[SOLVED] Best Infrared beam emitter and sensor?

Hey guys,

I am looking to buy a IR beam emitter (is this known as an IR LED?) and a photodiode sensor to go with it.
I am going to be using them outside and they will be used for a people counting device. The setup will be thus:
When people walk through the beam, IR radiation to the photodiode is momentarily stopped, and an arduino, which is connected to the photodiode, will recognise this as a count. The arduino will then output that data to a GSM modem, which will send the info to a webpage server.
What IR beam emitter and photodiode do you recommend I buy? I've noticed that there are a whole lot of IR LEDs which emit IR at a whole lot of different wavelengths, and likewise for photodiodes and receiving the IR. How do I know what wavelength I should buy? I need something that'll be very durable, because it'll be used outside in rain, hail, and shine.

Thanks!!

I do not think you have to be concerned on wavelength of the Diode. In situation you have to look for the beam strength of the LEDs capable of letting the IR SENSOR be triggered on the distance of your passage. The Emitter and the Receiver must both be IR, as the Photo transistor of normal light will not work IR transmitter.
Hope this helps.

hey thanks Raza,
That's helpful.
I have just been doing some research and found that in order for the sensor not to be triggered by natural light it needs to only respond to a specific signal. Apparently the conventional signal is a beam pulsing at 38KHz. IR 38KHz receivers are cheap and readily available, so that's good news!
Now I need to find a IR 38KHz transmitter.

Liamlambchop said:

hey thanks Raza,
That's helpful.
I have just been doing some research and found that in order for the sensor not to be triggered by natural light it needs to only respond to a specific signal. Apparently the conventional signal is a beam pulsing at 38KHz. IR 38KHz receivers are cheap and readily available, so that's good news!
Now I need to find a IR 38KHz transmitter.

Click to expand...

You could use your arduino as emmiter using the PWM module.

A popular receiver is the pna4602 38KHz very common in TV's

Infrared leds can cover more than 15 feet. You can greatly increase range only if a small lens is put on receiver side. In your application, can you differntiate between two people walking side by side or close enough to not let beam pass through between them?

Consider using an ordinary remote control as a sender. The kind that controls a tv, vcr, dvd, etc.

It's everything you need in one package.

You can clamp a button down and it will blink the IR led around 5 times a second, with pulses at around 38 to 40kHz, continuously.

Your receiver must be programmed so the interruption has to last a certain amount of time to be counted as a person.

This idea won't work if a person could walk fast enough to get by between pulse trains, uncounted.

This will be intensive use of the IR LED. It may start to dim after a few days, unless they're being made more robust now.

Akenafab, that's a good idea. It'd be a very simple blink sketch that'd do the trick. I'll look up that receiver, cheers!

Alertlinks, where can i find a small lense that'd do the trick? Would it be worth putting one on the transmitter too?

Brad I had considered the option of using a TV remote. However it would be cumbersome. If required I could take the parts out of the remote which I need, but I think the arduino would be a simple solution.
About the receiver... I don't know what is included in a standard package. What type of output do they have? Would it b suitable to go straight into the arduino? Can you program them?

For the transmitter, you can use the arduino to generate the pulse packets. That is probably the easiest way. If you want to use discrete chips, there are numerous NE555-based circuits. This recent thread has a couple shown:

https://www.edaboard.com/threads/223433/

Basically, you have one 555 generate the 38KHz "carrier", then use the other 555 to turn that oscillator one and off at about 1KHz.

Don't forget to use decoupling capacitors as mentioned in the thread.

John

I once used an IR receiver module from Radio Shack. A few dollars. Very easy to use.

Metal cube about an inch on each side.

It detected IR pulses at 38 or 40 kHz. The output went hi when they were present.

It had three wires. Two for the power (5V I think) and one for output.

Cheers John, I think that I will use the arduino. It is the easiest.
Brad that is good to know. I was hoping that it'd be a digital output.

Does anyone know where I could get some lens' for the IR transmitter and/or receiver?

The light you are dealing with is considered near infrared (IR-A). Many materials are transparent to it. If you search on "IR lenses" you will get many hits. Here is one manufacturer that seemed interesting. There is a lot of background information on the site. If you go to "materials" you will find pages for its online store:

IR lenses, infrared lenses and UV lens from Knight Optical

Here is a table showing transparency of various substances. Fuses silica is effectively quartz. Glass and some plastics may also work. The lens on IR controllers is plastic.

https://www.edaboard.com/attachment.php?attachmentid=61942&stc=1&d=1317024972

Before you get too far down that path, though, I suggest you think through what you are doing. How are you going to tell whether the beam break is by one person or many? As you know, people tend to herd, and if the counting zone will allow more than one person to pass at a time, they will do that. Your counts will be wrong. If you constrict the path, then you won't need much range. In fact, you may just need to put the detector at the end of a narrow, IR-opaque tube to restrict its angle of view.

John

Edit: I forgot to mention Edmund Scientific/Edmund Optics as a supplier:

The BK-7(NBK-7) substrate it uses for near IR appears to be just an optical glass.

John, mate, that is bloody brilliant.
I don't know if I'll need to use a lense. I was just asking so that if I did need one i would have the information already, and hence not need to start a new thread :S

I really appreciate it

Johns detailed compilation can be summed in one sentence: All standard optical materials used for visible light work for near IR as well. Slight differences in refraction index can be mostly ignored.

The high sensitivity of RC IR receivers and the directivity of small angle IR LEDs allows at least 10 m operation without additional optics. But they may be affected by direct sunlight and interfering modulated sources. The classical light barrier design in constrast uses optics at both sides and less gain to achieve high reliablility.

What do you mean by a classical light barrier design? You mean putting the LED and receiver in tubes?

"Classical light barrier design" refers to how commercial devices are build. I mentioned that they have usually lenses and won't be easily disturbed by interferring light sources.