Jones–Emberson 1 (also known as PK 164+31.1 or PN ARO 121) is a large and relatively evolved planetary nebula located in the constellation Lynx. It was discovered in 1939 by Rebecca Jones and Richard Emberson during a photographic survey of faint nebulae. The object lies at an estimated distance of about 1,600–1,800 light-years and represents the expanding gaseous envelope expelled by a dying low- to intermediate-mass star near the end of its stellar evolution. Because of its advanced evolutionary stage, the nebula exhibits a very low surface brightness and a fragmented shell structure dominated by emission from double ionized oxygen (O-III) and hydrogen-alpha (Hα). The central star of the planetary nebula is a hot white dwarf (16.6–16.8 mag), representing the exposed stellar core left behind after the progenitor star passed through the asymptotic giant branch (AGB) phase.

First Session back in December 2024

JoEm1 was actually one of my first targets shooting narrowband with the 10″ Newtonian back in December 2024:

Actual Session with the Celestron SCT 9.25″

On February 25, 2026 the installation of the Celestron C9.25 Schmidt-Cassegrain Telescope (SCT) was finally „somehow“ finished in the Observatory – so the new scope was ready to point on a prominent target. With the Celestron 0.63 reducers installed a focal length of 1.306 mm was finally realized, the scope therefore works at a focal ratio of f/5.56.

Within the first session I was able to collect some 4 hours and 45 minutes of narrowband data (3 hrs O-III and 1h 45min in H-alpha). The cause for not being able to capture the whole night was the Meridian Flip, which itself worked normally but revealed the SCT-problem of „mirror shift“…

Mirror Shift

In SCTs, focusing is typically achieved by moving the primary mirror along the optical axis using a threaded focusing mechanism. Because the mirror is mounted on a sliding baffle tube rather than a rigidly fixed cell, small mechanical tolerances are unavoidable. During focusing or when the telescope changes orientation relative to gravity, the primary mirror can shift slightly on this support. This phenomenon is referred to as mirror shift (or, in more pronounced cases during tracking across the meridian, mirror flop :-)). The shift causes a small displacement of the optical axis, which appears in the image as a lateral movement of the field or stars on the sensor. In imaging applications with long focal lengths, even minute mirror movements can produce noticeable star drift or changes in the image position. Modern SCT designs (such as the EdgeHD-series attempt(!) to reduce mirror shift through improved mechanical tolerances, mirror support systems, or mirror locking mechanisms. In astrophotography, the effect is often mitigated by focusing in a consistent direction, using external motorized focusers, or locking the primary mirror once focus has been achieved.

Of course (as always in the astrophotographers world there will be a possibility to implement the consistent focussing strategy somehow in N.I.N.A. as a workaround, but actually I am still searching …

An alternative way to avoid image shift in Schmidt-Cassegrain telescopes is the installation of an external focuser at the rear port of the telescope. Instead of achieving focus by moving the primary mirror along the baffle tube, the telescope is first brought close to focus using the internal focus mechanism. After that, fine focusing is performed with the external focuser, which moves the camera while the primary mirror remains stationary. Because the mirror no longer moves during fine focusing, mirror shift should effectively be eliminated.

Anyhow, the first collection of data (4h 45min) revealed a result that I found quite remarkable and somewhat unexpected. Although consistent stellar colors could not be derived from the narrowband data obtained through the O-III and H-alpha filters, the resulting image in the HOO palette nevertheless appears remarkably balanced:

2 Newtonian Sessions on the same Target

I ultimately found it interesting to make a comparison with more recent data acquired using the 10-inch Newtonian telescope. However, after two clear nights of observing became available, the comparison turned out to be somewhat uneven, since the Newtonian dataset accumulated nearly 10 hours of total integration time, and the star colors were also captured separately in RGB.

In the following, I briefly outline the image processing workflow in a few key steps.

Processing RGB stars

The RGB channels were acquired with integration times of 1 hour and 40 minutes each in order to derive natural star colors.

• Gradient Removal with GraXpert
• LinearFit (the weakest signal of channel B (mean = 7,016) as reference, R = 9,539, G = 8,142)
• RGB-ChannelCombination
• HistogrammTransformation
• SPCC
• ColorSaturation

Processing the Nebula

For the narrowband channel 4 hours was captured in H-alpha. The dominant O-III channel was planned with 6 hours, but only 5 hours and 40 min were available for integration.

• Gradient Removal and Background Extraction with AutoDBE (SetiAstro)
• StarXTerminator
• BlurXTerminator
• NoiseXTerminator
• HistorgrammTransformation
• GeneralizedHyperbolicStretch
• HDRMultiscaleTransformation
• ColorSaturation
• NoiseXTerminator

Bringing together a f/5.5 – SCT and a f/4 – Newton

• Mosaic-Script (better than StarAlignment)
• Dynamic Crop
• StarAlignment

Because the Newton Sessions (two sessions on March 6 and 7) provided 9 hours and 40 mins of narrowband data I combined the luminance data from the Newton with the Celestrons color data (only one session on March 25 with „just“ 4 hours and 45mins). What could be seen quite impressive are the increased details now visible in the combined version. And for the very first time – and I still dont really know why … – I decided to use the non-spike stars version for the final capture.

The final Image

What would the image look like when all the 2024 and 2026 sessions would be combined in one image… 26 hours and 20 min of total integration time!? Through the WBPP-Script in PixInsight the calibration, registration and integration of all the data is possible. The script provided an autocrop and I decided to combine the O-III channel (in total 125 frames) with the H-alpha channel (95 frames) using the H-alpha details as a Luminance channel (weight 30% only).

In the end, one sometimes cannot precisely explain why an image suddenly feels balanced and complete—but one recognizes the moment nonetheless; in this case it was the more restrained star appearance from the SCT, rather than the diffraction-spiked stars of the Newtonian, that seemed to harmonize naturally with the scene and quietly define the final image.