We finally have our first power graphs, and they are beautiful!
Originally we planned on measuring the compression of the spring using an LDVT that we made. Unfortunately, we ran into a few problems: a limitation on the number of data points the LabQuest can record, an annoying secondary wave from our output coil, and a broken sensor. Said sensor was a voltage probe, which we cut open and fixed, but it broke again. Micaela and I are both confident that the LDVT could work, but given the limitations of LabQuest and our lack of electronic resources here in Guwahati to remedy some unforeseen obstacles, we decided to brainstorm ways (no matter how unreasonable or crazy!) that we could measure the position of the spring:
- Mark the two ends of the spring with LEDs, videotape the spring in the dark, and use MATLAB to analyze the distance between the points (Shout out to Cindy Oh! A great solution, but not feasible if we want to get data on using rickshaws in real situations where the sun is shining.)
- Similar as above, but use the video analysis capability of a program called LoggerPro to manually click twice per frame of the video, marking the top and bottom of the spring
- Attach pencils to each end of the spring and feed a long strip of paper along the spring as the pedals rotate (Like a seismograph…my favorite solution!)
- Attach a magnet to the bottom of the spring and measure the change in magnetic field as it compresses (We only have one magnetic field sensor, and we need that to measure angular velocity.)
- Use a graphite linear potentiometer, which is a variable resistor (We like this idea, but it involves more time than we can afford.)
- Attach a horn that changes pitch when the spring compresses, and record/analyze the pitch (Luke loved this idea…surprise!)
Since we had reached an impasse with the LDVT method and desperately wanted to come home with some good power data in hand, we went with the most feasible and simplest solution we could think of, despite it’s total lack of elegance: videotape the spring and use LoggerPro to manually click the points on the spring. Desperate times call for desperate measures. Micaela cranked out an awesome Python program that takes all of the data we collect and calculates the power put in to the rickshaw for the recorded ride.
To calibrate, we applied a number of constant forces (measured with our hand dynamometer) on the pedals and recorded the resulting spring displacement (found by analyzing an image of the spring), in order to map the relationship between force applied to the pedals and the compression of the spring attached to the chain. As our luck would have it, the hand dynamometer (which was working great several days ago) decided it didn’t like us anymore. Unscrewing the case revealed two broken wires. With no soldering iron, we tried to super glue the wires to the chip and learned that super glue isn’t a good adhesive when you’re trying to perform circuit surgery. Thus ensued a frantic excursion into the sea of Guwahati traffic and shops to find a soldering iron. With Jaswanth as our guide and translator, we quickly located a shop that sold us a $3 soldering iron with a clear plastic handle that more closely resembled a cheap plastic flashlight than a device used to melt wires together. After a few suspenseful minutes of watching the iron fail to heat up and wiggling its wire, Micaela was able to fix the dynamometer when the iron decided to work for a moment. After that, the calibration worked without a hitch.
Yesterday we finally finished the calibration and did a test run, and last night we analyzed the data. This morning, after making a few more measurements of the crank and spring, we inputted all our data into Micaela’s program and were rewarded with a gorgeous graph of time versus power. It’s great to finally have some good hard numbers and graphs to show for all the work we’ve put in over the past three weeks! We plan on doing more tests and analyzing the corresponding data for the rest of our time here.