Let's Make Robots!

IR receiver feasibility?

Hello Y'all!

I have been scouring the web trying to figure something out. And I found your page, and you seem to be the group of people to talk to. So...

I want to make a bunch of simple circuits that will basically have wirelessly controlled switches. I am thinking the cheapest way to do this is to have IR receivers that operate at different frequencies that can be controlled with IR transmitters. So, my questions is, if you wanted to make, let's say three switches, would you use receivers? And say, some IR transmitters with different frequencies like 36, 38, 40kHz? Or would you use phototransistors and different wavelength IR diodes? Or radio?

I have found this page on Alibaba: http://www.alibaba.com/product-gs/541465116/IR_Receiver_led.html

Where could I find these chips in bulk? Seems like it might be a pain shipping them from China. But maybe the cost savings would make it worth it.

Well, basically, I am looking for advice.


Comment viewing options

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


I am seeing some controllers that look like they could be used here: http://octopart.com/partsearch#search/requestData&filters[category_id][]=4300&start=0&sortby[]=avg_price&sortby[]=asc

I am going to look around for some tutorials for setting up some test cases.


Thanks everyone!

Wow! It is so nice to see an active forum. Sorry I've been away for a few days.

So, this is more info about the concept. I want to around 1000 units, each with a few LEDs, let's say RGB. I want to be able to switch between the LEDs. My thought was to have a few phototransistors at different frequencies, one frequency for each color. So that if one were to shine an IR beam at the units then one would be able to "paint" with the corresponding color. The controls (emitters) would be embedded in flashlights with a corresponding color film, so red flashlight would trigger red LEDs, but really it would be an invisible light that would trigger the LEDs.

If the ATtiny can be had for cheap then it may make sense to use 1000 or so of those. But if phototransistors can be had for cheaper, and used as simple switches, then I may want to go that route.

I would probably still need some other unit to act as a switch so that only one color at a time would be triggered.

Thanks for the suggestions so far. I will update this thread as I work through the issue! Any further advice is appreciated!


building a huge paint display. If that is the case, you may consider something a bit more involved; a 'wand' that has a corresponding RGB led that can have its color adjusted and then a trigger to send that data via IR to the receiving wall. Just a thought. :)

PS - I failed to mention the wall component. With a uC on each LED/set of LEDs, you could read the incoming data and pulse the proper pins to get the requested color.

Or simply go the wii route. Make it work reversed. The wand does not signal the pixels directly, the wand knows where it points and updates the array in some completely different way. This way you could use fewer uC's controlling more pixels, and only need one receiver.


Yeah, I had thought of this, but decided against for a number of reasons. One is I would like for these units to be usable for many different projects. For instance balloon chains, or an outdoor ceiling made of paper plates. Also, I am not really up to wiring up a grid of 30x30 objects with unique identifiers. :D


But you are going to wire up power to them, right? Or will the units be battery powered?


CR2032s. I picture these being for temporary installations. Specifically I am creating something for OmniCorpDetroit's MakerFaire Preview Party (http://omnicorpdetroit.com/blog/). Last year I did a 1/3 mile long balloon chain that was lit with red LEDs. I want to do something that is interactive this year.

Instead of using different frequencies you could also make an IR master/slave bus where the master emitts IR commands and the slaves resceive the commands and act accordingly.

To achieve this you have to set up a protocol, where clients get an address and can read data from the bus if the own address comes in the envelope. Have a look at this one if you want to find out more.

You're basically flashing the IR LEDs in a binary pattern where address and commands are encoded into the pattern.

If you use a different frequency for each receiver then each switch will need it's own transmitter.If you use 38KHz for all of them then one TV remote can control 1000 switches independantly.

If the switches are used outside during the day then IR will be useless.

You need to give us more information about the intended use of the switches so we can give you good advice.


Ok, my issue is I want to make these as cheaply as possible, and I want to make maybe 1000 of them. So if I could just trigger these with the frequency, then I would save some serious money, right? It looks like including the logic component will cost an additional dollar per unit? Or can I get these for less?

Thanks for this tip though, I was not aware of the ATtiny! :)