And We Felt Bad About 11 Days Last Time…
So the last blog entry was about how guilty we felt about not updating our blog for 11 days. News flash! We beat the record. It’s been 14 days this time, but the guilt is significantly less. Last time, our reason for not updating our blog was because there was so much going on that we didn’t have time to. This time, however, it’s the opposite. Barely anything has happened since we last updated. We’ll explain.
So we left off last time thinking that we should insulate the sides of the specimen for even more accurate numbers. We were very wrong. Insulating the sides didn’t help; it made it a lot worse. Why? Thermal bridging. Before, when we ran the tests with the sides open to the air, the hot side loss a little heat and the cold side gained a little heat, but all was well and the change was not drastic enough to fret over. But by insulating the sides, we allowed the trapped heat that would’ve been lost to the surrounding to travel along the inner surface of the side insulation and heat up the cold plate, causing the temperatures to be off and messing up the data. Seeing how horrible the idea was, we threw it aside and ran the tests sideless like before.
Tired of documenting and calculating data by hand, Leanna created a spreadsheet that did all of the calculations for the thermal conductivity by inputting four temperatures. 5 trials were conducted for each weight (25 to 200, in increments of 25). The results were not so pleasing. This time, we found that as you increase the weight applied, the thermal conductivity increases as well. So we were getting the expected value at around 25 or 50 pounds, but any more weight than that and the percent error got up to 35%.
Finding this strange, Professor Kobus taught us how to calculate uncertainty of measurement tools to see how accurate of an experiment this was. Using this new knowledge, we found the experimental uncertainty to be over plus or minus 45%. This scared us a little, but it was also relieving to know that our messed up numbers were within the uncertainty range.
After finishing up all of the test with the Nylon specimen, we moved on to another type of specimen. Michael went down to the Machine Shop and cut us a new Aluminum specimen. Through a few trials with this new specimen, we learned that it was not possible to yield correct values from our apparatus because of conjugate heat transfer. Our experiment is limited to the building’s hot water and cold water and a large change in temperature is not possible to control using water. Knowing this, we theorized that if we were to replace the hot plate with patch heaters, we would be able to heat the specimen up to a higher temperature, yielding a larger change in temperature, and maybe get correct results in the process.
The patch heater is currently shipping so until then, experimentation and data will come to a halt.
Work never ends, because as we wait for the patch heater, we are going to be redesigning our box for easier insertion of the specimen. The new finished design is pictured in the blog image.