Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

eyesafety of an ir blink sensor

Status
Not open for further replies.

longjasp

Newbie level 5
Joined
Feb 12, 2012
Messages
9
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,357
hello,

i am doing a project of a sleepy driver warning system, i used ir led and photodiode as the sensor for the eyeblink detection. the project was done, but a problem exist, somebody said that our project is very dangerous from the eye because of the infrared rays transmited to the eye.. is that really true?
why is it that theres a lot of related project was done since it is not safety? how will i defend such issue>? really need ur opinions to defend our project. below is our full schematic diagram of our project.
for design ckt.JPG

the led attach to the rb0 port, represents the buzzer for the alarm indicator.
 

The answer can be only given by relating the involved optical output power to safety regulations.

What's the IR LED type, what's the minimal eye distance?

A modulated light source would allow to considerably reduce the output power.
 

Re: photo diode sensitivity for the eyeblink sensor

UPDATE: As root post was changed after incorrect topics merging, I add the original question:

hi guys,

i'm doing an eyeblink detector for our drowsy driver warning system, the project was done and very functional during night. but very sensitive at days especially at noon since the photodiode detects the light from the sun. how will i control the sensitivity of the photodiode from the sun. how will i modify the project as stable and functional even day and night and in any light situation. i already attached potentiometer at the sensor but still unstable esp at very expose places from the sun.
Attachment 70559

the led connected from rb0 represents the buzzer as the alrm indicator.

i really need ur opinions and suggestions.. thanx in advance,

Try to use polarizing filters on both LD. Adjust angle of polarization on one of them to receive reflected light of your light source. It will greatly suppress the sunlight.

https://en.wikipedia.org/wiki/Polarizing_filter_(photography)
 

Re: photo diode sensitivity for the eyeblink sensor

Modulate the TX LED with a 40kHz carrier
 

Re: photo diode sensitivity for the eyeblink sensor

1.

It may work to install a second photodetector to be a reference of ambient light conditions. Compare the output from your monitoring device to the reference level. It becomes easier to detect small changes from the monitoring device.

2.

Consider subtracting all DC component from your photodetector. A capacitor will block DC. Then you will be amplifying only the AC component. This will provide greater sensitive to MOMENTARY changes in light.
 

Re: photo diode sensitivity for the eyeblink sensor

As klystron said, modulate the transmitter AND filter the receiver for the modulation frequency. I am sure he simply forgot to add the latter aspect. Use a commercial product for the receiver like the Vishay TSOP3xxx units.

I made a device to detect movement of a 3 mm wire several years ago. It works in any light conditions, including full sunlight.

John
 

Re: photo diode sensitivity for the eyeblink sensor

@klug.. thnx for ur suggestion, but may i ask if how will i polarize the my ckt? what exactly the component or ckt will be added?.

---------- Post added at 20:14 ---------- Previous post was at 20:09 ----------

@klystron and jpanhalt.. how will i modulate the the transmitter and filter the receiver. what component will be added and where it will be connected exactly at the circuit?. big thnx for ur replies.. pls. help me in this problem.
 

Re: photo diode sensitivity for the eyeblink sensor

Longjasp,

you need a filter. These filters are wildely used in photography. It is a special glass.
Place it in front of your receiving diode. Then rotate this filter to achive maximal suppression of sunlight.

By the way, klystron's idea is also good. As jpanhalt has said you need to use ready IR reciever with filter.
In this case you need to feed transmitter diode by pulse current.

There is a circuit of IR repeater where you can find both receiver and transmitter of modulated IR signal:
**broken link removed**
 

Re: photo diode sensitivity for the eyeblink sensor

Be aware that sunlight is also slightly polarized (bees navigate by using it) and the windows of cars will affect the polarization too due to stresses in them. Thus, the direction of the car relative to the sun will affect the intensity of the reflected beam. You can use lenses from cheap polarized sunglasses to test the effect. Maybe a combination of polarization and pulsed IR will work best? In either case, I suggest doing it digitally and not depending on the intensity of the reflected light at all. In other words, your detector will be sensitive only to a coded refection. Sunlight will not have the appropriate code and will be rejected. It is just like the remote controller used for TV and other device.

Here is a link to one of many Vishay IR receivers: https://pdf1.alldatasheet.com/datasheet-pdf/view/26589/VISHAY/TSOP1738.html

Generally, one creates a pulsed beam using a typical IRED (IR-LED) at the center frequency of the device (in this case 38 KHz) and further modulates that into packets at 500 Hz to 1 Kz. The datasheet describes exactly what is needed. The acceptable limits are quite broad. That is most easily done with a microcontroller, but lots of analog circuits using devices like the NE555 to do the modulation are floating around. In a simple example, one has one 555 operating at 38 KHz and a second 555 pulses the reset pin of the first at 800 Hz.

John
 

Dark Blue wv(NM) min v max v le(mw/sr ) half-intensity angle
Transparent 850 1.5 1.8 80 mW/sr @ 50mA 24~30°

this is the specification of the ir led that i'am using, the ir led is placed just behind the sunglass at the inner part. does my design safe? if not, how will i cope with this? really need ur help. thank you
 

A simple way to guarantee eye-safety is to assure an optical power < 25 µW. It surely doesn't work with CW (DC supplied) IR LED.
 

This seems to be a duplicate of this thread:

Maybe they should be merged for efficiency?

John
 

With regards to eye safety, you need to get a copy of the regulations and do the calculations.

In terms of rejecting ambient light, firstly you need a modulated system as has already been suggested. I doubt polarising filters will help much. What you may need to do is supplement any built in optical filter in the photodiode with a narrowband interference filter. You have to be careful to make sure that the LED emissions are in the filter passband under all circumstances.

Keith
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top