Does constantly energizing a relay shorten its life?

It makes the coil run hotter, but as long as you're well below the maximum it should not appreciably affect life. AC relays run a lot hotter than DC ones of the same type.

What is that, a 150mW coil or something? It's not worth any effort at all unless you're operating very near the high temperature extreme and/or are going to be putting overvoltage on the coil.

If you're switching significant current compared to the ratings, note that the NC contacts are generally not as good as the NO for electrical life.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

pass

normally

Thanks for the response. That's a load off my mind (and a cost savings for my company).

Best regards in your direction, Ray

If you're interested in further detail on the topic, there was a recent thread on sci.electronics.design on exactly this question.

But, Spehro's answer summarizes it.

pass

I was diligent in that I googled "constantly" "energized" "relay" in groups and got no decent hits. I would be interested in more detail if you could provide a link or appropriate google key words.

Regards, Ray

It should be okay unless the ambient is very hot or you run it at more than 10% overvoltage. Vibration can be more of an issue, I've seen the wire break at the solder joint.

- YD.

--
Remove HAT if replying by mail.

It looks like the first message was dec 27, 2003. The subject was: "Do relay coils wear out if used continuously?"

The original poster was "Michael," with email " snipped-for-privacy@nowhere.com."

Here is the full text of his post:

You should be able to find the thread now. ;-)

Mac

Most relay data sheets quote mechanical life and electrical life. Both are in terms of switching cycles. Mechanical life is generally many times electrical life. A typical figure might be 5 million. Electrical life is quoted at certain load conditions e.g. 1A, 30V DC resistive load. Typical figures range from 1 million down to as a low as 20K. Some manufacturers give curves of electrical life versus load. These curves usually don't include the load conditions you are actually using so you have to guess.

There is also a specification for coil voltage range. The maximum voltage is limited by permissible temperature rise and coil temperature. In many relays it is actually only 10% over nominal. The minimum voltage is "pull in voltage", i.e. the smallest voltage at which the relay is guaranteed to pull in. This is typically 80% or 70% of nominal. Taking those limits into account means it is usually impossible to drive a relay off an unregulated DC power supply and guaratee you are within allowable limits.

You may have to guess. We *test* at the desired current, even if the manufacturer supplies curves. A million or five million operations doesn't take long, though you have to make sure not to switch so fast that contact heating prejudices the results. There is a WIDE variation in the improvement one can expect from derating resistive load current.

Nonsense! MOST DC relays operate from unregulated supplies! Billions (quite literally) of them!

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

It keeps the spring components under constant tension.

I work with relays that have been in service since the 1940's. Most of them are the originals. When we design a relay circuit, most cases we design for the relay to be normaly de-energized. This is mainly to save on power, the circuits are fed from batteries that are constantly charged. I can think of relays that are normarly energized, and have been for most of their lifetime (1940's)! Of course, these relays are industrial and hundreds of dollars.

message

news:j4dQb.10835$ snipped-for-privacy@newssvr23.news.prodigy.com...

info on the subject or pass

G5V-2-H-24VDC relay in a normally

worth the effort.

since the 1940's. Most of them

circuit, most cases we design for

mainly to save on power, the

constantly charged. I can think of

been for most of their lifetime

and hundreds of dollars.

Consider using latching relays.... No power except on a change of state!

groups

could

Thanks for the help! Ray

It won't do in some cases. An emergency halt relay for a chemical plant or something similar MUST be powered for the plant to operate and it MUST stop the plant if something happened. That's why all those big red buttons are ALWAYS break the circuit when pushed - they are in series with that relay's coil...

--
******************************************************************
*  KSI@home    KOI8 Net  < >  The impossible we do immediately.  *
*  Las Vegas   NV, USA   < >  Miracles require 24-hour notice.   *
******************************************************************

For long life, you can drop the supply voltage considerably after the relay pulls in. I would think the only life factor would be the operating temperature.

--
Many thanks,

Don Lancaster
Synergetics   3860 West First Street  Box 809  Thatcher, AZ 85552
voice: (928)428-4073 email: don@tinaja.com fax 847-574-1462

Please visit my GURU's LAIR web site at http://www.tinaja.com

Relays are known to last well over 50 years in intermittent operation, and well over 20 years in continuous operation - as long as the coil and the contacts are operated within design specs. For higher safety reliability, consider doing *all* of the following:

1) Put a (thin - 3 mils will do fine) piece of paper, mylar, or kapton between the pole and the armature to prevent the armature from touching the pole piece. Reason: the iron can get magentised and may prevent drop-out. 2) Mount the relay so that gravity can pull the armature away from the pole piece. Reason: if the spring breaks or otherwise becomes disabled, gravity will do the work; but only if there is that gap previously mentioned. 3) *TEST* each and every relay for 500mSec (or better if you are fussy) dropout *without* a spring; and then make *damn* sure you put the spring back in; someone else should double check that the springs are properly in place! Be advised the typical dropout time of relays is 100mSec. Reason: This verifies that the back-up gravity powered opening of the relays is functional and reliable.

There are two main constructions of power relays (which does not include shorting-bar type contactors).

1) A hinge plus spring to return the armature, flexible wires conduct the current. 2) A current-carrying flexure acts as the spring

In addition, European-spec relays often indirectly operate the armature with an insulating pusher to get sufficient creepage (typically several kV coil-to-contact). Some relays (eg. of the P&B T-70 construction) have awful breakdown ratings of something like

500V.

THe first construction is prone to rare failures from hinge sticking. I have seen an entire batch from a major manufacturer many years ago that had a 5% or so failure rate due to sticking- in the field. Very, very costly, and very subtle to find. The second is prone to possible failure from over-current, which destroys the spring characteristics of the flexure by annealing. The European type is probably subject to other problems from friction or contamination in the extra moving parts, but I've not seen them (yet).

I could go on (and on) but there are a lot of things that affect relay reliability and life (venting vs. sealing for example, a trade-off), but the point remains that continously energizing a relay has essentially no effect on life under normal conditions.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com