Relay Keeps Sticking Randomly

[Jaderman]

Hey all,
I've been working on a RFID project that turns on my lights when I swipe a card in front of the reader. But the problem I am having is that the two relays that are hooked up to my light switches (I have two lightswitches) keep sticking either on or off randomly. Sometimes they turn off or on together or at different times. Possible problems are: 1) the relays that I am using are not made for continuous switching (I'm not sure if there are such relays that can't do that but its a possibility), 2) the power source which is a 9v battery is about 5 feet away and I read somewhere that if the power source is too far away that it messes up with the something something don't remember haha, or 3) the relays are bad. Sooooo, any ideas or suggestions what could be wrong are greatly appreciated! Thanks in advance!

 

[Harald Kapp]

Can you give more details?
What do you mean by "they keep sticking"?
Do the contacts stick even if the current through the coil is turned on or off?
Do they stick in the open or close position?
Have you measured the current through the coil to ensure that the control signal is really turned on and off?
Are the relays rated for the load they are attached to?
What are the coil data of the relay? do the data match the battery you're using? Have you measured the voltage that is available at the coil pins?

Harald

 

[Jaderman]

Can you give more details?
What do you mean by "they keep sticking"? - Like the contacts inside will 'weld' together for awhile and I can reset it by touching another 9v battery temporarily on the coils and it will click back to normal.
Do the contacts stick even if the current through the coil is turned on or off? - Yes. I can unplug everything and turn off everything, even the power to my room and they will stay sticking for awhile. Then eventually they will come unstuck.
Do they stick in the open or close position? - Both. Completely random.
Have you measured the current through the coil to ensure that the control signal is really turned on and off? - No, I haven't.
Are the relays rated for the load they are attached to? - Yes. At 120VAC, they can handle 7A. The max one relay has to switch is .33A (three 13W high efficiency bulbs) and the other has to switch .54A (one 65W bulb, the light fixture is made for three bulbs but one is blown and the other socket is broken)
What are the coil data of the relay? do the data match the battery you're using? Have you measured the voltage that is available at the coil pins? - I don't know the rating of the coil. I just know that 5V does not make it switch. And no, I haven't measured the voltage.

 

[(*steve*)]

The normal "fix" for a sticking relay is to hit it with something. (only partially kidding).

Does the relay get stuck in the ON position or the OFF position? (OK, you said both)

I assume it's the off position, because if it was stuck in the on position it seems unlikely that applying voltage to the coil would help. (OK, I assume the battery trick only works if its stuck in the off position)

I think it's possible that the relay is damaged.

If you don't know the voltage rating for the relay and are just applying "enough" voltage for it to pull in, then you may have some arcing occurring because the contacts are not tightly closed.

Bulbs can have a high switch on current. If the relay contacts are not closed properly, this could easily damage them even though the current once the bulb is on is far lower than their rated current.

 

[Harald Kapp]

If the contacts stick, then it is not a problem of the battery or the coil but of a welding of the metal. Yes, this can happen.
According to the data the contact rating should be sufficient. But obiously something isn't. One possible cause:
If the driving voltage /current through the coil is not strong enough to quickly turn on the relay, the armature may move rather slowly and thus a small electrical arc could develop. The arc in turn melts the surface of the contacts and they may weld.

Two ideas to remedy this:
1) use a higher voltage to drive the coil (or a relay which is rated for a lower coil voltage)
2) put a ceramic capacitor (e.g. 2.2nF disc capacitor) suitable for use with 120 V (AC) across the contact side of the relay. This capacitor will, however, leak a small current even in the off-state of the relay. If you can't accept that, only 1) is suitable.

 

[wingnut]

Surely testing a relay is fairly straight-forward if one can remove the relay from the circuit or buy an equivalent one? One just applies the 9V across the input and listen for the click as the relay closes.

Are you sure the relay is rated for the voltage of your lights? If the voltage is too high then the relay may "weld". But then it would be unlikely to ever switch again.

I think the problem is more with the electronics driving your relay, especially if it turn on and off randomly. Or the circuit needs exorcising.

 

[GreenGiant]

The fact that they stick both ways, and the fix in both cases is hitting the coil with the 9V battery again indicates to me that they may be latching relays?

Just a thought

 

[donkey]

green giant I thought the same until I saw

Do the contacts stick even if the current through the coil is turned on or off? - Yes. I can unplug everything and turn off everything, even the power to my room and they will stay sticking for awhile. Then eventually they will come unstuck.

 

[Harald Kapp]

One thing is curious: they stick in the de-energized position, too, do they?
I can understand a welded conact in the energized position.
I can understand a relay armature not moving when energized.
But both simultaneously?

Can you give us the type of relay? Usually it is inscribed on top, rarely on the side of the relay. Maybe we can find out more this way.

 

[gorgon]

The only relay I could think of that could fit into this description, at lest partially is reed relays. But they are not made for power applications, so the point is moot.

TOK

 

[Jaderman]

No, they aren't latching relays.

I tried to switch the relays with 12V but my transistors couldn't handle that much so I just left it.

So, I ordered two new relays from:
http://www.jameco.com/webapp/wcs/st...Id=10001&ddkey=http:StoreCatalogDrillDownView
Sorry about the length. D= @Harald, These are the exact same relays i had installed.
BUT, when I installed them, it didn't switch. Each relay's coils voltage is rated at 9V EACH, combined its 18V. Arrrggghhhh D: So, I won't be able to tell what the problem was with the other relays now until i replace the transistors on my circuit cause, like i said, they won't be able to handle that much...

 

[Jaderman]

Am I right in assuming the combined relays do need 18V?

 

[MrEE]

A schematic here would be helpful. What kind of transistor did you use to switch the relays?

 

[Harald Kapp]

If I understand correctly the relays were in series, needing 18 V? In that case it is no wonder they sometimes stickj in the rest position. 9V can suffice to activate the relays, but not for sure. You should wire the relays in parallel.

Still, this doesn't explain the sticking of the relays in the ON position.

 

[gorgon]

We are then down to what do you switch with these relays, and how do you get 18V from a 9V battery. A schematic would clear some fog, I hope.

TOK

 

[Jaderman]

Here is the schematic:

The 'SIGNAL TO ARDUINO' pins are both hooked up together and plugged into one pin on my Arduino. The reason that I didn't hook both lights into one relay was because, later on, I was planning on making the light switch independently (like in the movies in a big warehouse where all the lights flip on one by one from where you're standing to the very back. =D) So, to turn on the lights, the Arduino pin turns to high to switch the relays. My brain is kinda cramped right now (I just made that schematic early in the morning and took me way too long then expected. haha) but, I think the relays are hooked up in parallel. Right?
The 'TO LIGHT 1' and 'TO LIGHT 2' pins go to either side of the actual light switch poles. A light switch is wired up with power and common, flip the switch, and it completes the circuit and the lights get power. I have hooked 'TO LIGHT 1' pins in parallel with one light switch (I have two which I already said earlier) and 'TO LIGHT 2" pins in parallel with the other light switch.
The 9V is being shared with the Arduino and the relays.

 

[Jaderman]

I don't think the pic worked. Here is the link:
http://www.flickr.com/photos/jadermaner/7787994752/in/set-72157627727738602

 

[MrEE]

It looks like you have the emitter and collector swapped. also you need a flyback diode accross the coil. (Cathode side to Vcc or (+)). Also you are missing a base resistor.

 

[Jaderman]

I see that now. But why did it work? I was trying other transistors that I had to see if they would work and I grabbed the same transistor that I was using in the schematic (the 2222N) and hooked it up the same way. When I touched the base pin with positive voltage, I could hardly hear the relay click and when I took the positive voltage off it didn't click again, not until I unplugged everything else did it click to off position. That's exactly what has been happening in my circuit. But, the odd thing is, when I tried another transistor, it didn't work and then when I tried the other transistor again, it didn't work. 0_o I have enough experience with electronics to know if I was just doing something noobish. haha What in the world is going on?

 

[MrEE]

you're not supposed to touch the base of a transistor with a positive voltage. You are likely to destroy it (that is if either the emitter or collector is tied to gnd and transistor is a NPN) . Check this link http://www.kpsec.freeuk.com/trancirc.htm. there is a section on driving relays as you scroll down a bit. Try that circuit and let us know how you are doing.