Let's Make Robots!

38khz 3-Pin IR Receiver and Wavelength of the IR LED

I just can't seem to find this through google except for actually looking up a diagram of the "full spectrum" so I need a confrimation here. The IR led's that glow a dim red are the one's in the 850nm range, correct? The ones that are "invisable" are the 940nm ones, is this also correct? Also, the 850 is a "stronger" light source? And finally, I can't seem to find a mention in the specifications of the receiver if it will respond to either wavelength.

I have been using this 3-pin "picaxe" IR sensor forever now, I can't believe these questions are only now coming up for me. I am working with 2 beacons now and I just now noticed that A) one could be seen and the other not, and B) the one that glowed dim red seemed to be "found" easier.  --Any thoughts here?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Here, I found this datasheet for TSOP3848 (which I bought, thinking it was the de facto standard). Fig 11 confirms: peak sensitivity at 940 nm. These ICs are done by Vishay and they offer a product selection guide or the professionals. It mentions their IR emitters as well. This might help you compare transmission powers and/or angles.


The 38 represents the carrier frequency BTW: 38 KHz.

"Picaxe" IR receiver datasheet is also here: http://www.rev-ed.co.uk/docs/TSOP18.pdf (there's a link to it in the "Picaxe shop"). So the "Picaxe" receiver does look like TSOP1838.

For some reason Vishay datasheet link doesn't work for me (neither from rik's post nor from Vishay's product page).


Yup, it seems our searches brought up pretty much the same data. I did the wiki on the spectrum and found the data sheet. I did not see a reference to 940nm myself --as you said, dunno if it is the same chip. I actually found a couple different data sheets for a few part numbers (this receiver seems to be made by a lot of people) and I only found one that mentioned 940nm --And at that, it was just a fleeting reference --no chart or anything including the wavelength of the sending unit.

At any rate, it is looking like the 940nm LED's are the ones I have been using. Oh, and I figured out why this has not come up before  --I have always used whatever radio shack had in terms of this sensor and the corresponding IR leds --Whatever they have been selling me has always worked (and matches each other). Now if only they would actually put some useful data about thier products on thier website or Heaven forbid, the packaging, I would be able to know what I had and thus could order more from a real electronics shop!

Hmm. Does this Wikipedia quote help?

The boundary between visible and infrared light is not precisely defined. The human eye is markedly less sensitive to light above 700 nm wavelength, so longer wavelengths make insignificant contributions to scenes illuminated by common light sources. But particularly intense light (e.g., from lasers, or from bright daylight with the visible light removed by colored gels) can be detected up to approximately 780 nm, and will be perceived as red light, although sources of up to 880 nm can be seen as a dull red glow in intense sources. The onset of infrared is defined (according to different standards) at various values typically between 700 nm and 800 nm.

IR is at the lowest end of the visibile light frequency spectrum. Remember your rainbow colors, Roy G. Biv (Red, Orange, Yellow, Green, Blue, Indigo, Violet). InfraRED is at a lower frequency than red, so the lower frequencies of IR are more invisible to the naked eye. Lower frequency = longer wavelength (frequency = 1/wavelength). So the longer wavelength IR LEDs (940 nm) should definitely be invisible.

You are working with the Picaxe supplied IR receiver? I've never worked with one, but some research shows they are the TSOP18 IR series of chips. You may be able to find a datasheet. I found one, but I'm not 100% sure it is the same chip you are using. Anyway if it is, it is most sensitive around 940 nm. At 850 nm it is probably down to 30% sensitivity.

Even if I am looking at a datasheet for the wrong chip, it is something in the same family. I suspect the receiver needs to be matched to a transmitter at very close to the same wavelength to be effective.

Hope this helps.