This code and setup were published as supporting information for A modified method for measuring pressure drop in non-medical face masks with automated data acquisition and analysis by Herman et al. in the Journal of the International Society of Respiratory Protection (ISRP Membership Required to View) in December of 2021.
An open access version is also published in PubMed Central under the following ID: PMC9126268
- This repository's contents are under a CC0 v1.0 Universal License. However, this license does not govern the software needed to run or open the files in this repository and may be governed under a separate license.
- The use of the files in this repository, the code presented in it, its dependent functions, or the software required to run it does not constitute an endorsement from the U.S. Food and Drug Administration or U.S. Department of Health and Human Services.
The code and setup outlined here in this repository can allow anyone with a 3D printer, a Raspberry Pi with Python 3, a computer that is able to run MATLAB to do the following:
- Print the main body of the Pressure Plate
- Record and acquire data from pressure transducers utilizing an Analog to Digital Converter (ADC) and a Raspberry Pi via Python code
- Analyze and summarize the acquired data using MATLAB scripts.
This repository can be broken down into 3 main sections:
- Pressure Plate STL
- Data Acquisition Python Script
- Data Processing MATLAB Script and Dependent Functions
Additional requirements, such as hardware, are explained in each of the individual main sections.
An STL file of pressure drop rig that we used in the manuscript. Units are in millimeters. Two prints of this STL are required for the full pressure rig. One half can optionally be mirrored on the ring side to have all of the ports on the same side. The gasket shape can be created by converting to a vector file of the ring side. The two halves and the gaskets, along with the sample under test, can be held together with 4x M3 x 0.5mm, 60mm long screws.
The main inlet and outlet ports are designed for 1/8" NPT fittings that can be Room Temperature Vulcanized (RTV) or epoxied in.
The pressure ports designed for 1/4"-28 thread fittings that can be RTV or epoxied in.
This python script (presslog.py) is a CLI based script that requires:
- A Raspberry Pi
- An ADS1115 16-bit ADC and its required libraries (See Below)
- A pressure transducer module with a linear output voltage of 0 to 5 Volts (and an impedance which is compatible with the ADS1115)
Obtain help in running the script by typing:
python3 presslog.py -h
Run the script by typing:
python3 presslog.py
How to install CircuitPython Libraries from Adafruit
Python Command Line Arguments Help
The output of the presslog.py script will output a TXT file that can be analyzed by the MATLAB script pAnalysis.m1. pAnalysis.m can handle multiple TXTs output from presslog.py. All TXTs, pAnalysis.m and all of the required files are in the pAnalysis directory. The scripts and functions required for processing are as follows:
pAnalysis.m- Script for Importing and Processing the TXTs outputted frompresslog.pypDataImport.m- Function that Imports a specified TXT to the MATLAB workspacenewMaterial.m- Lookup table helper function to standardize material namespSpike.m- Function to detect a sudden spike in pressurepCell2CSV.m- Function to output a MATLAB cell contents to a CSV file
PAnalysis.m is a script that import pressure logging files outputted from our pressure transducer setup. This Script imports a pressure transducer log file and calculates the average pressure after the index the script finds when the flow controller was turned on plus an offset. This is repeated for every log file in the same directory and a report can be generated.
This script is designed to read pressure data from two differential pressure transducers of different sensitivities. It averages the pressures from both transducers and only selects the reading from the transducer that is the most sensitive where it does not clip. This code can be used for any pressure unit specified (Quantity of Interest [QOI]). This script can also be used with one transducer
Requires: MATLAB R2006a (Tested with R2020b & R2014b)
NOTE: The functions used in this code are compatible with versions R2006a or before. However, some input parameters and Name/Value pairs might not be compatible. Make sure function inputs used here and with supplied functions are compatible with the version of MATLAB being used to run this script.
IMPORTANT: For the grouping of materials to happen properly, the nomenclature of the pressure files name must be the following format:
!Press#<material>#yyyy_mm_dd_HH-MM-SS.txt
This format is based on the output of presslog.py and the script is designed around it.
!Press at the start of the file name identifies if this a pressure logging file. Any files that do not have this tag and not have a file extension of TXT are not loaded for analysis.
<material> is the user entered material from the presslog.py and must be separated pound signs (#) at the start and end of the material. There is a validation of the material in the file name and the material name in the file. If there is a discrepancy, a warning is displayed.
After the material is a time stamp of when then the analysis started. It must be in the format listed above as there is a validation of the time stamp in the name and the time stamp in the file. If there is a discrepancy, a warning is displayed.
NOTE: Only one Flow Rate Can Summarized at a time. Code assumes all Pressure Logging files are cataloged under the same flow rate. This only affects the output of p_report (Variable in pAnalysis.m) and the saved CSV
pDataImport.m is a function that imports data from a text file outputted from our pressure transducer logger program. It reads data from text file name (filename) passed to it and output the data from the file in a structure. This function is based on the text file output from presslog.py and reads all of the different data fields from that text file.
| Input Parameter | Description |
|---|---|
filename |
char array of a TXT file name to import |
newMaterial.m is a function that compares the material name (oldname) passed to the function against a lookup table (oldnewnames) and either returned the standardized name from the table in the form of a Nx2 cell array with character vectors or returns the old name based off an input flag (std_name).
Lookup Table Format (oldnewnames)
| Column 1 | Column 2 |
|---|---|
| Old Name | New Name to Replace Old Name in Corresponding Row |
A standardized name will be returned if std_name is true and there is a match of the oldname in column 1 in oldnewnames. The standardized name will be in column 2 of the same row where the oldname is found in column 1 of oldnewnames. If no match if found or std_name flag is false, then the oldname is returned.
| Input Parameter | Description |
|---|---|
oldname |
Material Name to compare to the lookup table (oldnewnames) |
std_name |
Flag to return a Standard Name or Return the old one |
pSpike.m is a function that detects a sudden change in a 1D vector signal (Pressure) and outputs its index. This is done by picking a random point along the vector finding the cumulative variances on both side of the point, multiplying it by the index of the random point from each end, adding them together. The point with the smallest sum represents the point where the variances on both sides are least fluctuating and thus results in the most significant point where the signal changes.
| Input Parameter | Description |
|---|---|
P_data |
Pressure data vector |
-
"Find abrupt changes in signal - MATLAB findchangepts", Mathworks.com, 2016. [Online]. Available: https://www.mathworks.com/help/signal/ref/findchangepts.html#d123e60404. [Accessed: 04-Jun-2021]
-
P. P. Pebay, "Formulas for robust, one-pass parallel computation of covariances and arbitrary-order statistical moments,"; Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA (United States), SAND2008-6212; TRN: US201201%%57 United States 10.2172/1028931 TRN: US201201%%57 SNL English, 2008. [Online]. Available: https://www.osti.gov/servlets/purl/1028931
-
T. F. Chan, G. H. Golub, and R. J. LeVeque, "Algorithms for Computing the Sample Variance: Analysis and Recommendations," The American Statistician, vol. 37, no. 3, pp. 242-247, 1983, doi:10.2307/2683386.
-
R. Killick, P. Fearnhead, and I. A. Eckley, "Optimal Detection of Changepoints With a Linear Computational Cost," Journal of the American Statistical Association, vol. 107, no. 500, pp. 1590-1598, 2012. [Online]. Available: http://www.jstor.org/stable/23427357.
-
J. Muñoz and C. Luengo, "Which function allow me to calculate cumulative variance over a vector?", Stack Overflow, 2019. [Online]. Available: https://stackoverflow.com/questions/58343348/. [Accessed: 04-Jun-2021].
pCell2CSV.m is a function that saves a cell array to a CSV file. This function uses fprintf to output a CSV file. This was used to over the newer writecell for compatibility with older versions of MATLAB.
| Input Parameter | Description |
|---|---|
pCell_in |
Cell Array to be saved as a CSV File |
FN |
Character array of the file name of the CSV to be outputted less the file extension |
num_per |
Precision of the number to output in the output table |
Footnotes
-
Assuming the procedure outlined in the manuscript is followed. ↩