I was setting up a starter kit example using a relay to switch between red/yellow LEDs. I had the LEDs wired and just wanted to try switching them so I connected +5 to the coil and directly to GND. I hadn't wired in the transistor from the diagram yet. The datasheet said the coil was 330K Ohm, which is the same as the resistor on the LED circuits, so I thought this would be OK.
Bad decision.
The relay immediately began to smell funny and I cut power, but it was too late. Now even with the transistor it will only activate the NC LED, so I assume I fried the coil.
I'm new to the Arduino and haven't found any info on why the transistor is necessary. I attached an analog read to the collector and saw the output switch between 1023 and 840, so my understanding is that at 1023 (5v) there is no drop across the coil so the switch is open. At 840, there is about 1v across the coil and it should activate. But I'm surprised that the full 5v would burn it out. Why not use a relay that could just take HIGH/LOW across the coil to ground?
You make the mistake of thinking we know what starter kit you have, what relay you have and how you connected everything; we don't. Please read: General guidance and How to use this forum
If you connected the relay coil between 5V and ground and it fried then either you didn't connect the coil between 5V and ground or there was something wrong with the coil to start with.
The datasheet said the coil was 330K Ohm
I doubt that very much, more likely 330 Ohms.
I'm new to the Arduino and haven't found any info on why the transistor is necessary.
Because an output from an Arduino cannot supply enough current for most relays.
I attached an analogue read to the collector and saw the output switch between 1023 and 840, so my understanding is that at 1023 (5v) there is no drop across the coil so the switch is open. At 840, there is about 1v across the coil and it should activate.
None of that makes any sense. Please provide a schematic, hand drawn and photographed is fine. How to post an image
[Edit] Probably worth posting photos, including of the relay you burned out, showing the damage.
I've included the wiring page from the Vilros Ultimate Starter Kit. The relay is a JZC-11F.
You are correct that the coil resistance is much less, 320 ohm. I had not included the transistor in the diagram and simply connected b9 to GND. That's when I could smell something, but there is no visible damage to the relay. When I wired the transistor, I could not get the relay to operate, although I did see a change in voltage (measured at d4) from 1023 to 840. That should correspond to a drop from 5v to about 4v, correct? Wouldn't that mean 1v across the coil?
I'm not following what you were actually doing there, but unless you used some power supply seriously other than 5 V, I very much doubt it would have been the relay burning.
I think we need to know what you are using as a power supply and whether you have a (digital) multimeter because if you do not have one, you need to put all this on hold until you do.
Ah! I can guess!
You have connected the diode across the relay coil, the wrong way round!
And not surprisingly as the assembly diagram shows it with the marker bar (cathode) downward! Bad error!
I do not have a DMM, and I am seeing the value. Searching for one online now.
@Paul__B... excellent catch! I reversed the diode and all works as expected. The relay is not burnt after all. I'll get back to the kit supplier, as this is a very bad error, as you say.
I'm not sure what the purpose of the transistor is in this example, since it seems to work fine without it. Perhaps it is just to demonstrate the pattern when larger power supplies are required.
I'm still fuzzy about the theory for flybacks. The diode blocks current normally but when the relay is reopened the voltage spike is allowed to flow back to the 5v. I guess that is not likely to cause damage, whereas a big spike to ground could?
I'm still fuzzy about the theory for flybacks. The diode blocks current normally but when the relay is reopened the voltage spike is allowed to flow back to the 5v. I guess that is not likely to cause damage, whereas a big spike to ground could?
Any coil of wire, and that includes the electromagnet in a relay, forms an inductor. When you pass current through an inductor a magnetic field builds up around it (that's what operates the relay). When you remove the current the magnetic field collapses. When a magnetic field moves through a conductor, which in this case is the coil of wire forming the inductor, it induces current in the conductor, so when you disconnect the current from the relay coil the current doesn't just stop, the collapsing magnetic field tries to keep it going. If the connection to the relay coil is open circuit then the only way to keep the current going is to generate a high voltage in an attempt to maintain the current through the open circuit at the ends of the coil. The voltage will go as high as possible until something limits it, such as external insulation breaking down. That voltage is easily high enough to damage a transistor or other electronic device. The solution is to provide somewhere for the current to go, and a reverse connected diode does that. With a diode there is a clear path for the current from one end of the coil to the other, so the current keeps flowing until the magnetic field has collapsed, flowing harmlessly thorough the diode and without a voltage spike to damage anything.
That should lead you to another question...
BTW, burned out relays and other things with coils of wire smell different to burned out electronic components. Maybe worth remembering which smell is associated with which kind of device.
Most electronic components smell of burned epoxy, not so informative. Testing with multimeter or component-tester is more informative, especially after the event.
PerryBebbington:
BTW, burned out relays and other things with coils of wire smell different to burned out electronic components. Maybe worth remembering which smell is associated with which kind of device.
Not to mention the smell of burnt skin from said testing.
AJLElectronics:
Not to mention the smell of burnt skin from said testing.
Haha! Fortunately in this event there was no puff of smoke like when I forgot to switch the power voltage before plugging my US computer into an Irish grid. That was a more expensive mistake.
Thanks to all who replied, I think I've got this sorted. I don't expect to be dealing with bigger power than that supplied by the Arduino board. Mostly just creating blinkenlights for my grandson.
I've been continuing with my previous post ( Burned out relay - General Electronics - Arduino Forum ) and had it working for a while, then it stopped. The LED on NC works fine, and blinks off when the relay activates, but the other LED does not light. I can hear the relay click. I've switched the LEDs and both LEDs are wired properly, but I'm just not getting power from the NO pin. This is a JZC-11F relay.
I've tried reseating the relay in different rows of the breadboard without any luck. It seems unlikely that the internal contacts are failing but I can't think of anything else. The relay pins are pretty fragile and it doesn't seat very securely but no amount of jiggling seems to help. Anything else I should check?
Which is only half of the model number. What is its voltage rating?
You referred to 320 Ohm in the datasheet. Well, at 5 V, that would only draw 16 mA and I know that the common "Songle" relays in their "low power" version draw no less than 80 mA at 5 V - which is why they require a transistor to drive them. A relay that actually operated at 5 V with a 320 Ohm coil would be able to be connected directly to an Arduino.
My guess is that you have a 12 V relay which will only partially operate at 5 V.
peterkwa:
I thought this was really a different issue and the previous subject was misleading. But I'll move back to that thread.
Then you should have corrected the subject heading in your very first post. Obviously it is exactly a continuation of the original thread problem.
Whoa, big fella! Newbie here! Seriously, thanks for the guidance.
I've attached pictures of the relay and the schematic. The relay says 5v -- is that the missing piece of the part number. I know several companies make this model, but I didn't know how to interpret the logo in upper left.
The schematic is correct in it's diode connection, I think, but the wiring diagram incorrectly reverses it.
Right now, this wiring lights the NC LED but not the NO LED, although I can heard relay click. If I cut out the transistor and connect the coil to pin 2 (signal) and GND, then both LEDs work. It worked for a bit with the transistor but stopped.
My guess is that the transistor is decreasing the voltage across the coil rather than increasing it, and it's not sufficient to fully activate the switch. I have a DMM on order and will test the actual voltage when it arrives.
All that said, I've been running the experiment as I write. The NC LED stopped lighting as it occasionaly does, because the relay tends to lift out of the breadboard. It's very sensitive to being solidly pushed in place. When I applied pressure to reseat it, both LEDs started working.
I think, as in I am not 100% sure, that the diagram in that book is wrong. The pin marked NC is common, the pin marked common is NC and the pin marked NO is correct. Look at how it is drawn.
[Edit] I can't find a relay like that in my box of bits. I am not confident of the connections, check them (unless someone else posts to say they are definitely right or definitely wrong).
[Edit edit]
When I applied pressure to reseat it, both LEDs started working.
