Yes, this topic has been covered many times before and yes I have read the posts on IR alarms. However, I still have quite a bit of questions and so I am creating this thread.

I want to make an alarm system that uses two IR LED's. I bought an IR transmitter LED (clear plastic) and an IR detector LED (opaque dark plastic). I want both the transmitter and receiver on the same circuit.

The idea is that the Tx LED sends out short pulses (so I can run more power through the LED without burning it up) and when someone walks in front of it, these pulses bounce back onto the Rx LED that is right next to the Tx LED.

With that said, I was thinking of connecting a 12F PIC or a 555 timer to a transistor and then the transistor to the LED (to run quite a bit of current through the LED to increase distance). Then I would connect the Rx LED to and transistor to amplify the detected signal and trigger a relay or something.

How does this sound? Let me know what you guys think and if you have any suggestions. At this point I don't have any PIR sensors, only two IR LED's.

I would also like to know how I can know if the voltage that is created by the Rx LED is high enough to turn a transistor "on". I have looked at datasheets but I must admit that I am a bit dumb when it comes to reading those things.

Thank you in advance to anyone that helps me out with this.

__________________
My Blog: My electronic circuits and other endeavours.

 

I would suggest that before you get to far into designing any kind of circuit that you do the following first:

• On your breadboard, power up your IR led, biased correctly of course and just let it run.
• On same breadboard, hook up your detector and configure it to act like a switch.
• Use your logic probe on your switch to detect changes, DVM might not see quick pulses.
• You may need to place some form of light barrier between TX and RX leds so they don't couple across breadboard.
• When you have this going, you can test how it will work by walking in front of it, wave hands, reflection off dark light materials, etc.

Well, that is where I would start off from. Point is, see if the idea will even work like your wanting it too. No need to build anything fancy to test the basic IR functionality

__________________
Pay it forward.

 

So wise. This is why I love you Mike I will try this out. One other question:
The specs say that my LED has a max power dissipation of 150mW, max forward voltage of 2V and max current of 50mA. This being said, how do I figure out how high a power I can run the LED if I make an astable 555 with a duty cycle of 10%? I was told on these forums that you could run more power through LED's if the duty cycle is very low because the spec is for average power. An equation or explanation on how to figure out how far I can push the LED without burning it would be great

__________________
My Blog: My electronic circuits and other endeavours.

 

Here is a circuit I have posted here before. It works at short distances, I am not sure how far away you need to detect the reflected IR. I also have a 555-based design, without the error correction that the 12F509 allows. Let me know if you want it.

It is modified from this source:

http://users.frii.com/dlc/robotics/projects/botproj.htm

Edit: Board dimensions are 0.65" X 3" -- I wanted it to fit inside a square metal tube.

Attached Images

	12F509_IRTxRx.png (33.7 KB, 24 views)
	IR brd_post.png (15.2 KB, 12 views)

Attached Files

Infrared Proximity Detector3.asm (4.8 KB, 8 views)

 

For your first test, just run the led straight from 5v through an I limit resistor. Something around 82 ohms should give you 35ma. You could go smaller but do not exceed 50ma.

__________________
Pay it forward.

 

I believe your target going to have to be pretty close to reflect enough ir energy to be noticed.You might try some sort of lens to help. We used proximity switches similliar to this and the distance varied with material, color and texture.

__________________
gerty

 

I used an IR receiver module, like found in TVs and VCRs, well anything with a remote. Most will work with a 38kHz signal. I used an AVR, ATtiny13, and get about 2 feet driving the IR Led directly from the MCU pin, no transitors, or even current limiting resistor. See Nigel's tutorial, that's what I used for the hardware. The receiver modules aren't expensive to buy, easy to find from salvage. Only problem is that, it can be triggered by any remote, of course you can send out a specific pattern, only switch when received.

 

Hi hugoender,

the most critical item in this setup (reflective IR-barrier) is the transmitting power of the IR-TX diodes.

There are many noname IR diodes on the market. They all have in common a pretty low mW/sr value. (settled at 1 to 50mW/sr)

The OSRAM LD 274-3 has a comparatively high output of max 110mW/sr. Using three of them in series (forward voltage = 1.3V) you might achieve 300mW/sr (to be on the safe side), which should work reliably over distances of 10 to 15 feet.

Metal sheets reflect a lot of energy, but skin absorbs most of it. So that has to be taken into consideration.

Using PWM at duty cycle of 20% you might increase the forward current beyond the allowed 100mA per LED.

Use an astable MVV with adjustable frequency (or do that using the PIC) and PWM output. Receivers work best at 38KHz (TSOP1738).

Here is the data sheet of the LD274-3

Remark: If the receiver output is in the steady high state, change location where there is no ceiling!

Regards

Hans

Attached Files

OSRAM_LD_274-3.pdf (224.1 KB, 8 views)

 

Rather than trying to detect a reflection from a human, why don't you put a small mirror on the opposite wall and bounce the IR beam off that. Then you would trigger on the LOSS of the IR signal instead.

__________________
--- The days of the digital watch are numbered. ---

 

Why have a trip based on “reflection”? Just send your modulated IR to two IR receivers and have your alarm trip when the receivers lose a signal. You can use a comparator to set a delay (RC) 25Ms is typical. Use a logic gate like an “AND” gate so that the alarm is only tripped when both IR beams or obstructed simultaneously. The delay as well as the conditional trip “two beams “will help to reduce nuisance trips i.e your neighbours cat or dog.

The TSAL6100 has 1000 Mw/Sr at 1A pulsed current,a good detector for wavelength matching could be the PNA4602M. Both are inexpensive and widely available. The TSAL6100 provides a derating graph for the ambient temp you expect to operate your device in.

One problem you may encounter when using dual beams is one beam may flood both receivers this can be corrected with optics, or vertically separate the beams with enough spacing. You can use the tube from a dark BIC pen to help culminate the beam. Some experimentation would be required.

 

Thanks for all the posts guys. Still doing some research and testing on a breadboard. Will post back when I finish this phase of testing and have more questions.

__________________
My Blog: My electronic circuits and other endeavours.

 

Hugo, I was thinking, you could make your RX/TX section with some copper clad board as shown in pic, just solder clad pieces into place. This way TX and RX are isolated.

Attached Images

IRTXrx1.jpg (24.8 KB, 9 views)

__________________
Pay it forward.

 

I have made a lot of Beam Breaks with Logic IR modules, and I normally use a PIC with a hardware PWM port, and over-drive them with a lower duty cycle. If you have a few IR modules, you can write the software to prevent false detection, ie, only alarm if sensor 1 AND 2 are broken etc.

The module usually has a pullup resistor, and the presence of a steady beam, pulls it low, until broken, then it will swing high. I have found these very reliable, and never had probs with IR remotes affeccting them

Jim