From 8a80beb17505ee7ba55fd1f4b2c9b5e30e96f035 Mon Sep 17 00:00:00 2001 From: GraysonBellamy Date: Thu, 20 Jun 2024 10:28:20 -0400 Subject: [PATCH] Adds docs page for cone and hfm --- docs/cone.md | 17 +++++++++++++++++ docs/hfm.md | 9 +++++++++ mkdocs.yml | 2 ++ 3 files changed, 28 insertions(+) create mode 100644 docs/cone.md create mode 100644 docs/hfm.md diff --git a/docs/cone.md b/docs/cone.md new file mode 100644 index 0000000..a6e96b6 --- /dev/null +++ b/docs/cone.md @@ -0,0 +1,17 @@ +# Cone File Schema + +This documentation provides an overview of the schema used in cone calorimeter files. There exists a variety of manufacturers but the one we have and are most familiar with is made by Deatak. It is know that other manufacturers output file formats differ and will be the subject of future work on this project. + +## File Structure + +The output of the cone calorimeter programs is and excel ('.XLSM') file with two sheets. The first sheet contains the metadata for the test and is denoted 'Scalar Data'. The second sheet contains the actual test data in a typical tabular format with the name 'Scan Data'. + +### Header + +The metadata sheet contains information about the test, such as the laboratory, sample mass, operator, etc. This is the information which is contained in the file-wide metadata associated with the output Parquet files. + +### Data + +The data sheet contains the tabular data format. Column names are given on the first row and the next three rows give information about the acquisition parameters. These are dropped from the output as they do not contain useful information. The 4th row contains the units for their respective columns. The baseline row gives information about the state of the values before the test begins and is extracted and given a null time in the output Parquet file. The remainder of the rows contains the actual data and is extracted and stored as-is. Most of the data types are floats except for the 'Start Test' and 'Flame Verif.' indicators which contain a simple binary digit. + +The 'Ext Coeff' column is dropped from the data as it is a derived quantity from the 'Smoke Comp' and 'Smoke Meas' columns. diff --git a/docs/hfm.md b/docs/hfm.md new file mode 100644 index 0000000..3d7b2d0 --- /dev/null +++ b/docs/hfm.md @@ -0,0 +1,9 @@ +# HFM File Schema + +This documentation provides an overview of the schema used in heat flow meter files. There exists a variety of manufacturers but the one we have and are most familiar with is made by Waters/TA. + +## File Structure + +The output of the heat flow meter programs is and text ('.txt') file. This is possibly the nastiest file we have to deal with and resembles a word document rather than any standardized instrument output. The is a large section of metadata at the top and then also interspersed throughout. It is also questionable whether this should actually be stored in a Parquet file because it is literally like a 5x3 table but we are doing it to maintain consistency. We could move it to a JSON in the future if we wish. + +The way that we extract data and metadata from these files right down is extremely manually. We pretty much have to check every line and the text in it to determine what it is. I've developed better solution in the past for a small subset of data with complex regex but it is a pain to read and edit and hard to expand. As such, this is the solution at the moment but I would love a better one. diff --git a/mkdocs.yml b/mkdocs.yml index fd58cdc..015b61f 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -43,6 +43,8 @@ nav: - STA: sta.md - MCC: mcc.md - FTIR: ftir.md + - HFM: hfm.md + - Cone: cone.md markdown_extensions: - pymdownx.highlight: