question about light sensor

I am looking for a light sensor that measures outdoor sunlight, ideally from 0 ~ 200000 lux. The common mini photocell ranges from 0.1 - 1000 lux and the more expensive AMBI Light Sensor ranges from 3 - 55000 lux. I have seen some tutorial on using different pull-down-resistant values to modify the range. So for example changing a 10kohm pull-down-resistor to a 1 kohm will change the range from 0.1 - 1000 lux to 1 - 10000 lux. But the thing I don’t understand is why would the photocell still functions even though it is already saturated… Can someone help explaining to me please? And is there any light sensor out there can directly measure up to 200000 lux? Thanks!!

This flyer shows 2 parts
ISL29011
ISL29018
with max flux range of 64,000

Digikey carries them

Link to datasheet is hosed, I have reported it.

Maybe one of these will work
http://search.intersil.com/search?q=ISL29011&site=www&filter=0&btnG=Search&client=isilnew&output=xml_no_dtd&proxystylesheet=isilnew&sort=date%3AD%3AL%3Ad1&entsp=a__isil_policy&wc=1000&wc_mc=1&oe=UTF-8&ie=UTF-8&ud=1&exclude_apps=1&getfields=*

Thanks a lot! so the maximum is only 64000 lux... will it still work with direct sunlight? I have seen sensors like this: http://www.koubachi.com/products/outdoor/ which can sense up to 200000 lux, how do they do that?

S_sneeze:
Thanks a lot! so the maximum is only 64000 lux... will it still work with direct sunlight? I have seen sensors like this: http://www.koubachi.com/products/outdoor/ which can sense up to 200000 lux, how do they do that?

could it be that they have a (calibrated) filter in front of the sensor?

What do you mean by filter?

It is probably a film or glass cover over the sunlight sensor, and just attenuates the light intensity.

I would guess that it doesn't really have 200000:1 resolution. If I REALLY needed that much range I would use 2 or more sensors. And the high range one would be filtered. The Intersil device could possibly be used instead, but again, direct sunlight would saturate the sensor unless it was filtered.

I've been doing much the same but using the TAOS (now renamed) TSL257 and equivalent IR sensors. I like these as they are easy to use, just power and analogue output.
I'm not too interested in absolute light (lux) levels, but is it sunny, daylight or dark - the IR gives a good indication of the seasonal strength of the sunlight.
The TSL visible sensors are extremely sensitive and make good optical isolators or for picking up energy meter LED's. For analogue (non-saturated) use, they need to be heavily attenuated, and I use a length of translucent white nylon rod, 10-mm diameter as a light pipe and still have to keep the TSL edge-on to the pipe to keep in range. By adjusting the length of exposed rod, you can adjust the analogue value - you might still need an op amp to fine tune it.
The IR sensor is much less sensitive and needs to be placed closer to the light pipe.
I was thinking of compressing the range by exponentiating the raw data as the output falls off dramatically in low light levels.
Paulcets' idea of multiple sensors might be the best way of covering large dynamic ranges.
Commercial LUX meters all seem to have domed white diffusers - you could try half a ping-pong ball or the white cap from a bottle

Thanks Paulcet and tigger!! I would have never think of using filter if it wan’t you guys. I don’t know much about the filtering technique. So I am trying to make sure that I understand you guys correctly, is it like if I am using two sensors, one will be for the range of 10000 to 200000 ( the number is not exact just the idea), the other will be for 0-10000 and the higher range one is with filtering. There are things that you guys were talking about, like ping-pong ball and stuffs. Is there any good tutorial for the basic science and technique? I wanna know what material I should use and how to calibrate.

And I am doing this for a plant sensor, if that helps.

When "filtering" is mentioned, people usually think of color-selective filtering. What you need here, is what photographers used to call neutral density filtering.

Basically, you want to attenuate the light level by 50-80%. So that when the sun is shining very brightly, your 200000 lux gets cut to 20000 lux at the actual sensor. The sensor measures the 20000 lux and then you multiply that by 10 to get the actual 200000 before the filter.

You would probably need several sensors, and some way to calibrate them.

Thanks for giving me the right term! That actually helps a lot. I should now start going over some materials and come up with a plan with multiple sensors.