Why is relay latching? :(

[gorgon]

You should add a diode on each relay coil to reduce the reverse peak voltage that may ruin the switches on the first relay, and make them stick.
It may also be a good idea to add some sort of snubber circuit over the bell switch(es) if they are coil based. 47 ohms in series with 100nF should make a start value.

Regarding the problem with the LEDs in the switches, it is obvious that the current through the LEDs add up to the hold current for the relay, and makes it stick. To fix that you need to reduce the idle current through the relay. if the relay worked with one LED, you need to bleed the extra LEDs current away from the relay coil. A resistor of 100 ish ohm may be low enough to do this, but remember that this resistor will have 12V over it as long as you push the button and burn some 1.5 W. If you use 4 x 27ohm, 0.5W resistors in series you should at least have a safety margin on the wattage, even if they get hot.

 

[Kookie]

Hey THANKS guys! 100ohm 2W resistor got things to work. I very much appreciate your assistance!

I yanked out that bell that stopped working and threw in a buzzer. I picked it up at Grainger earlier this week when I happen to be driving by. It was only 10 bucks and works at 3-6VDC. That was easier than trying to get that bell to work when I've been pressed for time.

I was going to throw in that snubber circuit, yes they are coil based, but I had less than an hour to take apart the box, solder in the resistor, and throw up the buzzer. I think I will sit down this weekend some time and remake this circuit. I have enough of all these parts I used to make another.

BTW, measured the current on the LED SW's. Only 9mA each. Wouldn't have thought that was enough to hang up the relay.

 

[Kookie]

Ok, kind of new problem. Although I did mention it for the old bell. The new buzzer I put in on that 5V side now stopped working as well. I don't understand why these units keep dieing. I know there is 5v there. What would keep killing them? Looked over the old bell, looks fine.

 

[hevans1944]

Try removing the buzzer and connecting it directly to your five volt supply. Do the same thing with the bell. If both the buzzer and the bell work fine when directly connected to the power supply (no relay contacts, no fifty feet of wire in between) then there may be something wrong with either the relay or the connecting wire. Perhaps the relay contacts have arced and no longer provide a reliable low-resistance contact closure?

Most buzzers and bells work fine on low-voltage AC, which for a given RMS voltage provides more than twice the peak-to-peak voltage as DC voltage that is equivalent (in heating power) to the RMS AC voltage. For example, a 5 V AC transformer provides the same heating effect as a 5 V DC supply, but the peak-to-peak AC voltage is 2 x √2 x 5 = 14.1 V pk-pk. The higher peak-to-peak voltage will supply more magnetizing current at the peaks than the steady DC current, which may make the bell or buzzer work better on AC than the RMS "equivalent" DC.

Since you are using a relay to operate the "5 V" circuit, you could just purchase a small door-bell transformer and use it to power the bell or buzzer through the relay contacts in lieu of the 5 V DC supply. Or use a higher voltage bell or buzzer and power it from your 12 V DC supply.

 

[Kookie]

Yes. I understand. What I want to understand though is why these units will work in service for about 1 week and then began to fail. That just puzzles me. Regardless of of it being DC or AC. They are extremely simple units. Here is a picture of the bell unit:

Its ratings are 3-8vDC or AC

The buzzer is basically the same thing but with out the bell. That one was rated 3-6vDC

The contacts are clean.

Its almost as if the coil loses its ability the pull up the clanger. Does it somehow become weaker magnetically? That doesn't make any sense to me.

 

[Kookie]

P.S. I will measure the current it pulls. I don't remember off hand its rating. I know for a fact though I'm getting that 5v at the installation point. I will say the 12v bell works like a charm. It should, it was a $50 industrial bell. It needed to be a loud bell because its servicing a very large room. This bell I chose because its in the office and I wanted it to be relatively quite. Not so disturbing to office goings on.

 

[hevans1944]

... Its almost as if the coil loses its ability the pull up the clanger. Does it somehow become weaker magnetically? That doesn't make any sense to me.

It doesn't make any sense to me either.

Here is another suggestion: replace the bell and buzzer with an inexpensive 76 dB piezoelectric buzzer, like this one from Radio Shack, rated to operate from 6 V to 18 V DC. Or this 90 dB piezo pulse, rated to operate from 3 V to 28 V DC. There are others available. Maybe you could take a 6 V lantern battery, with two wires and alligator clips attached, to Radio Shack and convince the sales person to let you "try out" their stock before you buy. Tell them you want to make sure it will be loud enough for your purpose.

I have a 115 V AC version of one of these things attached to a charged-particle accelerator to let me know when the ion beam is on. It is so loud I had to encase it in a plastic tube and cap the "noisy end" most of the time . You can put a piece of tape over the "noisy end" of yours to muffle the sound.

 

[gorgon]

nWhat exactly happens when the bell/buzzer stops working?
Does the magnet pull the banger and holds it in, or does it not pull it enough to break the interrupter contact?

Have you measured the 5V when it is working/ stops working?
Is it still 5V or has it collapsed to a lower voltage?

What is th rated max current of your 5V power?

What is the DC resistance in the buzzer coil?

You could try to add an electrolytic capacitor over the output from your 5V power, before the relay, to reduce the impedance of the power supply. 470 or 1000uF should be ok, but the bigger the better, at least up to a point.

 

[Kookie]

Ok, heres the scoop so far:

That bell, pictured, stopped working after a week because it wouldn't pull in the clanger unless I nudged it. So, I decided to throw in that buzzer instead as a quick fix. (I have still yet to test out that bell to see why its malfunctioning.)

The buzzer stopped working after a week of service in an opposite way. It was pulling in the bar, but there wasn't enough clearance for those contacts to make and break (the buzzer is setup just like the bell). So, I bent down the metal tab with that contact to give it some room. Its working now again. We will see for how long. Not sure why after a week they seem to warp to not function. It seems my problems are all mechanical in nature not the circuit.

Measured the voltage with it pulled in at 4.8V 600mA. Well within the specs of the PS being output: DC 12V/5A, 5V/3A. Shoot, forgot to measure the resistance of the coil.

Here is the PS on ebay. This was my last one. I picked that up with some other stuff quite a while back all for super cheap. To me is was a quick, easy solution at the time. Throw it in a box with my relay and some wire terminals for in outs. Easy AC interface. (Would have done it differently by now, lol.)

I did apply some of hevans1944 advice and picked up a piezo on the way there. I put that in parallel with the buzzer in case it fails again. It will make some sound. Will see how long that cheapy buzzer will last.

 

[hevans1944]

The piezo may not like the competition of working in parallel with the buzzer. I've never tried that. But those I have tried will make a LOT of sound when continuously powered up. Oh, well, mission accomplished: you can now hear if someone is at the door pressing the button!

I have a front entrance doorbell button, and at one time I think there was a set of chimes in the hall. And maybe a rear doorbell button, too. I found the wiring in the basement after I bought the house but never bothered to replace the hall chimes or re-connect the wiring. I think the doorbell shared the same low-voltage transformer with the gas solenoid-valve and upstairs thermostat of my gravity furnace. So, for twenty-something years we always depended on a dog (there have been several, some louder than others) to announce visitors. The current dog goes bananas when a stranger approaches the door, but is especially excited by mail carriers.

A few years ago I decided it would be nice to provide a real doorbell again. So, on a trip to the adult toy store (Lowe's) I picked up a wireless doorbell set. The button outside has a battery and an RF transmitter. The noise-maker inside has a receiver and plugs into an outlet in the kitchen. This works okay except for two things: (1) visitors at the door seldom use the push button because the dog is barking and (2) whenever my wife uses the Vitamix to whip up a batch of vegetable smoothies the "doorbell" goes off. Not that you can hear it over the noise the Vitamix makes, unless you are standing right next to it.

 

[gorgon]

I think I would have tried the big capacitor on the 5V supply, to make the supply stiffer. Both the bell and the buzzer draw a lot of current when starting, due to the magnetic build up. You don't say how often the bells are used over a week, but I suspect you get some sort of magnetic hysteresis problem that weaken the pull. As far as I can tell, both the bell and the buzzer problem are the same. A low pull, or a too low speed in the pull, to disengage the interrupter contact.

 

[Kookie]

Ok. Cap makes sense. Will do on next visit. BUT... Mystery somewhat solved. (Kind of. - BTW, this is at my somewhat new job at a bakery. They discussed they wanted a doorbell for visitors and deliveries at the bakehouse, I said, "I'll do that for ya'!" As a side job under my DBA.) Went in early this morning and the floor was soaking wet in the office. Its been drizzling for a day now. I look up at the buzzer in the office and its directly getting soaked. Drip, drip, right on it! There ya' go. Problem kind of solved. I have to move it. It would be my luck to install it right where they happened to get a leak from the roof. It was the logical place though. Center of the whole wall / office. They failed to tell me of the leak that had developed. Wow, I wish I knew and now I can see why what ever I put there would fail over time. Go figure. Geez!

Oh. BTW hevans, didn't want to deal with batteries. They would be stolen so quick. Actually waiting for that to be the next problem. Switches gone, lol. This is in Detroit and not the best neighborhood.

 

[hevans1944]

... Oh. BTW hevans, didn't want to deal with batteries. They would be stolen so quick. Actually waiting for that to be the next problem. Switches gone, lol. This is in Detroit and not the best neighborhood.

I only suggested taking a 6V lantern battery to Radio Shack so you could use it to test their stock of piezo noise makers before buying one. The bozos there probably wouldn't let you do that... "pay before you play" policy at most retail stores, plus the stock is probably secured inside a blister pack or shrink wrap the sales personnel won't let you open without purchase. Doesn't hurt to ask though.