Extending the life of my relay?

[maniac84]

Hi guys...
I have a relay application like the circuit below.

My relay spoils after around 100 times of ON/OFF only.
My question is, is my design create lot's of voltage spike when turning ON/Off the relay? Is it the spike that caused my relay spoils?
I have put the diode to suppress the spike. Does it work?
How do we extend the life of the relay?

 

[thylacine1975]

The diode protects the transistor from the turn-off transient of the relay's solenoid inductance. This transient won't hurt the relay though - the only way the relay could be damaged is via the load it's switching. Are the relay's contacts appropriately rated for the mains load you're applying?

If the load is reactive (i.e. not a nice resistive load like a lamp or a heater), you'll need to increase the relay's switching ratings further - google "relay contact ratings" for a plethora of guidelines. If the load is particularly horrible, you might benefit from an RC "snubber" network across the switched contacts. The snubber ratings, values etc will again be dependent on what you're switching, so go a-googling

 

[maniac84]

The diode protects the transistor from the turn-off transient of the relay's solenoid inductance. This transient won't hurt the relay though - the only way the relay could be damaged is via the load it's switching. Are the relay's contacts appropriately rated for the mains load you're applying?

If the load is reactive (i.e. not a nice resistive load like a lamp or a heater), you'll need to increase the relay's switching ratings further - google "relay contact ratings" for a plethora of guidelines. If the load is particularly horrible, you might benefit from an RC "snubber" network across the switched contacts. The snubber ratings, values etc will again be dependent on what you're switching, so go a-googling

My load is actually a TV or a monitor. Is it a reactive load?
What do you mean by horrible load?
Below is the relay I use. Can it be use to turn on TV or monitor?

 

[crutschow]

Such a load may have a high inrush current which is frying the contacts. Normally it should handle the steady-state load without problem. You could try adding an inrush current limiter of the proper current rating in series with the load.

 

[jiripolivka]

With a load like your TV or monitor, the inrush current may be the problem as stated above. Likewise if you need to switch a motor or another inductive load.
Instead of trying inrush current limiters or snubbers, the easiest way is to add another "high-power" relay, with larger contacts and a wider spacing between them. The first relay shown in the picture will drive the second relay coil only, at e.g. 12 VDC.
There are also semiconductor high-power relays rated for e.g. 20 A at 240V AC, some optically coupled. Such devices are almost indestructible as there are no sparking contacts. Check Jameco or Digi-key.

 

[maniac84]

My relay problem now is after relay on, sometimes it can't be switch off... Seems like the contact had stick together.
Does this caused by inrush current?

 

[FvM]

Does this caused by inrush current?

Sounds like a reasonable guess.

You can use (unfortunately larger) relays respectively contactors with higher current rating, It's common to control loads with inrush current in automation systems without any contact protection. Contactor liefetime won't be mind blowing, but some 10000 or 100000 cycles should be possible. Remember that the manual on/off switch also uses mechanical contacts that don't fail after 100 cycles.

Inrush current limitation would be of course preferred. A RC snubber or VDR to reduce sparking during switch-off won't be bad too.

 

[maniac84]

I think I would prefer connecting a capacitor and a resistor. But how do we determine it's value?

- - - Updated - - -

I have found the formula to calculate the capacitance and the resistance:
https://www.illinoiscapacitor.com/pdf/Papers/spark_suppression.pdf
The formula is not quite clear in the pdf. Is it like below?
C=(I*I)/10
Rc= Vo /[10I(1+(50/Vo))]
The voltage for my load is 240VAC. Then what is my Vo?
How do we determine the I, which is the load current at contact opening? Is it the current rating value for my TV?

 

[crutschow]

I think I would prefer connecting a capacitor and a resistor. But how do we determine it's value?
.................................
The voltage for my load is 240VAC. Then what is my Vo?
How do we determine the I, which is the load current at contact opening? Is it the current rating value for my TV?

A RC circuit won't much affect the inrush current when the contacts close which appears to be causing the contacts to weld.

The load current when the contacts open would be near the TV power rating.

The best solution is likely to use a solid-state relay instead of a mechanical one.

 

[maniac84]

I see. I can't change to SSR cos budget problem.
I hv test a few time... but jz a few times only the relay can't be off ( contact open ). Like 10 times out of 100 times. How could that be?

The normal workable relay wil gv a clear click sound every time on and off right? Mine too. If it is spoilt, u would expect that it won't click when it can't off bcos the contact are weld together. But for my case, u can still hear a very faint click sound. Jz that the relay can't off. How come like that?

 

[crutschow]

The mechanical armature of the relay still likely moves slightly when the coil voltage is applied and removed, giving the slight click sound you hear, even when the contacts are welded. The contacts are on a flexible leaf connected to the armature and that allows the armature to move a small amount.

I believe we've identified your problem and suggested some solutions (inrush current limiter, larger relay, SSR). You just need to pick which one you want to implement. The inrush current limiter is probably the cheapest.

 

[maniac84]

I see.
But how come the relay can't handle big current? I don't understand.
The relay's (like shown in the picture in above thread) maximum current rating is 10A. I don't think my TV have so high current consumption.
If my TV current consumption is just 1A or 2A, does the Inrush Current will reach above 10A?

 

[crutschow]

A TV power supply can have a large input capacitance which causes a momentary large inrush current (many times the nominal current) to charge that capacitance when the power is applied if there's nothing else to limit the current. This can weld the contacts. It's the same principal as how a spot welder operates.

 

[maniac84]

A check on the back of my TV reveal that it use 50W of power only. Then, the TV current should be 50W/240VAC=0.2083A. That's around 200mA of current only. I think the inrush current can't be reaching above 10A or would it?
What is your opinion?

 

[crutschow]

The inrush current really depends upon the input filter capacitance and any input resistance from the input to the capacitance. It's not directly related to how much steady-state power the device uses.

Does the relay stay closed when all power is removed from the driver/coil side? If so that's pretty conclusive evidence that a current surge is welding the contacts, whatever the source of the surge.

I've noticed that you've posted you problem in a couple of other forums. I don't think you'll come up with a better opinion about its cause but good luck with that. ;-)

 

[maniac84]

The inrush current really depends upon the input filter capacitance and any input resistance from the input to the capacitance. It's not directly related to how much steady-state power the device uses.

Does the relay stay closed when all power is removed from the driver/coil side? If so that's pretty conclusive evidence that a current surge is welding the contacts, whatever the source of the surge.

I've noticed that you've posted you problem in a couple of other forums. I don't think you'll come up with a better opinion about its cause but good luck with that. ;-)

I have not switch off the power when the relay can't be off, so I'm not sure whether the contact is still welded. But during the time the relay can't be off, I try switch the relay on (by sending command using serial), I can hear a clear click sound. What does that means? Is it means during the relay can't be off time, the relay is not fully off contact?

 

[D.A.(Tony)Stewart]

It is precisely the inductive lag in current that creates the "wiping" corona arc across the contacts that burns them out in a few hundred or less breaks.

OMRON have the best specs and product anywhere in the world. THeir products are consistently higher rated for VA load switching in the same coil current form factor price etc. and still they fail from inductive arcs across the contact.

A snubber RC and MOV may be required for this type of load. COnsult with OMRON catalog for higher voltage rating relays at Digikey or Mouser or...

https://www.components.omron.com/co...6B03161CD86257538007424DB/$file/G5LE_0813.pdf

 

[maniac84]

It is precisely the inductive lag in current that creates the "wiping" corona arc across the contacts that burns them out in a few hundred or less breaks.

OMRON have the best specs and product anywhere in the world. THeir products are consistently higher rated for VA load switching in the same coil current form factor price etc. and still they fail from inductive arcs across the contact.

A snubber RC and MOV may be required for this type of load. COnsult with OMRON catalog for higher voltage rating relays at Digikey or Mouser or...

https://www.components.omron.com/co...6B03161CD86257538007424DB/$file/G5LE_0813.pdf

Thanks for the suggestion. But I don't think I have the room and budget for changing the relay model.
Now what I can do is just finding out ways to prevent my relay from being spoilt.
I have talk about the RC network in the above thread. Do you think it can suppressed the arc?
How about MOV? How do you determined the MOV's value?

 

[FvM]

A RC snubber usually doesn't help against inrush current related problems. Neither a MOV does.

 

[maniac84]

A RC snubber usually doesn't help against inrush current related problems. Neither a MOV does.

Then what do you suggest?