Let me rephrase. What dictaes the log in your formula? How would you derive this formula?
Light intensity fall as 1/square of a distance. I can bet you a beer on this.
Added after 4 minutes:
prodigyaj said:
no one mentioned about camera flash lights ? how do i elimintae that ?
well ok ! fine here it goes , then sensors are being used for external purposes as a part of automated robots , which will be in public grounds , so care has to be taken of all these factors
In public grounds most of the surfaces are not white and they will be at random angles. Light attenuation method is not going to produce any accuracy at all. Choose a different ranging method.
Guys what is this thing of camera flashes going on. I'll put the requirements clearly as follows:
1. Sensors will be at a distance of 3cm to 6cms from the reflecting surface.
2. The reflecting suface will always be white.
3. The sensors will be perpendicular to the surface at all time.
4. The sensors may have to work in presence of ambient light (not direct or bright
sunlight), so filtering of ambient voltages is almost necessary.
So guys please put forward your suggestions for this prototype now.
Looks like our friend prodigyaj made a good attempt to hijack the thread, although it lead to a useful discussion. I think, all of the key suggestions have been posted. In short: modulate+filter, make a calibration curve to find the distance from photodiode current.
Although I may have hijacked the topic , I think moving round and round in the topic its back to square one. I think even Mr.Correa the thread starter would agree to the fact that the thread just drifted and eventually didn't get any concrete output from this thread. I hope you take this as a constructive criticism and instead of indulging in petty fights we must and proving other as wrong we better give out correct data.