It's interesting, seen it before but I would think a photosensor designed for such a thing would work better. Most bots use CDS Cells anyway.
LEDs also work like a capacitor with light when reverse biased. **broken link removed** **broken link removed**
This method of using a 'reverse' LED as a Photodiode has more grey 'whiskers' than Santa Claus!..
As I scan the web, I am suprised at the number of very old 'new' ideas that people post as being original work.
There's an old saying 'that if you dont study your history you are bound to repeat it'. I would suggest that anyone who thinks he has a novel/original idea should take a little time to check patents or design release documentation, before he puts a 'Made in XYZ' sticker on it.
Regards
Hope your link up with 3v0 is now on line,
The article says that the signal from the LED is weak. It says to use an opamp that has an ultra-low input current and a 320M or 30M feedback resistor. They used a rail-to-rail Cmos opamp with an extremely low input current and a gain of about 3 million.
So why don't you use a rail-to-rail Cmos opamp?
The input resistance of an LM386 is only 50k and its DC gain is only 20.
leds as sensors is one of the founding topics of my blog. I was interested in using them as a human interface, and I don't think they have much application there. However, using them as a machine to machine or machine to environment interface seems a lot more practical and this rgb sensor is a great example of that.
I'm not sure all the op-amp overhead is necessary here, I'd be inclined to just connect the RGB directly to the analog inputs of a PIC, and use a high power warm / neutral white led (cheap white led has too much blue, it would skew the results) as the sample illuminator
Bill is right though, you can buy analog or digital RGB color sensor chips fairly cheap and they work awesome. just drop one on your robot's I2C bus and thats that!
hi,
David Cook has been kind enough to send me an email regarding his web page regarding the LED used as a Photo diode.
My earlier thread may have given the impression that I was saying that David was claiming his publication as being an original idea.
This in fact is not the case, I was referrring to the LED/PDiode technique only, his web page clearly credits the original source.
hi Hero,
He emailed me after reading this forum and I emailed him back.
It was my idea to add my last post, with the intention of clearing any possible misunderstanding.