Files
Abstract
The purpose of this report is to describe, explain and discuss the results associated with the author’s Milwaukee School of Engineering [MSOE] Master of Science in Engineering [MSE] capstone project. The purpose of the capstone project is to design a Time-of-Flight [TOF] sensor-based liquid level indicator/sensor [LLI or LLS] to measure fire suppression agents in pressurized storage cylinders associated with clean agent fire suppression systems. The conventional clean agents today are FM-200 and 3M Novec 1230, and in compliance with standard practice, the amount of agent in a storage cylinder must be regularly measured to ensure that the fire suppression system can provide adequate protection in the event of a fire. The current technology employed to measure fire suppression agents in storage cylinders features a Dip Tape design that is associated with inaccuracies because it is cumbersome to use and prone to human-operator errors. A review of literature was first undertaken for the following purposes: [i] to verify the relevant standards in clean agent fire suppression systems, [ii] to clarify the types of sensors available for the measurement of liquid levels, and [iii] to review relevant patent literature to determine the state-of-the-art in liquid level measurement, particularly with respect to fire suppression systems. A new design—which leverages the temperature and weight of the liquid agent to calculate the volume and mass of agent in a storage cylinder—was developed. The new design features a Time-of-Flight [TOF] sensor, a Texas Instruments LM35 temperature sensor, a Human Machine Interface [HMI], non-volatile system memory to store settings and data, and an ATMEGA328P microcontroller. System hardware and software programming was performed with Arduino. A POC system was successfully developed in five phases. In Phase One, components were selected, and a prototype unit was assembled. In Phase Two, the Arduino microcontroller firmware was developed, enabling the system to display the TOF sensor data and temperature data. Phase Three entailed the development of an algorithm for converting the TOF and temperature data into a volume and weight. Phase Four saw the development of additional firmware features [i.e., a pushbutton], and Phase Five was devoted to the challenging development of the user interface. The report features detailed explanations of each of these phases, including development and strategies associated with firmware and other source code. Following the development phases, accuracy and repeatability testing and verification of sensor data, and verification of the firmware were undertaken. In place of FM-200 and 3M Novec 1230, water—which has similar chemical characteristics—was employed. The project demonstrates that the new TOF system design can accurately determine the agent weight and is therefore a promising LLI alternative technology for clean agent fire suppression systems. However, testing and verification revealed an inconsistency in sensor data associated with temperature, along with the realization that the LM35 temperature sensor needs to be replaced with a more accurate sensor. It was additionally determined that the current hardware memory size is likely inadequate to support significant data recording. Firmware testing, moreover, revealed two software bugs, one associated with serial communications and one associated with data recording. Further testing and development associated with these issues—along with the need to verify the system with FM-200 and 3M Novec 1230—is recommended.