How far can an ADMX-like haloscope reach? A corrected sensitivity model, checked against published limits

I built a small, open tool that answers a simple question: for a microwave-cavity haloscope like ADMX, what is the smallest axion–photon coupling it can detect, as a function of axion mass? The textbook formula is easy to write down and easy to get wrong. This model includes five corrections that matter in practice: The cavity volume, form factor, and quality factor change with frequency The system temperature varies with frequency, with a quantum-noise floor Duty cycle and tuning overhead eat into the scan time The axion signal has a Maxwell–Boltzmann lineshape, with a matched-filter gain The overall normalisation is calibrated to ADMX Run 1B At the default settings (7.6 T, loaded Q of 80,000, system temperature 0.40 K, 100 s per tuning step), an ADMX-like search beats the DFSZ benchmark up to about 3–4 µeV and then falls behind, reaching about 4.3 times DFSZ by 40 µeV. The result is checked against published bands from ADMX, HAYSTAC, and CAPP. It agrees to within a factor of about 1.5 for ADMX and HAYSTAC Phase 2a, and it is honest about where it does not — for example, it is about three times too weak for CAPP-12TB at these settings, because CAPP ran colder and longer. This is a preprint, not a peer-reviewed paper. It is a scoping tool, not a replacement for a full experimental analysis. It does not include mode crossings, tuning-rod gaps, gain drift, RF interference, or a full likelihood analysis, and it should not be used to claim an exclusion. Files in this record: Rodger_ALP_Haloscope_Sensitivity.pdf — the paper alp_haloscope.py — the simulator, one file, command-line + Streamlit example_curves.csv — example measured curves (C, Q_L, T_sys vs frequency) to try the CSV import README.md — how to run the simulator To reproduce the paper figures, run the Streamlit app and use the Validation tab, which exports a provenance JSON file recording the exact parameters behind each figure.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-21
DOI
https://doi.org/10.5281/zenodo.24424160
Primary Topic
Dark Matter and Cosmic Phenomena
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

How far can an ADMX-like haloscope reach? A corrected sensitivity model, checked against published limits

Andrew Rodger
Zenodo (CERN European Organization for Nuclear Research)
Dark Matter and Cosmic Phenomena
preprint

How far can an ADMX-like haloscope reach? A corrected sensitivity model, checked against published limits

Andrew Rodger
preprint en

Abstract

I built a small, open tool that answers a simple question: for a microwave-cavity haloscope like ADMX, what is the smallest axion–photon coupling it can detect, as a function of axion mass? The textbook formula is easy to write down and easy to get wrong. This model includes five corrections that matter in practice: The cavity volume, form factor, and quality factor change with frequency The system temperature varies with frequency, with a quantum-noise floor Duty cycle and tuning overhead eat into the scan time The axion signal has a Maxwell–Boltzmann lineshape, with a matched-filter gain The overall normalisation is calibrated to ADMX Run 1B At the default settings (7.6 T, loaded Q of 80,000, system temperature 0.40 K, 100 s per tuning step), an ADMX-like search beats the DFSZ benchmark up to about 3–4 µeV and then falls behind, reaching about 4.3 times DFSZ by 40 µeV. The result is checked against published bands from ADMX, HAYSTAC, and CAPP. It agrees to within a factor of about 1.5 for ADMX and HAYSTAC Phase 2a, and it is honest about where it does not — for example, it is about three times too weak for CAPP-12TB at these settings, because CAPP ran colder and longer. This is a preprint, not a peer-reviewed paper. It is a scoping tool, not a replacement for a full experimental analysis. It does not include mode crossings, tuning-rod gaps, gain drift, RF interference, or a full likelihood analysis, and it should not be used to claim an exclusion. Files in this record: Rodger_ALP_Haloscope_Sensitivity.pdf — the paper alp_haloscope.py — the simulator, one file, command-line + Streamlit example_curves.csv — example measured curves (C, Q_L, T_sys vs frequency) to try the CSV import README.md — how to run the simulator To reproduce the paper figures, run the Streamlit app and use the Validation tab, which exports a provenance JSON file recording the exact parameters behind each figure.

Zenodo (CERN European Organization for Nuclear Research)
Dark Matter and Cosmic Phenomena
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon