Astrophotography diagnosis of NGC_7635_Bubble_nebula: Noise / underexposure, sensor tilt and 1 other
Master46×600s28 sept. 2026
The Doc examined this image of NGC_7635_Bubble_nebula (master, 46×600s). Estimated overall technical quality: 6/10. 3 defects found: Noise / underexposure (severity 3/5), Sensor tilt (severity 2/5), Clipped stars (severity 2/5).
&w=1920&q=75)
Info
- Cible
- NGC_7635_Bubble_nebula
- Date
- 28 sept. 2026, 20:25
- Position
- 23h20m49s · +61°12'44"
A waning gibbous Moon at 94% on the night of September 28: this is the worst possible configuration for broadband acquisition, and yet the measured background stays remarkably flat (plane gradient of only 3%, azimuthal anisotropy 0.02). Two readings are possible: either the Moon was low or far from Cassiopeia during the subs, or a selective filter was used, which would also explain the very dark background and the e_margin of 2.08. Either way, since NGC 7635 is an emission target dominated by Ha, dual-band or narrowband acquisition under this Moon is the right instinct. If the subs were shot in broadband without a filter, you got lucky on the Moon-target geometry.
Setup
- Type d'image
- Master
- Télescope
- iOptron CEM120,70,40,26, GEM, HEM, HAE, HAZ series and SkyHunter Mou
- Caméra
- Poseidon_C
- Exposition
- 46×600s
- Phase de lune
- Waning gibbous (94 %)
- FOV
- 50.7'
A coherent but demanding setup. At 1624 mm and 0.478"/px you are clearly oversampled for typical seeing: with a measured FWHM of 8.9 px at center, roughly 4.3", you can bin 2x2 without losing anything and gain a factor of 2 in SNR. The 0.84 degree field frames the bubble adequately, but NGC 7635 sits dead center while the associated nebulosity extends westward: an offset framing would have served the whole complex better. On settings, gain 125 and -15C are healthy, and the 600s exposure is justified at this focal length. The offset of 10, however, is too low, with 94% dark clipping: raise it to 30-50. Finally, this focal length makes the imaging train very sensitive to tilt, hence the measured asymmetry.
The diagnosis in detail
The most instructive aspect of this master is star shape. The center is nearly perfect (elongation 1.01, FWHM 8.89 px) and the PA dispersion reaches 45 degrees, which rules out any tracking component: across 46 subs of 600s, the mount held perfectly. The degradation is purely spatial and above all single-axis: horizontal FWHM asymmetry 0.212 against 0.029 vertically, with the left flank of the sensor (TL 11.09, L 10.82, BL 11.93) bloated by 26% against a right flank that stayed tight (R 8.79, BR 8.66). This is neither coma (which would be radial and symmetric) nor backfocus error (which would degrade all four corners equally): it is a tilt of the focal plane, modest but real at 1624 mm where depth of focus is very shallow.
The second axis concerns exposure. With an e_margin of 2.08 and 94.4% dark clipping, the background is not lifted off zero: the offset of 10 truncates the left tail of the histogram and you lose the faintest information, precisely the information carrying the bubble's extensions. The result is visible in the image, the outer halo of NGC 7635 appears only in fragments where 7h40 of integration should deliver it plainly.
The rest is healthy: flat background, no significant gradient despite the Moon, clean calibration, no walking noise. Both corrections are simple and belong to acquisition, not post-processing.
Priority actions
- Raise the offset from 10 to 30-50 to eliminate the 94% dark clipping and recover the bubble's faint extensions
- Correct the imaging train tilt by cross-tightening every connection, then adjust the left side with a tilt plate in 1/8 turn increments
- Bin 2x2 at acquisition to move from 0.478 to 0.96"/px, matched to real seeing and doubling SNR
- Add integration time (target 15h+) and dither between every sub
- Acquire a set of short exposures (15-60s) to rebuild the clipped bright star cores
