# Pm100d Device

```{toctree}
:hidden:
:maxdepth: 2
```

## Overview

A plesty device for pm100d built on the Plesty standard.

## Installation

```bash
git clone https://gitlab.com/plesty/hub/devices/plesty-pm100d.git
cd plesty-pm100d
uv sync
```


## Running as a TCP server

Make sure the device is connected and powered on, then:

```bash
uv run python -m plesty.pm100d --tcp-port 5555 -a "USB0::0x1313::0x8078::<serial>::INSTR"
```

Without hardware, `--mock` serves an in-memory SCPI simulator with the same
command contract as the real instrument.

## Troubleshooting

### Driver configuration

In Device Manager the meter must appear as **USB Test and Measurement Device
(IVI)** — confirm it is the PM100D under Properties → Details → Hardware Ids
(`VID_1313&PID_8078`). If the Thorlabs OPM/TLPM driver owns it, VISA reads
time out (`VI_ERROR_TMO`) even though the session opens; run the **Thorlabs
Power Meter Driver Switcher** (installed with the Thorlabs power-meter
software) to switch the meter back to the USBTMC/VISA driver, or rebind via
*Update driver*. Close the Thorlabs Optical Power Monitor software
completely — it lingers in the tray and holds the instrument exclusively. The
NI-MAX test panel is the arbiter: if `*IDN?` times out there too, the problem
is driver/instrument-side, not this module.

### Finding the VISA address

```bash
uv run python -c "import pyvisa; print(pyvisa.ResourceManager().list_resources('USB?*::INSTR'))"
```

Pass the full resource string including the serial number
(`USB0::0x1313::0x8078::P0001234::INSTR`) as `--address`.

### `list_resources()` / server startup hangs and Ctrl+C is dead

The process is stuck inside the VISA library's native code, so the interpreter
never gets to deliver `KeyboardInterrupt` — kill it from another terminal or
Task Manager. Then work through, in order:

1. **Wedged meter** — a halted USBTMC endpoint can stall the whole VISA
   enumeration. Power the meter fully off (hold the power button), wait ~10 s,
   replug USB (ideally another port), retest.
2. **Another wedged USBTMC instrument** — enumeration touches every device in
   the *USB Test and Measurement Devices* category; any one of them can stall
   the scan. Unplug or disable them one by one to find the culprit.
3. **Stuck NI-VISA services** — if the hang persists with no USBTMC devices
   attached, **reboot the PC** (resets the NI services and the USB host
   controller; this is the usual fix).
4. **Damaged VISA install** — if NI MAX also freezes expanding *Devices and
   Interfaces* after a reboot, repair NI-VISA via NI Package Manager; if a
   second vendor VISA (Keysight, R&S) is installed, set NI-VISA as primary in
   the VISA conflict manager.

### USB connection issues

- **Every command times out** although the session opened: power-cycle the
  meter and replug the USB cable — USBTMC devices wedge into a non-answering
  state that survives server restarts. (An open session proves nothing:
  `check_operatability()` only checks that the session object exists.)
- **Only the first command after (re)connect fails**: a previously timed-out
  query can leave an unread response in the USBTMC buffer. The failure clears
  itself after one successful command — retry before digging deeper.

## Reference

The **Parameters**, **Functions** and **Standard Methods** pages in the
navigation are generated automatically from the device schema by
`plesty docs` — you do not edit them by hand. Keep this page focused on the
QuickStart above.