Light Sensor disassembly report

From: Frank Angeli 
To: lego-robotics@crynwr.com
Subject: Light Sensor Dissection
Date: Wed, 16 Sep 1998 09:56:57 -0400

Hi People,

I dissected my light sensor last night and also made some measurements and tests.

1. As Kekoa mentioned the RCX sends a high ( 8 VDC) level for around 1.2ms and disconnects it for around 0.1ms which is when it reads the sensor. The 0.1ms voltage is proportional to the measurement where 0% is around 5 VDC and 100% is around 2 VDC.
2. The light sensor is closed y the six blue lugs you see at the bottom, four small ones at each corner which you don't see and a large tab used to hold the cable firmly. The put a drop of crazy glue at each corner and at the cable lug. I sliced the external lugs and removed the gray cover braking off all the internal lugs in the process.
3. The sensor and LED are placed on a circuit board which has an LM358D chip which if my memory is correct is an op amp. There are also a bunch of transistors, a couple of resistors and a 10uF 16V capacitor.
4. I removed the LED and tested the light sensor. WOW! With the LED removed I get a 0% to 100% dynamic range from dark to very bright light. A dim room is around 10%, pointing the sensor to a 75 watt incandescent lamp with shade some 9 feet away reads around 40% and an IR LED will deliver 100%. I will leave it like this forever!
5. I also made some tests replacing the small LED with another red LED and an IR LED. Unfortunately just by placing any LED in the circuit automatically biases and dampens the output. I thought most of the bias and damping was due to the closeness of the LED and sensor but most of it comes from the circuit design.

Note: Upon reassembly the sensor snapped perfectly and snuggly and apart from a few bruises works fine.