Date of Completion


Document Type

Honors College Thesis


Department of Civil and Environmental Engineering

Thesis Type

Honors College

First Advisor

Britt Holmen


This thesis centers around an analysis of two low-cost, small footprint non-dispersive infrared (NDIR) CO2 sensors for incorporation to lightweight drone-based sensing applications. The sensors of interest are produced by SenseAir and are the models K30 1% FR and K33 10,000ppm ELG. Low-cost portable sensors such as these are becoming increasingly common, but require thorough analysis of their accuracy, sensitivity to environmental noise, and response time in order to increase their reliability for accurate CO2 measurement applications. The sensors were tested against benchmark instruments in a flow through chamber. The goals of this project were to increase understanding of both sensors’ characteristics, especially sensitivity to ambient temperature and relative humidity conditions. It was found that the influence of humidity on the CO2 concentration produced contradicting trends within the two different humidity sensitivity tests conducted. It was also found that the relative humidity readings of the K33 were higher than the Hobo benchmark logger by 11 – 15 % throughout the tests. Additionally, the CO concentrations were offset from benchmark readings by a consistent amount in the tests conducted. This offset ranged from 70 - 90 ppm higher for the K30 and 80-100 ppm higher for the K33. This suggests that an algorithm could be applied to correct for this offset. In order to determine this correction procedure, the sensors’ should be tested following calibration and in greater quantities, meaning more sensors of the same model. For all tests conducted here, the manufacturer’s Automatic Baseline Correction (ABC) algorithm was not applied, leaving the opportunity to further analyze this calibration method.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.