IR LEDs too hot

I’m trying to get the IR LEDs temperature down when using 5VDC, currently at 5VDC they heat up to about 50-60 °C (also the current for this case is higher) which is no good because they are directly near the monitor. I would like to have the setup run at a more common voltage such as 5VDC and still be bright enough but not extremely hot. The total current is about 0.6A.

I currently use a adjustable 4VDC power supply as a work around which is very uncommon voltage.

See the setup here.

Components used:
4 PCBs with 3 IR LEDs and 3 resistors on each so in total 12 IR LEDs and 12 resistors they are about 0.6A (4*0.17)

Maybe these IR LEDs can be used to be bright and run at 5VDC without being extremely hot or would I need a different resistor?

Hi Tom
According to the specs that come up with these links your resistor value is spot on for 70mA foreword current @ 5V supply Max operating temperature is quoted at 100DegC but the LEDs should only be dissipating 112 mW (1.6 * 0.07) which I don’t think should get that hot.
0.6A is too high if you are only powering the LEDs. Should be more like 0.21A.
Are you powering the LEDs only? Measure the voltage drop across each LED. According to spec sheet should be 1.6V. Make sure the resistors are the correct value. You will have to measure these as measuring the voltage across them won’t mean much unless you are sure they are correct as the LEDs will have pretty much the same voltage across them with quite a wide range of current values.
Unless you are powering something else as well 0.6A is way too high just for 3 LEDs. Something not quite right.
Cheers Bob

1 Like

Yes, only powering the IR LEDs and nothing else.

Ah sorry, I missed the crucial information!
4 PCBs with 3 IR LEDs and 3 resistors on each so in total 12 IR LEDs and 12 resistors they are about 0.6A (4*0.17)

Just stuck don’t know how to go about it


1 Like

Where does the “0.17” come from. If you are using 5V and those LEDs and resistors each LED will draw 70mA (0.07A) so total 12 LEDs should be 840mA (0.84A). They are high power LEDs after all and maybe that running temperature is normal.
Are you running these continuously. They are described as IR communication devices so their normal duty cycle would be quite small. Probably less than 50% which would reduce the temperature considerably.
That 70mA is maximum foreward current so maybe you don’t need that much. You may find you get nearly as much brightness by halving that or even 50mA per LED. LED current is set by changing the value of the series resistor. To decrease current to 50mA increase the resistor value to 68 ohm. for 35mA 97 ohm.
By the way, these values are usually not super critical. There could possibly be more variation in the LEDs than resistor preferred values.
Cheers Bob


0.17 A is the reading of a single spot, not all four. This is smaller than the rated 0.21 A but this isn’t a uncommon deviation i think. Also the rating is the max current, not the usual current. I measured the current with a laboratory power supply.

They are running all the time, like you said maybe the type of LED.

I will try increase the resistor value, if I decrease it the resistor will take on the heat instead which I’m trying to avoid such high temps.

Hi Tom
My suggestion was to increase the resistor value. Decreasing it will push the LEDs over Max current (5V supply) and get hotter or even fail. You might find you don’t need anything like max current to get adequate brightness. Depends what you are using them for.
If you have a Lab power supply it should have current limiting so trying this out without changing resistors should be easy. Simply set the system up with supply voltage of 5V then use the current limiting function to reduce the current and observe what happens. The LEDs should run cooler. The power supply voltage will drop but take no notice of that. When you arrive at a point where you get acceptable results note the current and then calculate the resistors required to drive these LEDs at 5V or whatever voltage you are using.
If you are using a digital multimeter to measure current DO NOT use the mA scale. The voltage drop across the meter will likely be significant and throw your calculations way off. DO USE the 10A range. The drop will then only be a few millivolts and the resolution is usually OK.
Cheers Bob