Astrophotography diagnosis of NGC 6992_biss: Sensor tilt and satellite trail
Raw300s11 juil. 2026
The Doc examined this image of NGC 6992_biss (raw, 300s). Estimated overall technical quality: 7/10. 2 defects found: Sensor tilt (severity 3/5), Satellite trail (severity 1/5).
&w=1920&q=75)
Info
- Cible
- NGC 6992_biss
- Date
- 11 juil. 2026, 22:28
- Lune
- Gibbeuse croissante 91% (sous l'horizon)
- Site
- Bortle 2 · rural typique (VIIRS)
- Position
- 20h56m03s · +30°47'20"
Excellent conditions for this target: Bortle 2 gives a very clean background and maximum SNR, ideal for capturing the faint filaments of the Veil. The Moon, although 91% gibbous, is below the horizon and 98° from the target: no lunar pollution is possible here. In Hα on an emission supernova remnant you are broadly immune to sky glow anyway. The background map shows only a very slight gradient (4%) with no real impact. Nothing to fault on the sky side, you gave yourself every advantage.
Setup
- Type d'image
- Brut
- Télescope
- EQMod Mount
- Caméra
- ZWO ASI585MM Pro
- Exposition
- 300s
- Phase de lune
- Dernier croissant (8 %)
- FOV
- 1.70°
Well-matched setup: at 375mm focal length the ASI585MM yields a 1.70° field that frames the Eastern Veil properly, with margin around the main arc. Sampling at 1.595"/px sits in the comfortable range for typical seeing, and the measured central FWHM (~2.3 px, ~3.7") confirms decent sharpness without oversampling. Gain 220 is high for this camera but reasonable in narrowband to cut read noise on 300s subs; sensor temperature -10°C is healthy. Purely mechanical point of attention: the sensor-plane tilt to correct on the imaging train, independent of gear choice.
The diagnosis in detail
The image is a very good starting point. Tracking is not an issue: no coherent directional elongation (PA dispersion of 28° between zones, no uniform straight streaks), the EQMod mount holds the 300s exposure well. The real subject is optical, not tracking. The PSF plate betrays sensor tilt: center C is tight (elong 1.15, FWHM 2.3/2.0) and the right edge stays round (R elong 1.05), while the lower-left and lower-center zones degrade (BL 1.18, B 1.19, FWHM rising to 2.9/2.4). The asymmetry is single-axis (horizontal 0.206 against vertical 0.021), which rules out simple symmetric field curvature or radial coma: this is indeed a tilted sensor plane. The corner/center ratio of 1.32 stays moderate, the image is perfectly usable, but stacking many subs will make those soft corners visible. On the sky and exposure side everything is consistent: Bortle 2, no Moon, clean background, only a negligible 4% gradient. The background/black-point margin (e_margin 1.85) may look low but that is normal in narrowband Hα, the nebulosity already stands out well: no underexposure to fix. The satellite trail at the bottom is anecdotal and will vanish under sigma rejection.
Priority actions
- Fix the sensor tilt: check the optical train orthogonality and adjust with a tilt plate/shims until opposite-corner FWHM is equalized
- Verify the exact corrector/reducer backfocus, which can worsen edge-of-field degradation
- Stack the series with dithering and sigma rejection (Winsorized/Linear Fit) to remove the satellite trail
- Re-measure tilt across several subs before/after adjustment to validate the correction
