 1
    \$\begingroup\$ If it is electromechanical, yes. They bounce for ms, sometimes 10s of ms when closing and take 10s ms to open up. Whereas a 60Hz cycle is 16.6ms and the zero cross is much much shorter. \$\endgroup\$ – DKNguyen 2 days ago
  • 2
    \$\begingroup\$ It depends on the load, if the zero-cross is the best moment. If the load is a transformer, then it would be the worst case, but for resistive load is the best. \$\endgroup\$ – Marko Buršič 2 days ago
  • 2
    \$\begingroup\$ Oh, I didn't notice. You want zero CURRENT, not voltage. \$\endgroup\$ – DKNguyen 2 days ago
4
\$\begingroup\$

Relays are electro-mechanical and dropout time will be in the order of tens of milliseconds or more. If a snubber diode or resistor is use the drop-out time will be longer. In addition, mechanical differences between the contacts of several poles of the relay will mean that the contacts open at different times.

It will not be possible to control the relay to switch off at a specific point in the mains cycle.

\$\endgroup\$
4
\$\begingroup\$

Theoretically, it would be ideal, but you want to do it for zero CURRENT, not zero voltage. But the zero-crossing on 50-60Hz AC is far too fast to worry about that on an electromechanical relay which has inconsistent opening/closing times, takes tens of ms to open and bounces for ms, sometimes tens of ms when closing. It can't be achieved in practice. You can use an series R-C snubber instead parallel with the relay primary contacts however you lose the isolation of an open relay if you do this which matters in some cases.

You can worry about it on a solid-state relay, however.

\$\endgroup\$
2
\$\begingroup\$

If the current stays below the specified limit for the relay at all times, there's no reason to do anything: the relay is supposed to perform its rated number of cycles in these conditions. Making sure you switch at a lower current may increase the possible number of cycles beyond the spec, but you shouldn't count on that anyway.

If you're afraid you might exceed the rated commutated current, you should get a better relay, or limit the current. For instance, if you have big capacitors as a load, you could make a precharge circuit which limits the inrush current.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.