How to use a photo-diode to detect infrared?

Sup dudes!

For an up-and-coming robot, I need to be able to look for infrared light. I have a photodiode (it looks similar to the one in the picture)from an opto-interrupter from an old car CD player and was wondering how to hook it up to my Arduino to sense infrared light. It needs to be an analog signal. I'm fine on the coding end of things, just not sure how to use it is all. Will I put it in a voltage divider? Will I connect it to the base of a transistor?

Thanks dudes!

photodiode.jpg

Anode to ground, cathode to analogue input plus a resistor of 100K from the analog input to +5V.

Having said that a simple photo diode is not very sensitive and you will probably need at least one operational amplifier to increase the sensitivity. Google for some circuits.

EDIT corrected advice.

What kind of IR are you looking for? Just generic IR light? Near IR or heat IR? Pulsed IR with data? It makes a difference.

Grumpy_Mike, what did you mean by "analog input to +5V"?

Thanks for the replies dudes!

Grumpy_Mike, do you mean to use the photo-diode in a kind of voltage divider? The photo-diode and the resistor in series with the analog in between the two? That's what I was thinking of doing. (Great minds think alike! XD) Thanks!

polymorph, the sensor only has to detect infrared given off by infrared LEDs. They're not transmitting any kind of signal, they're fully on all the time. The are just going to be put onto an object so my robot can find it by tracking the IR light.

Thanks!

It would be much better to have the robots pulse the IR LEDs and use a detector that only looks for the pulses. Otherwise, it is hard to pick up one lone source of continuous IR out of a constantly changing background of IR. Lights, the sun, etc.

If you choose the right frequency, you can use an IR remote receiver IC. Vishay has quite a good selection.

No matter what you use, it is unlikely to allow you to see more than one robot at a time. You could get around this by having the robots send short ID pulses at randomish intervals. Then only occasionally would two robots be sending at the same time.

Are you only using one photodiode?

Thanks dudes!

First off dudes, I've done a bit more research since my last post.

Now I'm gonna use particular frequencies of LEDs. One for the object the robot has to find, and another frequency for an infrared node that will act as "home." The robot will have two infrared sensors, both infrared phototransistors. They are both set to different frequencies. One will detect the infrared "Home" and the other will detect the targeted object. The two different frequencies will help the robot find the object and then navigate home again.

Second, I think you may have misunderstood my question. The robot won't be finding other robots or anything like that, not needing any kind of data. I think it would be good to use an LED similar to those in TV remotes. I also think phototransistors would be a better method of finding the infared.

What do you dudes think of that? Any advice?

Thanks dudes! XD

They are both set to different frequencies

No you can't do that, you might be using the wrong words, are you talking about modulating frequency or emission frequency.
So that is a lot you don't want to do but it is short on what you do want to do.

Use the same carrier frequency, but modulate with a signal that identifies which is which.

Then use a receiver for remote controls like the ones that Vishay sells.

Sup dudes!

Sorry, I'm not great at explaining things... :confused:

What I meant, Grump_Mike, was to buy some Infrared LEDs that outputted different wavelengths of light, and phototransistors that detected these two different wavelengths. I.e, home LEDs would be at 880nm and target LEDs would be at 980nm or something similar. The phototransistors would detect 880nm and 980nm respectively.

Does that clear things up? Would that work?

Oh, and polymorph, I was looking at some of Vishay's infrared stuff on Radionics, I'll probably go with those. Do you mean to use the same frequency/wavelength of LEDs but just turn them on/off really quickly at different intervals of time to decide which is which? If I could, I'd like to keep the Home and Target free of circuitry, and just have the LEDs hooked up to a battery in series with a small value resistor or something. Could I do this?

Thanks dudes!

Does that clear things up?

Yes :slight_smile:

Would that work?

Well ...
While you can do that, the different wavelengths are still detected by both types of sensors, only with reduced sensitivity. Whether there is enough of a difference in sensitivity is hard to tell, it very much depends on your application.

Modulating different wavelengths at different frequencies offers the best way to separate them. Vishay do receivers in the two different wavelengths as well as a variety of different modulation frequencies. However most of them are for data. That means that the modulation IR must itself be modulated with data otherwise they drop out. The TSOP4038 however will accept continuous modulated light but I think that 38KHz is the only one they do.