Parallel LEDs - resistor needed?

[Cirkit]

I'm sure this question or similar has been asked asked before but I just wanted some clarification.

If I am connecting 7 LEDs in parallel with a forward voltage of 3.3V each to a supply of 3.6V, is a protection resistor needed? The datasheet mentions a maximum forward current of 25mA per LED.

The 3.6V is supplied directly from the output of a 358 Op-Amp.

 

[(*steve*)]

Yes. 7 of them.

 

[OBW0549]

If I am connecting 7 LEDs in parallel with a forward voltage of 3.3V each to a supply of 3.6V, is a protection resistor needed? The datasheet mentions a maximum forward current of 25mA per LED.

If you want to drive LEDs in parallel, it's usually best to put some resistance (enough to drop a hundred mV or so) in series with each LED to make their current consumptions more or less equal; otherwise, unless the LEDs are well matched you will end up with uneven brightness.

The 3.6V is supplied directly from the output of a 358 Op-Amp.

In that case, you're out of luck: an LM358 cannot supply more than about 30-40 mA, and you'll need a lot more than that if you want to run 7 LEDs at 25 mA each.

 

[Cirkit]

I'm not too worried about uneven brightness. I tried connecting the LEDs directly and it seemed bright enough. Is there a danger of the LEDs burning out? Since all the LEDs have already been assembled in parallel, it's not possible to add a resistor in series with each now. Should I still use a low value resistor with the parallel string?

 

[BobK]

If you are driving them from an LM358 it will be overloaded and unable to source the current the LEDs want. It is a bad idea, but you can probably get away with it if you think they are bright enough. Fortunately, the LM358 is rated for continuous short to ground, so you will not damage it.

Check the voltage across the LEDs when you are driving them and / or the current into them. I suspect it will be well below the rated max.

Bob

 

[Audioguru]

An LED is a semiconductor, not an incandescent light bulb. Light bulbs are easily made all exactly the same but LEDs have a range of forward voltage because some will be 2.8V, some will be 3.8V and many will be 3.3V.
If you connect a 2.8V LED parallel to a 3.8V LED then the 2.8V one will hog all the current and probably burn out but the 3.8V one will not even light until the other one is burnt out.

The datasheet of the LM358 and for every other opamp shows its range of maximum output current. For the LM358 it is a minimum of 10mA going high and is 5mA going low. The LM358 might get too hot driving an LED if its output current is high and its supply voltage is also high.

Maybe you are lucky that the voltage of all your LEDs is almost the same. Then the LM358 does not have enough output current to damage them without a resistor in series with them. Add a resistor if the LM358 gets hot.

 

[davenn]

I'm not too worried about uneven brightness. I tried connecting the LEDs directly and it seemed bright enough. Is there a danger of the LEDs burning out? Since all the LEDs have already been assembled in parallel, it's not possible to add a resistor in series with each now. Should I still use a low value resistor with the parallel string?

no, you need to redesign your board so that each LED has its own resistor

and I hope you have read and understood what 2 others have told you ...
here it is for a 3rd time .....

a LM358 cannot supply more than about 30-40 mA, and you'll need a lot more than that if you want to run 7 LEDs at 25 mA each.

 

[(*steve*)]

An LED is a semiconductor, not an incandescent light bulb. Light bulbs are easily made all exactly the same but LEDs have a range of forward voltage because some will be 2.8V, some will be 3.8V and many will be 3.3V.
If you connect a 2.8V LED parallel to a 3.8V LED then the 2.8V one will hog all the current and probably burn out but the 3.8V one will not even light until the other one is burnt out.

The datasheet of the LM358 and for every other opamp shows its range of maximum output current. For the LM358 it is a minimum of 10mA going high and is 5mA going low. The LM358 might get too hot driving an LED if its output current is high and its supply voltage is also high.

Maybe you are lucky that the voltage of all your LEDs is almost the same. Then the LM358 does not have enough output current to damage them without a resistor in series with them. Add a resistor if the LM358 gets hot.

The main issue is that bulbs resistance have a positive temperature coefficient and LEDs Vf have a negative temperature coefficient

 

[Cirkit]

Check the voltage across the LEDs when you are driving them and / or the current into them. I suspect it will be well below the rated max.

Bob

I measured the total current into the parallel string and it is 21mA. The voltage drops to 2.76V when connected to the parallel string.

After powering the LEDs for about an hour, the temperature of the 358 Op-Amp rose by 2ºC above ambient.

 

[(*steve*)]

If you're OK with 3mA per LED then your job is done.

At that low current, it could all safely go through 1 LED, so no great risk there either.

 

[Cirkit]

Great, thanks for everyone's advice on this. I just wanted to make sure that I wouldn't burn out the LEDs. I will heed the advice about separate resistors for each LED in future!

I'm curious how the current is being limited though, is it as a result of the load and the voltage drop on the output of the Op-Amp?

 

[(*steve*)]

I'm curious how the current is being limited though, is it as a result of the load and the voltage drop on the output of the Op-Amp?

The op-amp has an output impedance which limits the maximum current.

It's works pretty much like a resistor in series with your LEDs