The detector module is designed to output a logic level signal when a corresponding photodiode experiences a drop in light (i.e. is covered).

This is a light-adaptive photo-detector circuit based around the Maxim MAX944 comparator. They were nice enough to send me samples for prototyping.

interface connections

upstream: detector module --> IDC 2x5 male header --> (IDC 2x5 female header --> ribbon cable --> DB9 female) --> DB9 male --> control unit

upstream to control unit

IDC 2x5 male

  1. +5V
  2. start trigger OUT (active LOW)
  3. NC
  4. ready light (IN)
  5. GND
  6. detector out 1 (OUT)
  7. detector out 2 (OUT)
  8. detector out 3 (OUT)
  9. detector out 4 (OUT)
  10. NC

theory of operation

Warning: This detector module was designed for very high sensitivity. The detector has been tested to trigger with the shadow of someone walking by. Please adjust your hysteresis levels accordingly.

The photodiode charges up a capacitor to the ambient lighting levels. When the light level is suddenly reduced, the photodiode will turn off, pulling the comparator's positive input high. At the same time, the capacitor connected to the comparator's negative input remains at the previous lower voltage and the comparator output goes high. The schematic may be found here.

Note that the minimum ambient light level is determined by the sensitivity of the photodiode. On my setup, some super-bright orange LEDs overdriven with only 110 ohms of resistance is a barely sufficient light source.

To change the pulse length, reduce the size of the capacitor, but note that smaller capacitors require more rapid decreases in light level. at 33uF, the capacitor should be big (and leaky) enough such that even very slow triggers will trigger the comparator. I've tested with 10uF to 100uF, and anything above 22uF seems to work best. Essentially, increasing the capacitor size increases the RC constant which is how fast the circuit reacts to changes in ambient light levels.


As you can see here, here, here, here, and here, I don't think layout is critical for this circuit (new pictures taken with the camera's macro mode). The photodiode is the shiny white thing at rows 36-37. Hopefully, these pictures should be sufficient for me to recreate the circuit if I should ever lose the schematic.


The circuit runs at 5 volts. I've scoped it and from what I've seen, it doesn't oscillate any more. This shows the voltage levels at dimmer ambient light: the line on top is the capacitor voltage (which is higher than the voltage on the photodiode's cathode). When triggered, the photodiode's cathode voltage goes slightly above the capacitor voltage. This picture and this picture shows what happens when the ambient light levels increase: in both cases, the capacitor voltage is the higher one.

Just in case the light source is non-DC, I also tried putting a 0.01uF capacitor in parallel with the photodiode. It changes the reaction time of the circuit, but it's not too bad. I think that even a 0.001uF capacitor should be sufficient.

To make sure that I'm not getting any transient logic level changes, I've also tested the circuit by connecting the output of the circuit to the gate of a general purpose (note: not a sensitive gate device) SCR. The SCR stayed untriggered when it should which meant that any transients were shorter than the required SCR turn-on time.