Hello Alan,

> First, I would dispense with the 10uF capacitor, this is likely to have
> leakage current that will upset the reading, especially as the resistance of
> the LDR gets higher under low light conditions, which is probably where you
> want to be most accurate.

     I was already thinking about the needs of that capacitor. I've 
removed it and the behaviour of the circuit seens a bit better.

> Second, the circuit you show will have a very non-linear output against
> light level. Again it will show the greatest change at reasonable light
> levels, and be worst at the dark end, which again, is probably where you
> want the most accuracy.

     Yes, this is why I have that formula [lux = (63750/adcResult-250)] 
on the software. The resistance of LDR is almost linear with incident 
light. Something like R=500/lux (for R in kOhms).

     So, luminous intensity on the output of the voltage divider is:
	Vo=(4.7/(4.7+(500/lux)))*5

     And with some algebric manipulation and scalling to ADC reading 
values (0-255) I've came with:
	lux = (63750/adcResult-250)

     Obviously, I'm mistaken somewhere...

 > Whatever circuit you use, you may need to have some tables in your
 >program that you can use to stepwise linearise the output.

     Yes, but I preffer a more general formula. There is a PIC18F252 for 
this simple task, so I have plenty of processing power to make calculations.

     Thank you,

     Brusque
-- 
---------------------------------------------------------------------
Edson Brusque                     C.I.Tronics Lighting Designers Ltda
Research and Development                   Blumenau  -  SC  -  Brazil
http://www.suporte.ind.br/ryan/netiqueta.htm     www.citronics.com.br
---------------------------------------------------------------------
_______________________________________________
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist