pyNGB - NETZSCH STA File Parser

A high-performance Python library for parsing NETZSCH STA (Simultaneous Thermal Analysis) NGB files with comprehensive metadata extraction and analysis tools.

⚠️ Disclaimer: This package is not affiliated with NETZSCH-Gerätebau GmbH. NETZSCH is a trademark of NETZSCH-Gerätebau GmbH.

Features

• Strict Binary Parsing: a record-grammar tokenizer parses every byte of the file — no pattern-match guessing (tens of milliseconds per file)
• Complete Metadata: Extract instrument settings, sample info, and experimental conditions
• Baseline Correction: Automatic baseline subtraction with validation, including the correction run embedded in Sample + Correction (.ngb-ds3) files
• DTG Analysis: Derivative thermogravimetry calculation with smoothing options
• Batch Processing: Parallel processing of multiple files
• Data Export: Convert data to Parquet or CSV; export metadata to JSON
• CLI Support: Command-line interface for automation

Installation

pip install pyngb

Quick Example

from pyngb import read_ngb
import polars as pl
import json

# Load NGB file with embedded metadata
table = read_ngb("experiment.ngb-ss3")
df = pl.from_arrow(table)

# Access data
print(f"Loaded {df.height} data points")
print(f"Temperature range: {df['sample_temperature'].min():.1f} to {df['sample_temperature'].max():.1f} °C")

# Access metadata
metadata = json.loads(table.schema.metadata[b'file_metadata'])
print(f"Sample: {metadata.get('sample_name', 'Unknown')}")

# Baseline correction
corrected = read_ngb("sample.ngb-ss3", baseline_file="baseline.ngb-bs3")

# Sample + Correction files (.ngb-ds3): raw sample run by default, the
# embedded correction run, or the corrected curves Proteus displays
raw = read_ngb("run.ngb-ds3")
corr = read_ngb("run.ngb-ds3", run="correction")
corrected = read_ngb("run.ngb-ds3", run="corrected")

# DTG analysis
from pyngb.api.analysis import add_dtg
table_with_dtg = add_dtg(table, method="savgol", smooth="medium")

Command Line Usage

The CLI is available as the pyngb console script (or equivalently python -m pyngb):

# Convert single file
pyngb convert sample.ngb-ss3

# Batch processing with baseline correction
pyngb convert *.ngb-ss3 -b baseline.ngb-bs3 -f parquet -o ./processed/

# Inspect file structure / run data-quality checks
pyngb inspect sample.ngb-ss3
pyngb validate sample.ngb-ss3

# Get help
pyngb --help

Documentation

• Getting Started - Installation and first steps
• User Guide - Core functionality and examples
• API Reference - Complete function documentation
• Troubleshooting - Common issues and solutions
• Contributing - Development guidelines

License

MIT License. See LICENSE for details.

Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions

Made with ❤️ for the scientific community