Photosensors

[irkgreen]

Hey all,

I'm trying to detect the status of a device's LED via the ioBridge.

My dad happened to have a box of miscellany that had 2 photosensors.  One is about twice as big in diameter.  The smaller one is this:

http://www.datasheetarchive.com/CL703L-datasheet.html

The other looks similar but has "21K288" on the side and I can't seem to find a datasheet for it.

I've wired them up to the DI and GRD on two different channels.  The large one works great.  It detects room light and detects an LED when placed on top of one.

The smaller one is either bad, or I am wiring it wrong.

I'll admit I am shooting blind here.  I'm not sure what the circuit should be to accomplish this. 

Ideally, I'd like to find a similar sensor from digikey so I can order a few of them.  Although, I only need two working so if putting the right value resistor on the CL703L would solve it then that is cool too.

Any help would be appreciated.

Thanks

[iobridge]

If you have a multimeter, you can connect it directly to the black and red leads. Measure the resistance as you change the light level. This would be  a quick test. This is of course assuming it;s a photo-resistor.

[irkgreen]

It is a photo-resistor I think).  The large one (the one that works) reads 0 ohms when no light and 200 or so when lighted.  The other reads 0 either time (I assume it is bad).

[texasclodhopper]

The CL703L is a "Photoconductive Cell". You can find plenty of info on that type of "light dependent resistor (LDR)". "CdSe" is the material it is made of (cadmium selenide.)

One place with a few descriptions is http://answers.yahoo.com/question/index?qid=20090120053413AACBtXi . That might get your thought process going on basic electronics.

Your measurements seem to be opposite to what I would expect, because LDRs have an inverse relationship to light intensity; more light creates lower resistance.

[jason]

irkgreen,

Are you using the photosensor in a "voltage divider" arrangement?  I think that might work better than just connecting straight to the DI and ground pins.

http://itp.nyu.edu/physcomp/sensors/Schematics/VoltageDivider

 

[irkgreen]

Ah, that makes sense.  I will try that, although I am pretty sure the 'small' photosensor is bad.  I have ordered these from digikey:

http://search.digikey.com/scripts/DkSearch/dksus.dll?WT.z_header=search_go&lang=en&site=us&keywords=PDV-P9002-1-ND&x=0&y=0
http://search.digikey.com/scripts/DkSearch/dksus.dll?WT.z_header=search_go&lang=en&site=us&keywords=PDV-P9200-ND&x=0&y=0

What value resistor would be recommended?  At 5 volts and 20 mA, seems like 250 ohms would be fine.  Basically, I would just need impedance less than the photoresistor (500 K or 5 M) so that current flows to ground when the photo resistor detects light, right?  Otherwise, current will flow through the photo resistor when there is no light.

[jason]

I'm not sure the perfect resistor value to chose for your case.  CdS photocells respond to differently to various wavelengths of light.  So the color of the LED factors in there too.  If I remember correctly, a CdS photocell can barely even see green light.

For our own ambient light sensor, we use a 100k potentiometer with this photocell (range of 0.5M to 16k): http://www.advancedphotonix.com/ap_products/pdfs/PDV-P8103.pdf

The pot makes it easier to hit that "sweet" spot.   

[irkgreen]

I got these in:

http://search.digikey.com/scripts/DkSearch/dksus.dll?WT.z_header=search_go&lang=en&site=us&keywords=PDV-P9002-1-ND&x=0&y=0

and I hooked them up to GRD and digital in and they work great with my LEDs.  Since the ioBridge has 25k pull up resistors, there's no reason to create a voltage divider for these sensors, right?  So far it seems completely reliable.  I guess I fail to see the reason to complicate it.  But I would do it if it would make it more reliable or safer.

[jason]

The extra voltage divider resistor would be required only if you were using the analog input.  But you're correct, since the digital input already has a pull-up, you've effectively made a voltage divider with just the addition of the photocell.  No need for any other parts.

[irkgreen]

I actually am now using some photo-resistors from digi-key that have a pretty wide swing.  These:

http://search.digikey.com/scripts/DkSearch/dksus.dll?WT.z_header=search_go&lang=en&site=us&keywords=%09PDV-P8001-ND&x=0&y=0

I have it connected to ground on one end and "digital in" on the other.  I realize that a voltage divider would be a better way to approach this, but I'm curious if I am doing any harm this way?  It seems reliable so far.  Would this be damaging to the ioBridge in anyway?  I also want to test them in extreme temperatures.

If not, my thought was to use the sensors with a voltage comparator (LM339 since i have 4 sensors).  I'm hoping that what I have will be just fine though.  I also have  TSL12S's that convert light intensity to voltage that might work too.