Astrophotography diagnosis of SH2-112: Light pollution gradient and backfocus error

ProcessedMULTIBAND136×300s25 mai 2026

The Doc examined this image of SH2-112 (processed, MULTIBAND, 136×300s). Estimated overall technical quality: 7/10. 2 defects found: Light pollution gradient (severity 3/5), Backfocus error (severity 2/5).

Annotated image
Click to zoom

Info

Cible
SH2-112
Date
25 mai 2026, 00:00
Position
4h02m60s · +51°19'12"

Moon at 90% on the shoot night: this is the main source of stray background and largely explains the measured gradient amplitude (42%). With a multiband filter on an emission target like SH2-112 this choice is defensible since narrow bands reject part of the lunar continuum, which let you bank 136x300s of usable data. On a broadband RGB target this full Moon would have been a dealbreaker. To improve further, favor moonless nights or frame the target far from the lunar disk to reduce the background ramp to model.

- the Doc

Setup

Type d'image
Traitée
Télescope
Sky-Watcher Quattro 200P
Caméra
Player One Ares-C Pro
Filtre
MULTIBAND
Monture
Sky-Watcher NEQ6-Pro
Exposition
136×300s
Phase de lune
90%
Notes
Logiciels : PHD2, Siril, Nighttime Imaging 'N' Astronomy (N.I.N.A. / NINA)
FOV
1.80°

The Quattro 200P f/4 with the Ares-C Pro gives a 1.80 deg field, well suited to SH2-112, which sits in the central region without being lost: the framing centers it correctly with margin for the background. The Newton f/4 sampling stays comfortable for this target. The only setup point to fix is the coma corrector backfocus: the PSF panel shows FWHM rising symmetrically in the corners (ratio 1.21), typical of a slightly out of tolerance sensor distance on this demanding f/4. Fine adjustment in 0.5 mm steps will make corners match the center.

- the Doc

The diagnosis in detail

The image is well crafted: the SH2-112 core is well defined, stars are point-like at center (FWHM 5.86/5.51 px, elong 1.06, PA dispersion 31.8 deg indicating fairly round stars) and tracking on the PHD2-guided NEQ6 is not a problem. No trace of tilt (very low horizontal/vertical asymmetry, 0.02), drift or guiding oscillation.

The dominant flaw is cosmetic and fixable: a linear background gradient at 42% oriented 292 deg, whose plane dominates the radial component by a factor of 4.9. The 90% Moon is the primary cause, amplified by incomplete gradient subtraction. This ramp matches no DSS structure, so it is not IFN: it must be cleanly removed on the linear image before any stretch.

The second point is optical and mild: FWHM rises symmetrically in all four corners (corner/center ratio 1.21, elong up to 1.18 in BL), without true coma, pointing to a slightly out of spec field curvature/backfocus on this f/4 Newton rather than a collimation fault. A fine corrector adjustment will be enough.

Priority actions

  1. Remove the gradient on the linear image (GraXpert or Siril) before stretching to flatten the background
  2. Adjust the coma corrector backfocus in 0.5 mm steps until corner and center FWHM match
  3. Favor moonless nights (or frame away from the lunar disk) to reduce stray background