Astrophotography diagnosis of M20: Focus miss and moon gradient
RawUV IR CUT OPTOLONG15s
The Doc examined this image of M20 (raw, UV IR CUT OPTOLONG, 15s). Estimated overall technical quality: 4/10. 2 defects found: Focus miss (severity 3/5), Moon gradient (severity 2/5).
&w=1920&q=75)
Info
- Cible
- M20
- Site
- Bortle 2 · rural typique (VIIRS)
- Position
- 18h18m48s · -13°48'00"
The Bortle 2 sky is a major asset for signal-to-noise, but the 74% gibbous Moon largely negates it: it raises the background and imposes the measured oriented gradient (20%). For an emission target like M20 you can limit the damage with a dual-band filter, but the ideal is to shift the session outside moonlight, or wait for the Moon to set. M20 culminates low in the southern sky at our latitudes, worsening extinction and turbulence: shoot it near meridian transit to minimize the airmass traversed.
Setup
- Type d'image
- Brut
- Télescope
- DOBSON FLEXTUBE 300/1500 GOTO
- Caméra
- SmartEye IMX 533 color
- Filtre
- UV IR CUT OPTOLONG
- Exposition
- 15s
- Phase de lune
- Gibbeuse décroissante (74 %)
- FOV
- 36.0'
The 300/1500 at f/5 samples at 0.718''/px with the IMX533, which is fine or even slightly oversampled for typical seeing: the 8-14px FWHM reveal optical/atmospheric bloat more than a lack of resolution. The 0.60 degree field frames M20 comfortably, target centered and complete. Gain 350 is high and the 15s exposure short: a logical choice to freeze tracking on an alt-az mount (GOTO Dobson), at the cost of a barely lifted background (e_margin 2.32) and a need to stack many subs. The black point does not clip (0% dark clipping), so the offset is sufficient. Hardware priority: nail the focus (Bahtinov mask).
The diagnosis in detail
The frame is a single 15s sub of the Trifid Nebula (M20), clearly identifiable by its central multiple star and its radiating dark dust lanes. The provided DSS note actually corresponds to a different field (M16, the Eagle), so it cannot be used as a framing reference here, which has no impact on the diagnosis.
The dominant defect is star shape. The PSF plate, once the false noise spikes are discarded (the 1.4px zones), shows genuinely wide disks: 8.1px at center, 8.5px and 14.4px in the lower zones, i.e. about 6 to 10 arcsec. That is well above what the 0.718''/px sampling allows. Elongation stays moderate (1.27 to 1.36, barely above the 1.17 OSC floor) and the axes draw no clear directional streak: the problem is bloat (focus to redo, possibly worsened by seeing) rather than a tracking or field-optics defect. On a single sub one cannot separate focus from seeing, but redoing focus is the safest lever.
On the sky background, an oriented 20% gradient is measured and consistent with the 74% gibbous Moon, the only notable light source under this Bortle 2. It removes easily in processing. The short exposure leaves a barely lifted background (e_margin 2.32), normal in alt-az: compensate with a large number of stacked subs.
Priority actions
- Redo focus with a Bahtinov mask and aim for 2-3px FWHM before launching the run
- Stack many 15s subs to compensate for the short exposure and deepen the background
- Remove the lunar gradient in processing (GraXpert / DBE)
- Schedule future sessions outside moonlight to exploit the Bortle 2 sky
