Skip to content

Restore strict flux-scale QA on 2026-01-25 hour 22 #157

Description

@jakobtfaber

Question

Why does the 2026-01-25T2200 hourly-epoch mosaic fail strict flux-scale QA at median ratio 0.688 after same-date BP generation and epoch-gaincal LOW_SNR fallback to the static daily G table, and what minimal correction makes the same checkpoint pass 0.8–1.2 without threshold changes or --lenient-qa?

Acceptance: rerun the same 11-MS hour with strict QA; flux scale, completeness, and noise gates all pass; photometry runs; the public dashboard shows the QA verdict, mosaic/diagnostic previews, and light-curve data. Preserve the measured 12.658 Jy peak and compare the same checkpoint before/after.

Evidence: /data/dsa110-proc/products/mosaics/2026-01-25/run_2026-07-21T02_39_45Z.log, 2026-01-25_run_summary.json, and /stage/dsa110-contimg/images/mosaic_2026-01-25/2026-01-25T2200_mosaic_qa_diag.png. Related calibration finding: #104. Blocks #151.

Part of #144

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions