xif-fr / experiment-monitoring

A software package for automated monitoring of lab equipment, including time series visualization and automatic e-mail alerts.

Experiment Monitoring Software

A software package for automated monitoring of lab equipment, including time series visualization and automatic e-mail alerts.

Architecture

A central server gathers data from different sources, writes them into a database, hosts a graphic interface for visualization and sends automatic alert e-mails based on user-defined criteria.

Currently supported interfaces

• Serial (e.g. Pfeiffer TPG261)
• TCP/IP (e.g. Arduino Due)
• Analog [via ADC on Arduino Due] (e.g. Pfeiffer TPG300)
• SNMP (e.g. Eaton UPS)
• Phidgets (e.g. Thermocouple module)

Setup

• Hardware requirements & server setup: See server_setup.md.
• ADC setup: See adc_setup.md. Only needed if you want to monitor analog signals.
• Working with existing interfaces:
  • config.py is all you need to modify.
• Adding your own interfaces:
  • Write a subclass that extends the abstract Sensor class defined in sensor.py to drive your sensor/equipment and instantiate it in config.py.

Guide to the repository structure:

• calibrations: Contains calibration data and scripts for all calibrated equipment.
• classes: Contains driver classes for all interfaces. Put your new driver classes in here.
  • sensor.py: Abstract base class for individual sensor classes.
  • adc: Contains Arduino sketch, its Python Class and its setup guide.
  • ups: Implements EatonUPS Class for batteries and their setup guide.
• tests: Contains tests for the Phidget class.
• utilities: Contains interface-independent classes to be used by all sensors.
  • spike_filter.py: Spike filter for instances of Sensor subclasses. Enable in config.py by setting sensor.spike_filter.spike_threshold_perc for any given sensor.
• config.py: Main configuration file.
• exec.py: Main execution file for Linux service and command line execution.