This was the first almost entirely electrical project I got to work on at Emphysys. I was part of a project that involved the extension of a rod, and used a motor to move the piece out in a screw action. In order to get precise data on how far the rod has lengthened, we needed to measure how many rotations the motor turned, and then extrapolate. When first ideating this project, it started like many do... at a whiteboard with a fellow co-worker.
​
I know it looks like a bunch of scribbles, but trust me a LOT of knowledge was transferred during this discussion. I learned about package sizes for electronic components, how everything should be wired together, knowing which resistors to use and we also talked about how the encoder works fundamentally. The circuit consisted of three phototransistors and an LED in the center. Technically only two phototransistors are necessary to get accurate readings, but we wanted a third one for redundancy.
Below some of the notes I took at the time of doing this project, using a tool called Notion. This is an example of the type of documentation I like to do for my work; it helps me navigate all the dense information easily.
The view is a little bit wonky because I had to capture the images through a microscope. The components I was working with were extremely small and required a lot of precision to place and solder so that it was still functional after. I burned out one LED on accident, so once this was done it was a relief seeing the expected resistances between connections on the multi-meter. Below are some images showing one of the phototransistors next to my fingertip, along with the completed final circuit.
With the circuit finished and connections tested with the multimeter, it was time to do a full test with this setup. My goal was to be able to measure the rotation of a motor shaft, and see if it is accurate to the motors rated RPM, specified in its data sheet. I found a fan from an old project that was in our spare parts storage, and modified it slightly to sit over the sensor at a certain distance. I connected the fan to a power supply, and connected leads from each phototransistor to an oscilloscope.
The output voltages for each sensor were too high, resulting in oversaturated readings of current. To fix this I soldered each 10K output resistor in parallel with its own respective 1.2 K resistor for an equivalent resistance of ~1.05 K ohm
​
Ended up getting pretty good readings and the distance from the spinning disk on the fan and the sensor was roughly 5 mm. Optimal readings will be at about a distance of .5 mm. I ordered some small motors online that spun at the same RPM as the motor in the implan would be, and fashioned a small fixture to hold it out of aluminum (this was before we got the 3D printer).
​
Moving on to smaller encoder disk sizes, and closer distances to the encoder LED, meant that I was able to progressively test the performance of this circuit.
Below are images showing oscilloscope readings from the last few rounds of testing. The data collection process for this experiment was pretty informal, since this was just a proof of concept more or less. Further implementation and feasability testing of a motor encoder was done in the benchtop prototype phase of this project. With the data I collected, it was clear to see the efficacy of the discrete circuit. When working out the math, the period in the below images yield an RPM in the 12,000-14,000 range, aligning with the specified RPM of the motor in its data sheet.
