Photodiode Solar Sensor

[kmason14]

Hi all, I am working on a project at Univ. of MD for high altitude-balloon research. My specific project is to make a "solar detecting" sensor(s). I am using the photo-diode in "photo-conductive" mode which means I am reverse biasing it, and when light strikes it, the photo (leakage) current increases (R of diode decreases). I am using it with an LM2902 quad op-amp (see below) and a 30K resistor.

My concern is choosing the value of the resistor. I chose 30K based on something I read online about these circuits. The idea is I want a LOW output when it's in ambient light and HIGH when it's directly in sunlight. I have a direct relationship between solar lux and photocurrent Ip from the data sheet, and from wikipedia, I know my range of lux (for morning balloon launches) will be about 400 for sunrise to 10-25,000 for overcast midday and 125,000 full sunshine at noon.

Combining these two facts, my range of Ip will roughly be about .04mA - 12.5mA for sunrise - full sunshine @ noon. Now, I will need to amplify the output since this is small, so it can be read by the Arduino.

I will (hopefully) be outputting to an Arduino ('Freeduino' actually) microprocessor board (recommended input 7 - 12 V). I will also pull 5V directly from the Arduino to supply the reverse bias voltage Vr. In addition I was thinking of adding two capacitors onto the top and bottom of the op-amp to reduce noise.

Any suggestions/help would be great! Thanks


Here is my photodiode and op-amp, as well as a basic idea of the circuit:


( sfh2505 )


LM2902KDR Texas Instruments Op Amps

https://www.aptechnologies.co.uk/PDF/Photodiode%20Typical%20Circuits.pdf

 

[MikeMl]

I'd believe 0.04 to 12 uA.

Assuming your ADC range is 0 to 5V, why not do something like this: The current source represents the photo-flux induced reverse leakage Id, and is the independent variable in the simulation. If powered from 5V, the opamp must be rail to rail output.

If you cant figure out how to negate the ADC reading in software, then hook it up as shown in the second circuit.

If you want to maintain ~5V across the photo diode while subjecting it to illumination, then hook it as shown in the third circuit.

 

[kmason14]

The third circuit looks like what I would need. Cool simulations what are you using for that?
I got the milli-amp numbers from the data sheet of the diode, the graph (attached) showing photo-current vs. lux


Although I'm not sure what the max photo-current is with a 5 V bias.

thanks for the help

 

[MikeMl]

It is not milliAmps; it is microAmps.

LTSpice. Free download at Linear.com

 

[kmason14]

1,000 µA = 1mA

from the graph, at 10,000 lx = 1,000 µA (1mA), so I extrapolated the rest with that same slope.

But is that the max value from the graph or do they just show only that range? That's where I'm kinda confused.