Characterization device and gears


The original idea was to determine how the intensity of light affects some property of a sensor, in case of a CdS-photoresistor the electrical resistance. In order to set different intensities of light in a controllable manner I aligned two polarization filters (for cameras) in line with a LED and the CdS-resistor and read the resistance values at different angles.
Very soon I figured out that I wasn't even able to reproduce an angle with 1° precision by hand, so I thought of using a stepper motor to rotate a polarizer in well defined steps. For the first prototype I used a ready made gear set, a ball bearing, a piece of aluminum tube, plywood , angle joints and a lot of hot melt glue, besides a cheap stepper motor starter kit [1]. The images may give you an idea of how I proceeded and what it looked like.

lessons learned

  1. the concept proved working
  2. we can differentiate between angle differences of less than 0.1°
  3. the stepper motor kit is ok, as it provides 2048 steps or 4096 semi-steps per circle, but it cannot hold a position, it wiggles, I guess, due to the internal gearing.
  4. if gears are not absolutely centered, the motor cannot drive the polarizer gear evenly, this can even be heard. In effect we can see (slightly) asymmetric curves and displaced maximums.
  5. I need to improve my woodworking skills...

In order to overcome the issues described, I wanted to switch to a bipolar stepper motor with 1.8° step width and gears with a 1:9 ratio. I got in touch with Franz-Erich Schmitz through our local Repair Cafe. He was so friendly to introduce me to gears and provide me with STL files for according gears, which I share here, for reference: motor gear, module 0.5, polarizer gear, module 0.5, motor gear, module 1, polarizer gear, module 1

I had them printed out in FDM technique at a local store, but had to realize some drawbacks with that. Mainly the structures, especially for module 0.5 were not smooth and the teeth on the ground plate side needed to be cleared by grinding. Also the insertion of the aluminum tube didn't go too well. Check out the tutorial on 3DHubs for the different 3D printing methods [2] .

So for my instructable, I literally re-invented the (analyzer- and motor-) wheel and added the shaft going into the ball bearing and a cavity for the polarizer sheet. I had them printed out in SLS technique this time and this worked better (even though there IS an imbalance in the analyzer gear, which I need to compensate for with a flexible motor mount). You can find the freeCad construction file and derived iges files on instructables [3]. In addition to this I also switched to a motor with 0.9° step width, allowing for smaller gears and gear ratio.


  1. unipolar stepper motor kit (German)
  2. 3D Printing tutorial on
  3. Polarimeter with RaspberryPi on instructables