bapsf_gaspuff

Monitoring of the gas-puff and pressure system of LAPD at the Basic Plasma Science Facility (BAPSF). The repository covers two largely independent subsystems:

1. Pfeiffer Vacuum gauge logging — continuous pressure acquisition and real-time monitoring (see below).
2. Gas-puff acquisition — a Raspberry Pi pipeline tying together a Sensirion flow meter, a piezo valve driven through an AD/DA board, and a pressure transducer.

---

Pfeiffer Vacuum Gauge

Everything for the Pfeiffer subsystem lives in pfeiffer/. It talks to a Pfeiffer Vacuum MaxiGauge TPG 366 controller over Ethernet, continuously logs every connected sensor's pressure to daily HDF5 files, and provides a live plotting GUI.

Hardware & connection

• Controller: Pfeiffer MaxiGauge TPG 366 (up to 6 pressure sensors).
• Transport: TCP socket over Ethernet — default controller IP 192.168.7.44, port 8000.
• Protocol: Pfeiffer serial/ASCII mnemonic protocol (PRX, TID, GAS, etc.) with ACK/NAK handshaking, tunneled over the socket. Line termination is CR+LF.
• Units: Pressure logged in Torr.

Files

File	Purpose
pfeiffer/PfeifferVacuumCommunication.py	Low-level driver. Defines the MaxiGauge class (socket connect/disconnect, mnemonic send/enquire, ACK/NAK handling) and the MaxiGaugeError / MaxiGaugeNAK exceptions.
pfeiffer/Pfeiffer_control.py	Acquisition loop. Continuously polls all sensors and appends readings to a daily HDF5 file.
pfeiffer/Pfeiffer_GUI.py	PyQt5 real-time plotting GUI that reads the latest HDF5 file.

MaxiGauge driver (key methods)

• connect() / disconnect() — open/close the TCP socket (2 s timeout, up to 300 retries).
• get_all_pressure_reading() — returns (status_list, pressure_list) for all sensors via the PRX mnemonic, with retry on garbled responses.
• get_device_id() — sensor model IDs (TID).
• get_gas_type() — gas calibration setting per sensor (GAS), mapped through the GAS_TYPE table (Nitrogen, Argon, Hydrogen, Helium, Neon, Krypton, Xenon, CAL).
• pressure(sensor) — single-sensor reading (PR<n>), with status decoded via PRESSURE_READING_STATUS (e.g. Underrange, Overrange, Sensor error/off, No sensor).

Acquisition: Pfeiffer_control.py

Run this to start logging:

cd pfeiffer
python Pfeiffer_control.py

What it does:

• Writes to C:\data\gauge\pressure_data_<YYYY-MM-DD>.hdf5, rolling over to a new file automatically at the day boundary.
• HDF5 layout: a PfeifferVacuum group containing one resizable dataset per sensor ("1", "2", …) plus a timestamp dataset (seconds since epoch). Each sensor dataset carries Model, Gas, Unit, and Modified time attributes; when a gauge or gas setting changes, the new value is appended to the relevant attribute list rather than overwritten.
• Uses HDF5 SWMR (Single-Writer/Multiple-Reader) mode so the GUI can read while the writer is running.
• Resilience: logs connection STARTED / LOST / RECOVERED events to C:\data\gauge\connection_log.txt, transparently reconnects on MaxiGaugeError/TimeoutError, and auto-recovers from stale SWMR locks by invoking h5clear (tries h5clear on PATH, then a vendored fallback path).

Configuration: the controller IP, HDF5 output directory, and h5clear fallback path are set near the top of pfeiffer/Pfeiffer_control.py. Adjust them for your machine.

Live monitoring: Pfeiffer_GUI.py

cd pfeiffer
python Pfeiffer_GUI.py

A PyQt5 window that opens the most recent HDF5 file in C:\data\gauge (SWMR read) and refreshes every 0.5 s in a background thread. It shows two stacked panels:

• Top: the last 30 seconds of pressure.
• Bottom: the full day binned into 5-minute averages.

The sensor to plot (sensor_number) and the rolling window length (n_points) are configurable at the top of the file.

Dependencies (Pfeiffer)

h5py, numpy, portalocker, PyQt5, matplotlib, and the HDF5 h5clear command-line tool (bundled with the HDF Group HDF5 distribution).

---

Gas-Puff Acquisition (overview)

The gas-puff side runs on a Raspberry Pi and coordinates the valve, flow meter, and pressure transducer. The data flow:

flowchart LR
    subgraph gas[gas puff]
    direction LR
    ADC[AD/DA board] -->|DAC out| amp[Amplifier]
    amp --> piezo[Piezo valve]
    amp --> |optional|scope
    psi[Pressure transducer] -->|ADC in| ADC
    end

    Pi === ADC

    subgraph central[" "]
    in["input.txt"]-->Pi[Raspberry Pi]
    Pi-->out["output.csv"]
    flow[Flowmeter] -->|serial| Pi
    Trigger -->|GPIO pin| Pi
    LabTimeServer --> Pi
    flow .->|data| out
    end

    in .->|data| out
    psi .->|data| out

Loading

• Flow meter: Sensirion SFC5xxx, accessed over the SHDLC serial protocol. Driver docs: https://sensirion.github.io/python-shdlc-sfc5xxx/index.html
• Trigger: hardware shot trigger read on a Raspberry Pi GPIO pin (via a small C library, gpio_detect.so).
• Output: per-shot flow data saved to HDF5.

Arduino firmware for the trigger/status display lives in arduino_src/, and hardware datasheets (AD/DA board, amplifiers, Arduino) are in doc/.

---

Repository layout (src/)

The src/ directory holds the gas-puff Raspberry Pi code. It is a separate pipeline from the Pfeiffer logger above.

File	Role
src/flowmeter_main.py	Main acquisition entry point — GPIO trigger handling, multiprocessing-based flow capture, and per-shot HDF5 saving.
src/FlowMeterCommunication.py	FlowMeter wrapper around the Sensirion sensirion-shdlc-sfc5xxx driver.
src/wavegen_control.py	Driver for the waveform generator / AD/DA board used to drive the piezo valve.
src/kernel.py	High-level GasPuffValve interface coupling the valve, trigger, and waveform output.
src/input.py	Waveform definition (generate_pulse_waveform) and a small Tkinter control panel.
src/main.py	Standalone example that programs and bursts a pulse waveform.

Note: some wavegen-coupled paths are partially archived/commented; flowmeter_main.py is the current acquisition driver.

---

License

See LICENSE.