What will happen if we switch an electro mechanical relay more than the specified frequency? Eg if we switch a relay from On and then back to OFF before it’s operate time and vice versa , how relay would behave .

  • \$\begingroup\$ That would be a buzzer. You could adjust the amount of travel to change the volume. Many old style intercom sets use a buzzer like this. \$\endgroup\$
    – John Canon
    Apr 2, 2022 at 22:39

2 Answers 2


more than the specified frequency

That's by the very definition of "specified" ... not specified. It might switch, it might not switch. It might arc too much, it might break earlier, it might not.

  • 1
    \$\begingroup\$ Yeah, it might not have enough time to completely make or break the connection as it takes some time for the switch to physically move to the open or close position. If you really need a faster switch, use something electronic instead like a MOSFET \$\endgroup\$
    – Miron
    Apr 2, 2022 at 19:03
  • 1
    \$\begingroup\$ What type of load are you switching, it will have a bearing on the life of the relay. \$\endgroup\$
    – Gil
    Apr 2, 2022 at 21:20
  • \$\begingroup\$ can it possibly get contact weld permanently ? load is smps. on pressing switch, smps turns on and off. \$\endgroup\$
    – Design4ec
    Apr 3, 2022 at 9:01
  • \$\begingroup\$ @Design4ec that's the thing: nobody can ever tell you. You're using it outside specifications. \$\endgroup\$ Apr 3, 2022 at 10:34

When talking about “switching”, one must distinguish between toggling electrical input vs. toggling the contact.

You can wiggle input voltage to a relay as fast as you want, but that doesn’t mean that the contact will actually close or even move much.

For the contact to close quicker, you need to increase the drive voltage to counteract the coil inductance and to increase the magnetic field strength to overcome the inertia forces.

In most relays, you can thus close the NO contacts almost arbitrarily quickly by raising the initial coil voltage – until the wire insulation in the coil dielectrically breaks down, and/or until a ham radio operator living 2 miles away calls the FCC on you :)

I’ve driven some 12V rated reed relay coils from short 300-500VDC pulses, bringing the contact closure times into the 10-30us ballpark. This has limited uses – mostly just a silly exercise to demonstrate ultra-low-latency arithmetic on relays. You can make the relays do it about as fast as early microprocessors, but only once every few seconds :)

The downside really is that you can’t do this to the poor relays very often. The contact opening times cannot be generally improved without constructing the relay differently – so the “recycle” frequency is still rather low. Close fast, open slow.

Relays have mechanical self-resonance frequencies at which their contacts close with more force than normally. This usually improves contact life vs same non-reactive load switched at a lower frequency.

The resonant frequency is higher than the maximum operate frequency specified in the datasheet. It’s not very practical though, as due to drift/aging, the relay requires closed-loop control to stay on the resonance.

It also forces you to essentially have the relay act as a fixed duty cycle, fixed frequency PWM driver. Most relays are used in circuits where this mode of operation just doesn’t arise and is of no use.

The effective slew rate of current at contact closure and contact opening has big effect on contact life – the faster the rise/fall time, the less wear on the contact. The faster and more forcefully the contact closes, the better. Thus, generally, it makes sense to close the relay as fast as possible without over stressing the armature – this is applicable to small armature relays and reed relays. Larger armature relays and contractors typically fail due to fatigue fractures of the armature if you try that.

Many relays have electrical time constants not much better than the mechanical ones, so faster switching requires higher coil voltages to improve contact life. This trades off thermal fatigue on the coil for contact life. Whether such trade off is warranted depends on the application.

And, finally, what any particular relay does in “off-datasheet” application is up to you to figure out. You have to develop a relay test setup not unlike the ones used by the relay manufacturer R&D and production test departments. There’s lots of good experimental science here that integrates mechanical, electrical, electromagnetic and statistical knowledge. If you want to build a product around such off-label use, you have lots of work ahead to make sure you understand the relay behavior almost better than the manufacturer does.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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