Usage

Overview

PyRth offers several evaluation methods to process thermal transient data. The main functions are:

  • standard_module() Provides a basic evaluation for processing input data to compute thermal impedance and structure functions.

  • standard_module_set() Enables batch processing over multiple parameter configurations. This method iterates over provided iterable keywords and generates a set of evaluation modules.

  • bootstrap_module() Applies statistical bootstrapping to estimate variability and compute confidence intervals for the calculated thermal properties.

  • optimization_module() Optimizes model parameters, refining the fit between the theoretical model and the measured data.

  • theoretical_module() Generates theoretical predictions based on given resistances and capacitances, computing the theoretical thermal impedance.

  • comparison_module() Compares evaluation results, e.g. between theoretical predictions and experimental data or between different evaluation strategies.

  • temperature_prediction_module() Predicts temperature profiles from measured power data using computed impulse responses and convolution.

Data Sources

The following examples use measurement data provided under tests/data in the repository, loaded via the helper module tests/data/measurement_data.py:

  • MOSFET_DRY_DATA and MOSFET_TIM_DATA: Voltage transients from MOSFET devices, with and without thermal interface material.

  • LED_DATA: Voltage transients from LED devices for thermo-optical calibration testing.

  • TEMP_DATA: Time and temperature differences extracted from ASCII data files (.asc) for thermal response.

  • Calibration arrays: MOSFET_CALIB_DATA and LED_CALIB_DATA, used to convert voltage or resistance readings into temperature.

All data are returned as NumPy arrays of shape (n, 2), with columns [time, measurement]. In your own workflows, replace these with your own arrays (e.g., from CSV, T3ster exports, or other sources) before passing to the evaluation methods.

Input Modes

PyRth supports four input modes to describe how raw measurement data is interpreted. Set the input_mode key in your parameter dictionary to one of the following:

  • impedance - Data columns: [time, Zth] (thermal impedance vs. time). - No conversion applied; data are assumed to be impedance already. - Use when you have precomputed impedance values.

  • temperature - Data columns: [time, temperature_rise]. - Converts ΔT to Zth = ΔT / P, where power_step (W) must be set. - Use when your measurement yields temperature vs. time.

  • volt - Data columns: [time, voltage]. - Converts voltage to temperature via a calibration array (calib) and then to Zth. - Requires: calib array (e.g., MOSFET_CALIB_DATA), kfac_fit_deg to set polynomial fit degree. - Use when measuring a voltage-based temperature-sensitive parameter (TSP) (e.g., a MOSFET).

  • t3ster - Specialized mode for Siemens T3Ster raw data files. - Parses infile (.raw), infile_pwr (.pwr), and infile_tco (.tco) to extract voltage and calibration. - Use when working directly with T3Ster instrument exports.