1
\$\begingroup\$

I'm confused weather I should go for a standard electromechanical relay (EMR) or a sold state relay (SSR) to switch CFLs @ 220V with 100W max load. This is for a commercial product with the following objectives:

  1. Have at least 7 years of life (of continuous use)
  2. Resilient to short-circuits at the outputs of it's switching device (SSR, EMR or TRIAC).

The main concern here is the failure of the device: SSR will provide an extremely long life (if other components are up to it) whereas a EMR will have a relatively shorter life-span. However, it seems EMRs are more rugged to a short-circuit (till the circuit breaker or fuse pops, at least). A failure mode of CFLs I've found is that they sometimes short-circuit - either the internal wiring insulation fails or some other type of fault due to sub-standard components).

I suppose the question becomes: how long can I expect an EMR with a resistive electrical contact life of 100,000 cycles to work if the relay is switched 10 times a day?

Alternate approach that I had in mind: use a large TRIAC (rated for 30+ amps continuous) and a resettable fuse, rated for 500 mA, in conjunction with it. The fuse interrupts the current flow thereby saving the TRIAC from damage. Since the TRIAC is a heavy-duty one it has a higher likelihood of surviving such a failure.

Will the TRIAC allow me to have my cake and eat it too?

\$\endgroup\$
  • \$\begingroup\$ Common SSR is just a TRIAC with optocoupler, so anything you said about TRIAC true for SSR too. (But SRR is more expensive than TRIAC with comparable preformance). \$\endgroup\$ – Vovanium Aug 5 '14 at 12:13
1
\$\begingroup\$

It might be on the optimistic side to expect the resistive life to apply to a CCFL load, however an overrated (eg. 10A or 16A) mechanical relay with any kind of a rating for a tungsten load will probably last just fine with a 0.5A CCFL load.

If it was purely a resistive load you'd probably get in excess of 10^6 operations. Check the life curves in the relay datasheets.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Is the main problem the inrush current? One could limit that with an NTC? Also, most 10A relays seem to need a 100mA min @ 5VDC. load or more. It would work fine with a 100W CFL but wouldn't the life reduce if the load is less, say, 10W? \$\endgroup\$ – Saad Aug 5 '14 at 17:17
  • 1
    \$\begingroup\$ Yes, but it may not actually be a problem. Inductive loads are a lot nastier. You could easily perform tests. At one operation per 2 seconds you can get 10^5 operations in a couple of days, and I suggest you do this before thinking about NTCs. \$\endgroup\$ – Spehro Pefhany Aug 5 '14 at 17:19
1
\$\begingroup\$

10 cycles per day isn't much. Simple math says 100000 / 10 / 365 gives a life of 27 years for your EMR. Even if we take the 100K as a mean and assume failure can occur at 50%, you're still over your 10 years.

If it were me, I'd just go with the mechanical relay.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ In terms of mechanical operations you are right. EMRs will last a life time if they are not cycling load. But if you are breaking load over and over the contacts will wear out much faster than the headling 'max # of operations'. It may still be acceptable, but it might be an order of magnitude fewer operations. \$\endgroup\$ – Nick Aug 5 '14 at 15:46
  • 1
    \$\begingroup\$ As a reference, see the DC switching cycles plot on page 2 of this data sheet: gigavac.com/pdf/ds/pp/gx11.pdf#page=2 \$\endgroup\$ – Nick Aug 5 '14 at 15:48
  • \$\begingroup\$ But I agree with you - I am kind of old school, I like EMRs when space isn't a concern. The 'click' is satisfying feedback to know things are working. \$\endgroup\$ – Nick Aug 5 '14 at 15:54
  • \$\begingroup\$ @Nick Nice very specialized contactor. Good to know about. \$\endgroup\$ – Spehro Pefhany Aug 5 '14 at 18:44
1
\$\begingroup\$

Since the question has no mentioning on space/weight restriction, how about using a circuit breaker with a suitably over rated relay (says, a few A is reasonably cost, weight and size)?

In case of external short, circuit breaker is faster than fuse so the short circuit overload energy is lower and stay within limit of a 'suitably over rated' relay.

As illustration (not that this particular one suits you), this randomly chosen relay has formal spec. on pulsed rating. "steady State Current Capability = 3.31 Amps. For pulse durations less than 0.5 s, it can carry up to 16.55A, factor of 5. For 50 ms, it can carry up to 33.1A, factor of 10."

For 0.5 second pulse, current rating 5 times higher than continuous

| improve this answer | |
\$\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.