DAPHNE Interface CLI

A friendly command‑line wrapper (based on Typer) to configure, calibrate and debug DAPHNE front‑end boards from NP/LAr test‑stands.

$ daphne --help
Usage: daphne [OPTIONS] COMMAND [ARGS]...

---

Installation

git clone <your fork>
cd daphne_interface
python3 -m venv venv
source venv/bin/activate
pip install -e .

---

Shell completion

# Bash
eval "$(daphne --install-completion bash)"
# Z‑sh
eval "$(daphne --install-completion zsh)"
# Fish
eval "$(daphne --install-completion fish)"

---

Command cheat‑sheet

Area	Skeleton	Note
Configure	daphne configure json FILE	apply full JSON config
Spy buffer capture	daphne capture plotly [opts]	interactive Plotly HTML
Offset calibration	daphne calibrate offsets [opts]	adaptive self‑centring
Read‑backs	`daphne check bias	trim
Low‑level helpers	`daphne set bias	trim …`

---

Offset‑DAC example

daphne calibrate offsets --details details.json \
                         --target 4000 --band 2 \
                         --samples 4000 --n-wf 20 \
                         --step-init 50 --max-iters 20 \
                         --save-json best_offsets.json

Key points: dynamic step (min 1), median baseline on first 128 samples, safe DAC window 1500‑3000, log+PNG artefacts.

---

Spy‑buffer example

daphne capture plotly --details details.json \
                      --samples 4000 --n-wf 10 \
                      --trigger aligned --html wf.html

Working with DAPHNE spy-buffer data

These short recipes show how to capture raw wave-forms, save them to disk, and plot / store mean-FFT spectra for any DAPHNE endpoint.

---

1 · Quick one-liner from the CLI

Capture 10 wave-forms × 4000 samples from the channels listed in details.json, trigger the spy-buffer in aligned mode, build an interactive Plotly viewer and keep a compressed copy of the raw data.

daphne capture plotly \
       --details details.json      \
       --samples 4000              \
       --n-wf 10                   \
       --trigger aligned           \
       --html waveforms.html       \
       --save-wf waveforms.npz

2 · Programmatic capture inside Python

import numpy as np
from daphne_app.hardware.daphne import Daphne

ip_suffix = 7                                  # board 10.73.137.107
dev = Daphne(f"10.73.137.{100 + ip_suffix}")

wave = dev.read_waveform(afe=0, ch=0,
                         samples=4000,
                         self_trigger=True)

np.savez_compressed("AFE0_CH0.npz", wave=wave)
dev.close()
print("✅  Wave-form saved to AFE0_CH0.npz")

3 · Mean FFT (20 averages) and PNG export

"""
Compute the mean FFT of 20 spy-buffer captures (AFE0 / CH0),
display the plot and save the spectrum to a PNG.
"""
import numpy as np
from daphne_app.hardware.daphne import Daphne

dev  = Daphne("10.73.137.107")

waves = [dev.read_waveform(0, 0, samples=1000, self_trigger=True)
         for _ in range(20)]

# Compute + plot + save
freq, mag_dbfs, rms = Daphne.compute_mean_fft(
    waves,
    dt=16e-9,
    plot=True,
    save_path="fft_AFE0_CH0.png"
)

print(f"RMS amplitude  = {rms:.3f} ADC counts")
print("📈  Spectrum saved to fft_AFE0_CH0.png")
dev.close()

Trigger modes:
software ⇒ write 0x2000 once per read.
aligned ⇒ write 0x2020 & 0x2021 for centred buffer.

---

Contributing

Run tests with pytest. Pre‑commit hooks via pre-commit install.

---

© 2025 DUNE collaboration – MIT licence.