LAB-Mangetometer

Code related to readout and analysis of data taken with a Bartington Magnetometer

Requirements

Required submodule daqhats need to be installed in the virtual environment. See DAQHATS.

Other required libraries for spectral analysis include:

1. dash for web-ui.
2. jupyter for notebooks.
3. polars for large csv reading.
4. h5py for hdf5 file handling.
5. scipy for spectrograms.
6. argparse for python argument parsing.

Scripts

CheatSheet

1. Start Screen session:

screen -S <session_name>

2. Environment activation:

cd gdzhao
source .env/bin/activate

3. Start Acquisition (each in individual screen sessions):

Monitor raspberry-pi health using python monitorpi.py -s <scan_rate> <save_dir>. A recommendation is -s 0.5 use default. Saves in save_dir/logs/ directory.

Scan magnetometer using python scan_save_rawh5_fault_tolerant.py -s <scan_rate> -t <t_measure> <save_dir>.

4. QuickStart for watchdog session. python watchacq.py -s <scan_rate> -t <t_measure> <save_dr directly starts an acquisition in script scan_save_rawh5_fault_tolerant.py in a separate screen session. This script will watch the running status of the acquisition script and attempt to restart the screen session and the script once error occurs.

Scan to binary files

• scan_save_rawh5.py saves one hdf5 file with continuous writing. Data is raw adc code without calibration in uint 16 format.
• scan_save_rawh5_fault_tolerant.py scan at given scanrate, by default dumps hdf5 file parts every 1 hour. Data is raw adc code in uint16 format.

Scan to save (old)

• continuous_scan_save.py saves with either hdf5 or csv. The acquisition length should not exceed the memory limit of the raspberry-pi.
• continuous_scan_saveh5.py saves with hdf5 only. The acquisition length should not exceed the memory limit of the raspberry-pi.
• continuous_scan_savecsv.py saves line by line the readout. Has low memory usage and can handle longer durations.

Raspberry-pi health logs

• monitorpi.py saves a csv of cpu temperature, external voltage, inbox temperature, inbox humidity, inbox pressure.

Spectral Analysis

• magnetofft.py uses fft to compute fft amplitude. New version of magnetofft.py contain Power Spectrum (PS) and Power Spectral Density (PSD) calculation from FFT_report. Also contains plotting routines for PS and PSD.

Batch Processing

• pspsdbatch.py computes and stores 1-hour ps-psd results, have options to save either each segment or the mean of each hour.