What am I doing wrong? I have 3 LEDs connected to 2 AAA batteries and the batteries become warm

Thread Starter

xbonet

Joined Nov 19, 2018
10
I turned a small bike light into a night reading lamp. It runs off 2 AAA batteries, and it has a simple circuit with
  • a chip-on-board that runs through 2 functions: blinking and steady (I'm not interested in the blinking one, so I'm thinking of getting rid of the chip, but it's kind of a hassle to dissolve the epoxy covering, I'm sure),
  • 1 cold white LED (though it has connections for 3), and
  • 1 resistor for that LED (100 Ohms 10% according to its colours, though when I measured it it's 3.3 Ohms... guessing there's something wrong with it?)
Anyway, I needed warm white light instead of the cold one, and I figured I might as well take advantage of the 3 LED slots and have myself a brighter light. I got rid of the resistor because my warm white LEDs (5mm - 3V - 20mA) would simply not light up when connected to it. I figured they'd be OK anyway, because the AAA batteries supply 3V and these LEDs have a minimum 3.2V forward voltage. (Here I didn't take current too much into thought, because as a newbie I'm still trying to wrap my head around that, but I figured I shouldn't have any problems.)

The thing is: when using the lamp last night, it started getting a little warm. Not too much, but enough to let me know there might be something not entirely right. It's not the LEDs that get warm, but rather the batteries. Now, I've checked it thoroughly and can't find any place where the circuit might be shorting out. My guess is that the weird resistor was limiting the current and that now it's simply putting too much of a heavy load on the batteries. I think it doesn't have anything to do with voltage unaccounted for (which in my mind would be a bit like connecting the positive and negative terminals on a battery, which also starts getting them warm as they short) because these LEDs require for more than 3V and they're getting 3V, so there should be nothing else left that a resistor would need to handle. Am I entirely off base here?

Thanks in advance for your help!
 

WBahn

Joined Mar 31, 2012
29,979
The only way that this can work at all is if that chip-on-board (COB) is a voltage converter pumping up the battery output to something that will drive the LEDs. So it all depends on what output current it is programmed for -- and that resistor you removed might have played a role in programming it.

It would really help if you posted a sketch of the circuit and as much information about that COB unit as possible.
 

cmartinez

Joined Jan 17, 2007
8,220
Hello, and welcome to AAC.

Questions:
  1. When you measured the value of the 100 ohm resistor, did you measure it while on the circuit, or separately?
  2. What kind of multimeter are you using? Have you tried using its current measurement function, if available?

Batteries getting warm is usually not a problem, as long as they don't get too warm. When they do get warm, it's because a significant amount of current is being drawn from them.
 

Thread Starter

xbonet

Joined Nov 19, 2018
10
Hi guys! Thanks for the lightning-fast replies! And thanks for the welcome!
Your speed caught me by surprise, so I'm only just coming back to the Forum.
Hello, and welcome to AAC.
Questions:
  1. When you measured the value of the 100 ohm resistor, did you measure it while on the circuit, or separately?
  2. What kind of multimeter are you using? Have you tried using its current measurement function, if available?
To answer your questions: (1) I measured it separately and (2) I'm using a Chinese multimeter I bought off Aliexpress (I think, or Gearbest) branded B-Side, model ADM02; but I do know it works rather well, and I did test out other resistors simultaneously after this one resistor gave me a weird reading, and all 3 other resistor were spot on their listed value. And, yes, I was using its current measuring function; it has 3: µA, mA and A, so I used mA.

And I do think the warming up of the batteries might be because there's now 3 LEDs pulling out current instead of 1. But I want to make sure, as I'd rather not (a) burn the LEDs and have to work in it again and (b) not have the think melt on my head at night! (Perhaps this should've been A )

And about both of your comments about the sketch... yeah! I'm sorry! I was about to include a photo of the circuit itself but I was sort of ashamed of the messy, workaround way I connected the LEDs and soldered it all... For what it's worth, this is it:



* Oh, I forgot there was that other SM component there, which I can't recognise. (It's right next to the lone big screw hole around the middle.)

(In my defence, the original soldering job wasn't that good either. And the resistor was literally placed just as I've placed the connecting wires: the lead going to the LED positive was soldered to the LED positive pin itself, hovering right over the ground pin.)

I've also "sketched" the circuit and how it was originally setup. Sorry for the crude sketch, as electronic sketches go; you were probably expecting a real electrical diagram but I'm still learning to understand how to read them, so making one is a little ways off.

This is the basic PCB. (As you can see, there are only 2 components on it, the second just says 09 DG.



These are the places setup for 3 LEDs, which I'm now using; though the only one that was on it originally was the central one.



And this is how the thing was originally. As you see, the central LED was soldered and the resistor I mentioned bridged input positive with the LEDs' positive.



As you can also see, there really isn't that much space for 3 resistors on the thing (at least not for the left-most LED), so in case I really do need a resistor, what other option do I have? I know having just one resistor for 3 parallel LEDs is a real no-no, but by what I can see, it seems that's the only option, as I realise the resistor shown above is also potentially connected to the positive pins of the other 2 LEDs...*

(* Having said this I have now realised it was absurd of me to connect 3 wires, one to each LEDs' positive pin, as only the one wire, setup in place of the resistor, would've accomplished the same thing... )
 
Top