What is coma and why needs to be corrected. Coma is a special type of an optical aberration. It’s mainly visible in the astrophotography – the stars are deformed into comet-like shapes:
Fast Newtonian telescopes suffer from comatic aberration, mainly in the corners of the photos. These telescopes are nowadays very popular between amateur astrophotographers just like me, because they offer the best price/aperture ratio. There is easy way how to suppress the comatic aberration – coma correctors. There are many coma correctors on the market, but there are just a few published reviews or comparisons between them. Therefore I decided to perform the comparison myself. Quite big collection of coma correctors has assembled in my drawer ready to be tested:
I tested five correctors: ASA 0.73 reducer, Baader Multipurpose coma corrector (MPCC), GPU coma corrector, Explore Scientific and TS Quatro. The details are summarized in the table below. I expect the question why didn’t I test TeleVue Paracorr. Well not mentioning that it’s quite expensive, therefore out of the budget for this exercise, it has been proven to be probably the best 2″ coma corrector on the market. Only disadvantage is that it increases the focal length by factor 1.15. So, you cannot go wrong with this corrector.
Probably the mostly used sensor for astrophotography is Kodak KAF 8300. It’s sufficiently large, it has sufficient resolution (8MP), and it’s available in mono or color version. I have the mono one, because I live in sub-urban area with relatively strong light pollution. Therefore narrow band filters are very useful. There are many manufactures producing astro-cameras with this sensor. I picked Moravian Instruments, because they have very good references.
My main telescope used for astrophotography is ONTC 10” f4 Newtonian from Telescope Service. It has carbon fiber tube, which is very stiff and it has very low thermal expansion. The optical tube assembly (OTA) is equipped with 2” Feather touch focuser – probably the best 2” focuser on the market. In order to reach the focus, the primary mirror has to be shifted. For ASA reducer towards the secondary mirror, for Baader MPCC and GPU correctors was in the middle and for Explore Scientific and TS Quatro outwards from the secondary mirror.
Mount and polar alignment
My primary equatorial mount for astrophotography is Gemini G53f with friction drive, which means no backlash. It’s excellent mount and I am very happy with it. In case of OTA or camera I can imagine an upgrade, but in case of mount I cannot think of anything better. It never let me down and number of bad pictures due to bad tracking or backlash is basically zero. Because every time I have to move the mount outside, I have to perform always polar alignment. I tried many methods, and ended up with PolMaster from QHYCCD. The advantage of this method compared to conventional polar finder is that the OTA and counter weights are assembled prior to polar alignment. The best way how to check the quality of the polar alignment is to autoguide and checks the corrections in declination axis. Here is a screengrab of the PHD II guiding performed by Orion Mini Autoguider. As you can see there are almost no correction in declination axis needed and correction in right ascension axis are even bigger. And yes, as I said I am very happy with Gemini mount.
Collimation of the Newtonian telescope could be a separate topic and strongly influences the quality of the pictures. Therefore I tried to be as precise as possible. This means, for “rough” collimation I used Howie Glatter laser collimator with TuBlug. This is probably the best laser collimator one can buy. For the “fine” collimation I used CatsEyes auto collimators. Anyway only very small adjustments were needed between these methods.
In order to confirm the precision of the collimation, the defocused star test was done prior every imagining session. Here is the picture of defocused star – I think the collimation is nearly perfect:
Here again, I tried many methods – FWHM or Bahtinov mask, but I ended with Gold focus mask. This focusing technique is unique, because for the evaluation of the focus the special software is used, which tells you whether you should focus in or out and how good focus you achieved. It’s quite fast and very accurate method:
The telescope was always properly collimated and tempered for 2 hours prior imaging session and camera’s sensor was cooled down to -30 °C. For the comparison is best to choose the area of the sky where are many stars. Therefore I chose constellation Cassiopeia. The bright star Cassiopeia Gama was always centered in the middle of the sensor, first defocused in order to check the collimation and then focused with gold focus mask. Autoguiding was switched on and 3 pictures on L filter with exposure time 120 s were taken for each corrector. Since it’s necessary to shift the primary mirror for different correctors I had to split imaging sessions every time I had to shift the mirror. This is mainly caused by the fact that I cannot shift the mirror outside in the dark, but I had to move the telescope inside, which means I get a lot of dew on the primary mirror and temperation is gone. Therefore I used one night for tensing the ASA corrector, second one for MPCC and GPU, and third one for Explore Scientific. The pictures were taken by SIPS software in linear form (.fits) and post-processed by Pixinsight. Bias and Dark were subtracted from the picture and 1% of highlights and shadows were clipped by histogram transformation.
So, let’s have a look at the real pictures. First without the corrector:
As you can see the coma is present in every corner, whereas in the center the stars are still quite round.
Upper left corner – no corrector:
GPU coma corrector:
Upper left corner – GPU coma corrector:
Upper left corner – Baader MPCC:
Upper left corner – TS quatro:
Upper left corner – Explore Scientific:
Upper left corner – ASA reducer:
Well and the winner is… Subjectively I would say that the best star shapes has Explore Scientific CC. The stars are circular in every corner, but it’s obvious that the focal length was increased. Second best subjective rating reached GPU. Only in upper right corner has slightly oval stars. Third best (still subjective) CC is Baader. The stars are slightly oval in every corner, but it’s nothing dramatic. Compared to uncorrected pictures the coma was nearly suppressed. On the other hand TS quarto CC performs very badly. I would say that the stars are more deformed than in case without the corrector. The comets are flying outwards the center. One could say that it’s not coma corrector, but coma creator 🙂 And yes, I didn’t forget to use 15 mm spacer between the camera and the corrector, because back focal distance is 70 mm and not 55 mm as for every other CC. With ASA reducer it’s quite tricky. Bottom 2/3 of the picture looks great. However, upper 1/3 of the picture is out of focus. The upper bright star has double diffraction spikes. This is caused by the double spider wanes and it implies that the picture is out of focus. How this could be? Collimation is perfect, so it must either be something between the corrector and sensor or between the corrector and primary mirror. In order to disprove the first variant I rotated the camera by 180° and I got this:
Basically the same outcome – the upper 1/3 of the picture is not focused. Very bright star in the upper right corner has double diffraction spikes, which is the indication of defocus. This means there must be something with the mirrors. I realized that another source of problems could be misalignment of the secondary mirror with the focuser. There is one instrument, which can detect this problem easily and it is Spheretec 2” Concenter eyepiece. It is basically Chesire tube with transparent plate with circular markings. It works like standard film container Chesire – primary mirror must be covered and it is better to place white paper underneath the secondary. My mirror was slightly out of center, but it was nothing dramatic, so I corrected it and here is the view through the focuser:
Unfortunately it didn’t help:
Non-subjective analysis of the picture quality can be done by CCD Inspector. 2D curvature map provides the information about the collimation and the tilt of the sensor. So let’s have a look:
GPU coma corrector:
Well, it’s quite interesting that CCD inspector cannot detect very bad stars in case of TS quatro corrector. Moreover there are quite big difference in collimation between GPU and MPCC, but in reality, I just removed the camera from the focuser, changed the corrector and placed the camera back. Probably this could be caused by the tilt during fastening the corrector in the focuser. ASA corrector has higher FWHM in the upper part of the picture compared to no corrected picture – this is very disappointing.
Another very useful feature of CCD inspector.
GPU coma corrector:
The lowest vignetting occurs logically without any corrector, but in case of Baader MPCC, TS Quatro and Explore scientific the vignetting was not increased significantly. On the other hand GPU and ASA are vignetting strongly.
If you are meticulous person just like me, get Explore Scientific cc. It has pin point stars in every corner and very acceptable vignetting. Only disadvantage is slightly increased focal length. If you want to avoid that and keep the original focal length get GPU cc. This corrector has two disadvantages: it’s long (100 mm), so it may stick out of the focuser inside the OTA and block some light and it has quite big vignetting – flat field is a must. If you are on budget, get Baader MPCC. Advantages: cheap, short, low vignetting. Disadvantages: the stars are bit oval in the corners. TS quarto cc was not correcting the coma at all, therefore not recommended. ASA reducer was a big disappointment. Even though this was the most expensive corrector in the test, I was not able to get it working properly and 1/3 of the picture was defocused. The problem is most probably caused by the tilt; however none of the other tested coma corrector suffers from the tilt. Anyway, I will get the tilting mechanism and try to make it working, so stay tuned. Initially it supposed to be a comparison of several coma correctors, but it turned into “how can I make ASA reducer working”.
Update on 31.10.2017
As I promised, I continue with my effort to make ASA reducer working. Well, the last option is the tilt between the corrector and the sensor. So, I got myself a tilter from TS:
This one is completely useless for me, because it has male T-thread and simultaneously the adjusting screws on camera side. If you screw it on the camera, the push-pull screws are completely blocked by the filter wheel. If you have a filter wheel, don’t buy this tilter!
Another tilter can be purchased from ZWO. This one has the male T-thread, which can be screwed off and screwed on the other side. This is exactly what I need. In the end it looks like this:
Back focal distance (BFD) should be 55 mm for ASA reducer. Somewhere I found that the corrected diameter can be increased if this distance is reduced by 1 – 3 mm. My BFD is: distance from the chip to the surface of the filter wheel 33.5 mm, I changed T-thread adapter to shorter one, which has 7.5 mm, tilter has 11 mm + 2 mm male T-thread. This means 33.5+7.5+11+2=54 mm – perfect.
Let’s go to tilt some sensors
Based on analysis of the pictures by CCD inspector it was clear that I have to tilt the sensor along the axis, which is parallel to the wider side of the sensor, as it is illustrated on following sketch:
This means to turn the upper screw and leave bottom ones untouched or vice versa. I pointed the telescope to Vega and started carefully to adjust the screws. I made small marks on the screws, in order to know how much I turned. Initially, I turned two bottom screws by ¼ of a turn and refocused by gold focus mask:
Upper side of the picture has donuts and double diffraction spikes – as bad as before. Vega has obviously diverging difraction spikes, which is caused by not parallel spider wanes. This I will correct later.
What CCD Inspector says? Well, he gently advised me not to be afraid and push more:
So, I took the advice and turned by one full turn:
Well, it was too much, so I turned ¼ turn back and here we go:
This should be nearly perfect. I was so excited and happy for a short time. If we have a look at the two brighter stars at upper center part of the picture, we can see that these stars have still double diffraction spikes. On the other hand I got rid of donut shape stars even in the corners.
Tilting ring definitely improved the situation and it’s only way how to get acceptable pictures from ASA 0.73 reducer. However, the problem was only reduced, not fully eliminated and some stars in the corners have still double diffraction spikes. Very helpful tool in this process is CCDInspector and I would advise to purchase it, if you want to perform similar exercise.