Procedures

A procedure is a single measurement: it declares parameters and data columns, connects instruments, and runs a startup → execute → shutdown lifecycle, emitting results as it goes. Procedures are ordinary PyMeasure Procedure subclasses with a Laser Setup base class added on top.

This page explains the class hierarchy and catalogs the built-in procedures. To write your own, see Creating a New Procedure.

The class hierarchy

classDiagram
    Procedure <|-- BaseProcedure
    BaseProcedure <|-- ChipProcedure
    BaseProcedure <|-- FakeProcedure
    BaseProcedure <|-- Wait
    ChipProcedure <|-- IVg
    ChipProcedure <|-- It
    ChipProcedure <|-- IV
    ChipProcedure <|-- Others["… all device measurements"]
    class Procedure {
      +startup() execute() shutdown()
      +emit(topic, value)
      +should_stop()
    }
    class BaseProcedure {
      +instruments: InstrumentManager
      +procedure_version, show_more, info
      +skip_startup / skip_shutdown
      +connect_instruments()
      +patch_parameters()
    }
    class ChipProcedure {
      +chip_group, chip_number, sample
      +Telegram alert on finish
    }

Procedure (PyMeasure)

Provides the engine: the lifecycle hooks, emit() for streaming results/progress, should_stop() for cooperative aborts, and the Parameter/Metadata/ DATA_COLUMNS machinery. Read the PyMeasure docs for the full API.

BaseProcedure

The root of all Laser Setup procedures (procedures/BaseProcedure.py). It:

• adds common parameters: procedure_version, show_more, info, skip_startup, skip_shutdown, and a start_time metadata;
• owns an instruments = InstrumentManager();
• implements connect_instruments(), default startup() (connect) and shutdown() (shut down all instruments);
• wraps startup/shutdown so they're skipped when skip_startup/ skip_shutdown are true (vital for sequences);
• is @configurable('procedures'), so YAML overrides apply automatically.

Key class attributes you'll set in subclasses:

Attribute	Meaning
name	Human-readable name shown in the GUI.
INPUTS	List of parameter names to show as inputs (order matters).
DATA_COLUMNS	Column names; the first two are the default plot axes.
EXCLUDE	Parameters to leave out of the saved file header.
SEQUENCER_INPUTS	Parameters a sequence may sweep.

ChipProcedure

Base for all device measurements (procedures/ChipProcedure.py). Adds chip_group, chip_number and sample inputs (so every file records which device was measured) and sends a Telegram alert when a run finishes. It also defines two mixins used across procedures:

• LaserMixin — zeroes the laser voltage when laser_toggle is off.
• VgMixin — evaluates the dynamic gate-voltage expression (DP + 15 V), substituting the sample's latest Dirac point.

Both hook into patch_parameters(), which runs just before a measurement.

Built-in procedure catalog

Procedure	name	Measures	Notable instruments
IVg	I vs Vg	Current vs. gate voltage (transfer curve); estimates the Dirac point	Keithley, TENMA ×2
IV	I vs V	Current vs. drain–source voltage sweep	Keithley, TENMA
VVg	V vs Vg	Voltage vs. gate voltage (constant current)	Keithley, Bentham
It	I vs t	Current vs. time, 3-phase laser off/on/off	Keithley, TENMA ×3, PT100, Clicker
Vt	V vs t	Voltage vs. time (constant current), 3-phase laser	Keithley, TENMA, PT100
ItVg	I vs t @ Vg steps	Time-resolved current at several gate voltages	Keithley, TENMA
ItWl	I vs t @ wavelengths	Time-resolved current vs. light wavelength	Keithley, Bentham
Tt	T vs t	Plate-temperature sweep over time	Clicker, PT100
Pt	P vs t	Optical power vs. time, 3-phase laser	Thorlabs PM100
Pwl	P vs wavelength	Optical power vs. wavelength	Bentham, Thorlabs
LaserCalibration	Laser calibration	Optical power vs. laser-driver voltage	TENMA, Thorlabs
Stress	Stress & relax	High bias/temperature, then monitored relaxation	Keithley, TENMA, Clicker
Wait	Wait	Does nothing for a set time (sequence spacer)	none
FakeProcedure	Fake Procedure	Synthetic data for demos/tests	none

Backward-compatible aliases

procedures/__init__.py defines IVT, ITt, IVgT as aliases of IV, It, IVg for older configs/scripts.

A parameter-level breakdown of each procedure is in the Procedure catalog reference.

The lifecycle in detail

flowchart TD
    Init["__init__<br/>override_parameters · wrap skip flags"] --> Patch["patch_parameters()<br/>(mixins transform inputs)"]
    Patch --> SU["startup()<br/>connect_instruments() + configure hardware"]
    SU --> EX["execute()<br/>measurement loop:<br/>emit('results') / emit('progress')<br/>check should_stop()"]
    EX --> SD["shutdown()<br/>instruments.shutdown_all()"]

• startup() configures hardware (reset meter, set ranges, enable sources).
• execute() is the measurement loop. It must check self.should_stop() regularly so Abort works, and call self.emit('results', dict(...)) with keys matching DATA_COLUMNS.
• shutdown() safely powers things down. With BaseProcedure, the default shuts down every connected instrument.

Worked examples (FakeProcedure and It) are in Creating a New Procedure.

Parameters come from config

Most procedures don't hard-code parameter definitions — they pull them from the shared parameters.yaml via laser_setup.procedures.utils.Parameters:

from .utils import Parameters
class It(...):
    vds = Parameters.Control.vds          # a FloatParameter defined in YAML
    laser_v = Parameters.Laser.laser_v

This keeps units, limits and descriptions consistent across procedures. See Defining Parameters.