Imaging test (8th June 2024)

So given it was a non-work night and predicted to be clear Dave invited me over to test the gear/setup prior to the BASEG Tenerife 2024 trip.

Given we are fast approaching the summer solstice, the nights are getting shorter and lighter. The Annual Darkness chart for IMT3 shows that we currently have no astronomical darkness and the nautical darkness lasts only for some 3.5hrs (left image) and now compare that the darkness graph for Mount Teide in Tenerife (right image) and although we lose some dark hours around the Summer Solstice it’s not as much as is lost being located at 52°N.

Setup – added BlueAstro stick station to measure pressure as Pegasus have not exposed the pressure measurement from the NYX101 mount to the ASCOM layer … why not ? Added weight bag to NYX-101 and GL.inet travel router to top of the scope as MS Windows keeps messing up the Wi-Fi hotspot on the Mele Quieter3C if it does not detect a internet connection ….. how stupid is that. Now I have a permanent hotspot thanks to the instructions given by Cuiv the Lazy Geek on his YouTube channel.

When attempting to polar align when using the QHY PoleMaster I noticed that the sky brightness below 15 mag2/sec (measured by a Unihedron SQM) produced a white screen due to over exposure – the minimal exposure in the now aged the QHY PoleMaster software (> 4yrs since last release) was 50ms which is too long even though I could eyeball Polaris in the early evening sky.

NINA 3.0 start up had not detected the QHY native driver and after I shutdown and restarted the app it then was able to detect the QHY 268C camera. However, it disconnected when it could not cool to -10℃ which I’ve never encountered before. I did eventually managed to get it cool and stay connected.

During guiding calibration OpenPHD2 would constantly complain about losing the star. Again the star was clearly visible on the PHD display and after downgrading from 2.6.13dev4 to dev3 and then I suddenly realised the value the error was referring to. I changed the minimal star HFD down to 1.0 from 1.5, also recreated the dark library to remove the possibility of the guider attempting to guide on a hot pixel.

Once guiding, the guide graphs were reporting 0.08 – 0.19 arcsec total polar alignment error. Hopefully I will learn to improve that and maybe repeat the polar alignment procedure or use PHD drift align to refine it.

I also forgot to change filter at the start of the evening from the Antila Quad Band filter to the Baader UV/IR filter. I noticed a halo from the bright star Arcturus in the constellation Bootes which I was testing autofocus on but I decided to continue regardless as I was only testing.

The goal for the night was to grab a base image of T Coronae Borealis (“The Blaze Star”) before it goes nova – it’s a reoccurring nova with a 80 year cycle. The star is currently hovering around 10.2 magnitude and it is predicted based on previous eruptions to reach around magnitude 2. As I ultimately wanted to perform some photometry on the star using the UV/IR cut filter I did want to blow out the cores of any bright star so I opted to use the ExtendedFullWell2CMS mode at gain 0, offset 30 with 60 second exposures. During the sequence I forgot to watch HFR/Star count graph and it rose above 5 which meant I have to refocus to (HFR ~2.0) and so due to my oversight due to chatting I will have to dump a lot of the early subs.

At the end of nautical darkness I stopped the sequence and used the NINA flat wizard (dynamic exposure) with the old PegasusAstro 120 flat panel at 100% brightness to create 25 flats and flat darks for a target of 33% ADU. After packing up and the with the pre-dawn temperature hitting 4℃ just before 4am I was looking forward to getting to bed.

At the next clear non-working evening I will attempt to grab some subs again but this time using the correct filter. Thank you to Dave and family for hosting me once more.

NB – NINA 3.1 was formally released the following day (09/06/2024) !

Baader UFC, QHY OAG-M and QHY268C

My current setup for OSC on the Takahashi FSQ-85 uses the QHY-OAG-M and the various QHY metal spacers which you screw through the OAG onto the CAA that fronts the QHY286C.

However, I’ve never liked the screw adapter approach for this sensor as unnecessary back focus is wasted due to the weird CAA to QHY286C not being a screw thread. The lack of an IR cut window on the QHY268C sensor means I need to sit an additional filter in the QHY 020055 M48 (F) adapter, in order to change the filter I would have to take the imaging train apart to remove it !!!

Baader Universal Filter Changer

Enter the Baader Universal Filter Changer (UFC) , the chart below details the core base changer alongside the adapters required to match to most common setups.

Baader Universal Filter Changer – Adapter Ecosystem

You can purchase additional sliders than can hold 50mm, unmounted and threaded 2 inch (M48) filter. The slider is magnetic so it will not dislodge during slews or during scope assembly or disassembly.

The 4 assembled components of the Baader UFC

Proposed Set-up

The Takahashi Flattener (TKA37582) has a metal back focus distance of ~56.2 mm so I need to get close to that but being under that distance is better than being over as I can add additional spacers if needed. The effect of adding a 2mm thick 2-inch filter decreases the optical backfocus by 0.6mm (2mm/3) so I will need to add 0.6mm spacing to account for the filter effect which results in a total distance of ~56.7mm.

OrderComponentThickness (mm)
2QHYCCD CAA M54(f) (#020002)6
3Baader UFC M54(m) camera side adapter (#2459117)2
4Baader UFC base (#2459110)13
5Baader UFC SP54(m) telescope side adapter (#2459134)1
6QHYCCD OAG-M M54(f) (#020079) adapter3
7M54 Metal Spacer0.5
9QHYCCD Metal Spacer0.4
10QHYCCD OAG-M M54(f) adapter3
11M54 Metal Spacer0.3
12Baader M54 x 0,75 Inverter Ring 0
New Imaging Train

I should now be able to change filters much easily albeit at an additional cost but the flexibility should be worth it. Of course I will need to test the star shapes in the field to make sure that I have the correct back focus and add extra distance if needed.


The majority of the Baader UFC components were purchased from FLO (First Light Optics) and the remaining adapter (#2459134) from Tring Astronomy. Delivery and communication by both vendors was great, shame about the Royal Mail industrial action messing deliveries about during the Christmas holiday.


My gratitude to Stuart1971 from StargazersLounge who helped me with suggestions to try to sort out the QHY268C noise banding issues.

Imaging Report 15th-16th December 2022

As it was a work night and Dave was working I was imaging from the back garden and not the dark site. I started late in the evening and had issues sorting out the autofocus in NINA which meant I was even later than I had planned.

Using the OptoLong L-Pro filter and the native camera driver (not ASCOM) :

QHY268C Nina Native Mode

The target was decided by Dave as M31 (Andromeda Galaxy), I decided on 120 second exposures at -15℃, gain 0 and offset 30 (Photographic Mode). I finally started to capture data at 19:55 when M31 had already transited 45 minutes earlier.

M31 single sub (stretched) via NINA Imaging Session

I believe the dark shadow to the left is the QHY OAG prism, I may have to reduce it’s depth in the light path or rotate it 45 degrees so it sits on the long horizontal axis.

Once M31 had reached approximately 50 degrees altitude and was starting to set in the West my guiding started going awry especially in RA. I put this done to either balance issues or cable snagging so that’s something I need to look at in the future.

After shutting down earlier than I thought due to the guiding issue and the pending rising Moon I decided to do my darks, flats and dark flats. However my Pegasus Flatmaster refused to connect to NINA so I decided to fix and complete this task on a different day.

I only managed to capture 62 subs which totals just over 2 hours. I was hoping to get at least 6 hrs for a single session and something approaching > 12 hours which means I will need several more nights to capture my required amount of data for M31.

Not a great first outing but given I hadn’t imaged in ages this was expected. I can honestly say that although a OSC coupled with the new harmonic mounts such as the ZWO AM5/Pegasus Nyx are very convenient as part of a lightweight holiday travel setup. OSC also is easier processing but still I can say I’m not a total fan due to the loss of high frequency data, high HFR focus issues and in the case of the QHY268C the noise banding issue which although appears to be removed by darks etc still is unnerving.

Update (17/12/2022)

Have resolved my Flatmaster 120 panel issues but I feel I’m better off with a fully controllable fixed panel instead of a manual panel so I will look into that.

My Darks, Flats and Dark Flats are now done (25 of each) all performed at -15℃ to match this session temperatures but there is still not enough actual data to start processing.

Equipment Check-up

So after ages I’m planning to perform some imaging after being challenged by Dave to produce a OSC image of M31 with the QHY268C that can be of print quality.

The first step was upgrading all the necessary software to make sure both the NUC and laptop would not be interrupted during my planned imaging session.

  • Windows10 latest Updates (Dec2022)
  • SGPro (4.2.0936)
  • NINA (2.1 HF1 Beta002)
  • ASTAP (2022.12.09)
  • ASCOM (6.6SP1)
  • Pegasus Ultimate PowerBox (1.6.1230.46)
  • iOptron commander (

The iOptron CEM60 mount hadn’t been updated in ages and it showed :

  • HandControllerV2 : 210106
  • MainBoard : 190716
  • RA board : 201005
  • DEC board : 190716

After performing the firmware update procedure, the mount was now showing the following levels :

  • HandControllerV2 : 220119
  • MainBoard : 211018
  • RA board : 211018
  • DEC board : 211018
  • Catalog : 150429

Of course I will need to perform a new zero mount position following the upgrade. Tonight I will also be changing the OpenPHD algorithm to PredictivePEC as described by Cuiv the LazyGeek :

Processed Image (M45)

So visiting Dave one evening as we have not met for a while whilst I was updating software Dave processed the QHY268C data we took of M45.

M45 – PixInsight Processed Image

So I’m disappointed that although the image is a good first start I forgot to change the change setting on the gain which in SGPro is in the event settings and not in the top level sequence display 🙁

Dave ran the image through the annotate function of PixInsight. The galaxy PGC13696 near the bottom of the image is actually 232 million light years away.

M45 – PixInsight Annotated Image

Viewing Report 10/11th December 2021

Unusually it was a clear Friday evening. I did plan to be ready to go as soon at the pole star was visible but my imaging PC insisted on updates and the local hard disk was running at 10MB/s (replacement SSD on the way).

By the time I was ready, mount setup, polar aligned and balanced it was already late. I decided not to use the latest SGPro or NINA beta but just use the existing SGPro version. I was delayed starting as I was having issues with SGPro hanging when it couldn’t talk to the SQM (ASCOM Conditions Observing Hub) on a previous COM port, I need to report this back to the devs as a bug.

At this point Peg-Leg Dave joined me on a video call and we discussed imaging M45 in different modes on the QHY268C OSC. So we moved the scope to Alp Ari and proceeded to plate solve in SGPro, sync’d the scope Cartes Du Ciel and calibrated OpenPHD2.

Using the SGPro framing and mosaic wizard to decide on the framing for the target sequence I wanted as much of the reflection nebula as possible rather than being dead center.

M45 – SGPro Framing Wizard (FSQ85/QHY268C)

I’ve used the multi-star guiding in OpenPHD2 since it was first released in an earlier beta and I know Dave is looking forward to using it when he moves from using an OAG on his 12-inch RC to a 90mm guide scope to make it easier to get more guide stars or even one star.

Multi-Star Guiding in OpenPHD2

Whilst trying some mode/exposure tests the guiding started acting up in RA, so parking the mount and disengaging the clutches I redid the balance of the scope. It was only marginally off but it was enough to cause issues for the CEM60 …. it is not forgiving !

We decided to increase the Gain/Offset to 15/75 and use the Extended Full-Well mode (#2) of the QHY268C, testing the star brightness levels of various exposure times we opted for 180 seconds as that was under the maximum brightness level.

As I currently have no IP camera outside I like to see the mount position using GSPoint3D as I like to view where it is especially during meridian flips. NINA has this built-in now in the recent version 2.0 betas. As SGPro lacks this functionality I can use the view via this is standalone version that connects to the ASCOM mount.

Mount/Telescope virtual view
M45 – Pre Meridian Flip

SGPro paused the guiding just prior to the meridian flip. Following the automated flip, the guider and the imaging sequence automatically restarted after a plate solve and auto centering were performed.

Inverted shot of M45 (180s) – Post Meridian Flip

It gradually got cloudier just after midnight and the quality of the subs declined so I decided to stop acquiring data even though we really wanted over 4 hours of exposure.

I proceeded to take calibration frames. Using a target ADU of ~23,000 the SGPro flat wizard on the Pegasus FlatMaster (100%) gave an exposure time of 9.68s for the Optolong L-Pro filter, 25 flat-frames were taken followed by 25 dark-flat frames of the same exposure time and finally 25 dark frames of 180 seconds.

It was at this point that I realised that the FITs header showed a gain level of 0 and not 15, the offset was correct but I can’t be sure if the EFW mode was used as it’s not in the FITs headers. Only when using the native driver in NINA can you set the mode within the sequence, in SGPro the mode is set in the external ASCOM driver when the camera is not active in SGPro even though though it’s in the ASCOM API as the Camera.ReadoutModes property.

Also for some reason the default setting in the QHY driver is to NOT disable the overscan area which means I have black borders on my images which will make processing the data in Pixinsight a challenge !

I actually got to bed after 3am even though I had planned to stay up until the dawn. Next morning I noticed that my counter-weight had slipped and rotated on the bar. This may have also caused some of the issues with the guiding so I need to set-up earlier and check things more thoroughly in future to avoid these mistakes.

So although it’s not the data we planned it will be worth processing over a wine. The evening was a really a useful experiment and hopefully lessons will be learned …. if I remember the next time.