Testing Overview
The final testing of TCoMS will take place during the CWU spring quarter of 2021. Prior to testing if TCoMS is able to accurately measure thermal conductivity, the basic design requirements will be checked including the weight, size, and power supply choice for the device. To test the functionality of TCoMS, materials with known thermal conductivities will be used and the obtained valued compared to the expected value. If the difference between these values is less than 5%, then the main goal of this project has been accomplished. The testing data will be recorded on a premade data sheet and input into the final report. Due to changes in the operations of TCoMS, the temperature data was imported into an Excel file which has information on the testing time and the thermal conductivity.
First Test: Potato Test 1
The Potato Test was the first test run on TCoMS to determine if TCoMS would provide a valid value for the thermal conductivity measurements. As potatoes are not a perfect substance and may have material properties vary, a value between 0.545 and 0.957 [W/mK] for thermal conductivity measurements would be acceptable. This test resulted in an average thermal conductivity value of 0.0063 [W/mK], only about 1% of the expected value.
​
TCoMS failed the accuracy test for the potato measurements at this time. The largest expected source of error is the onboard power source not supplying enough current. Changes were made to fix this issue before the next test.
Potato Testing Video
Subsequent Testing
Following the first test, revisions were made to TCoMS increasing the heating circuit voltage to 9 [V]. The test attempted was a cold water test using a 99% water 1% gelatin solution. During testing it was found that the 9 [V] power supply caused damage to the Arduino Due on which TCoMS relies. Trial tests just prior to the water test were reporting low thermal conductivity values so it was surmised that there still was not enough power being delivered. The data obtained in this test resulted in zero values for thermal conductivity.
​
The Arduino Due was replaced and a diode was added to the heating circuit to prevent damage to the Arduino. The heating circuit voltage was increased to 18 [V] To check for possible coding errors a temperature readout was also obtained from TCoMS and saved to an excel file. Tests were performed on water and a potato prior to another issue cropping up.
Final Testing Images
This image shows a hole being made in a potato using the poker from TCoMS.
The hole was made to allow for easy insertion of the testing probe.
TCoMS has been initialized in this image and awaits test start command for the water test.
This image shows a hole being made in a potato using the poker from TCoMS.
Results
At the end of this session of testing, the thermistor burnt out, there are many possible causes. The most likely cause is the thermistor was not secured well enough on the end of the probe causing its connections to fatigue. The temperature plots are shown below.
Using the temperatures, the thermal conductivity values of the various materials were then calculated and plotted through Excel as well. The results for these values are shown below. The expected values for the thermal conductivity of these materials are as follows:
​
-Water: 0.6 [W/mK]
-Potato: 0.545 to 0.957 [W/mK]
​
The potato thermal conductivity values have a large range of possible real world values. This is due to potatoes not being a consistent testing material. As such, the value compared to is near the average value of the two extremes, 0.75 [W/mK]. Realistically, results should be based off of the water test.
​
The equation that is used to determine the thermal conductivity from the temperatures is shown below. This equation is for the pulse decay method which TCoMS uses.
Downloadable Result File
To find an average thermal conductivity value, the averages for each test were used and then average for each material. The data used for the averages start two seconds after the heating cycle has finished, which is four seconds on the test timer. This is to avoid the high error section of the initial values. This type of initial error is characteristic of a pulse decay test.
​
The averaged values and their associated %error are as follows:
​
-Water
Thermal Conductivity = 1.26 [W/mK]
%error = 110%
​
-Potato
Thermal Conductivity = 0.75 [W/mK]
%error = 0.46%
​
​
Below is the results of the final requirement checklist. The final %error aspect of TCoMS did not meet the 5% error on the requirements for the water test. As the water test is a better analysis of the true accuracy this indicates that despite the potato test succeeding, TCoMS failed to meet the requirement.
