Marco, This article is old and dusty. Using LED's as photodetectors was in vogue 20 years ago. There is no reason to use a LED as photodetector even for the expence reason claimed in the article and nowadays the aray detector is detecting in one shot the whole spectral range. A white light source is guided through an optical fiber, passing the probe, difracted on a grid difractor and spreaded on the multi coupled photodiode array. In that way the whole light spectre is analysed very fast. The whole device is small like a match box and not very espensive for it's performance. greetings, Vasile On 8/17/07, Marco Genovesi wrote: > > About two years ago I started to play with an inexpensive light detector using a led as sensor. This argument has been discussed sometimes, this is one of the documents about an application related this topic: > > http://www.merl.com/papers/docs/TR2003-150.pdf > > In the referenced document the authors speak about a led photocurrent of 1uA (under strong light) and 1pa (in dark) while at normal room illumination is 50pa. > These values may be reasonable, the point non very clear is that They say of a leakage at the input pin of 0.002pa (the PIC used is a 16F876). > I have seen on the datasheet much higher values (up to a max of +/- 1uA) and also at room temperature a value of 1nA seems to consider. Now I have two questions: > > Are they in error or there is something true in this document? > > Eventually, could I obtain better results using a pic with a comparator (instead of measuring Hi/Lo threshold on the pin) ? > > > regards > Marco > > > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist