Astronomy Specialist


Adriano Massatani

Posted on January 16 2020



Testing a telescope could be a complicated and costly job considering the necessary equipment, the appropriate metrology to use, the measure of the error and technical issues that may arise. On the contrary, the star test can be easily performed by anyone analysing a star at high magnification. The atmospheric turbulence however can highly distort the Airy disk, making this test prone to misjudgement. The Ronchi test is easier as it is not affected by turbulence but as per the star test, it does not provide a numerical result.

Star tests performed indoor or the use of an interferometer provide a more comprehensive and precise analysis but they are only going to analyse the performance in the center of the optical axis and they don't provide information on the extra-axials aberrations.

On the contrary, the software CCD Inspector only needs a starry night and a camera to measure the collimation, vignetting and aberrations using a set of photos. Differently from the other tests, this software provides information across the entire area of the focal plane rather than on the optical axis only. 


How it works

CCD inspector analyses the size of the stars in a photo and it draws a map that clearly shows the aberrations and the distortion across the entire image. To have consistent and valid results, some precautions have to be taken:

  • The photo should contain at least 100 stars evenly spread in the field so no globular or star clusters.
  • No bright, saturated or elongated stars as they would alter the measure of FWHM (Full Width Half Maximum).
  • To reduce the error, take a set of multiple shots in three different areas of the sky. Ten shots per area would be ideal.
  • Focus accurately and guide during the exposure. 
  • Use photos with a FWHM >2"
  • The sampling should get a FWHM of at least 2 pixels. For example if your image scale is 2"/pixel and the FWHM is 4", the FWHM is equal to 2 pixels. 
  • The photos should have an exposure between 30 and 60 seconds.
  • Let your OTA cool down or tube currents could produce asymmetrical results due to tube currents.
  • To verify that your camera does not have a tilted sensor, take 4 shots rotating the camera 90°. If the camera has a tiled focus plane, the direction of the tilt will move with the camera.
  • Use always the same camera to have results that can be compared possibly with a full frame sensor. In all these tests a Canon 6D has been used.



Once you have captured enough shots, set the right parameter in order to produce consistent results that can be compared with other telescopes.

  1. First, enter the correct image scale in "Setting>Default Image Properties..."The saturation level cuts all the stars that are overexposed, the default value is 45,000 ADU.
  2. Drag and drop all the images on the main window of the software, select them all and click on "Measure All" 
  3. This first analysis shows some useful measures such as the average FWHM. Discard any frame with an abnormal value of the FWHM that could be the result of a poor guiding or a blurry image. 
  4. Select all the frames previously selected and click on "curvature" to have an average value of the telescope aberration across the field. 


SharpStar 61EDPH + corrector & reducer SharpStar 76EDPH + coreector & reducer
SharpStar 100 Q-II SharpStar 121 SDQ
SharpStar 150 HNT
Skywatcher Quattro 250 f/4 with coma corrector Skywatcher Quattro 250 f/4 without coma corrector



CCD inspector will show a map with a colour gradient that represents the field pr"curvature". The software will scale the colour range accordingly so any amount of curvature will be shown in every image, even if its value is small. In order to have some consistent result that can be easily and visually compared to one another, it is suggested to uniform the colour range.

Click left on the curvature image > Display Range > Minimum=2.0 ; Range 1.0

On the top left corner of the curvature image, there are some important numerical numbers that fully describe the performance of the tested telescope across the entire focal plane:

  • Min FWHM: this is the minimum size of the FWHM that is generally in the center of the field. 
  • Max FWHM: same as above but in this case the largest values represent the elongated stars on the border of the field.
  • Curvature: this is the most important value that describes the maximum increase of the FWHM across the field. However, it does not measure only the field curvature but all the aberrations. We may link the concept of barrel or pincushion field distortion with the word "curvature" but this is incorrect. For example, Newton telescopes are not affected by field curvature but the stars are elongated due to coma and astigmatism; CCD inspector will reveal a "curvature" that actually refers to these two aberrations. 
  • Tilt in XTilt in Y: self explanatory.
  • Total Tilt: it measures how much the size of the stars increase asymmetrically across the field. It is not related to the curvature as a telescope could have a large curvature but no tilt as the aberrations of the telescope increase symmetrical from the center of the field to the borders. Every optical configuration will have a bit of tilt that can be caused by a tiled optical element, a thumb screw that creates a minor rotation of the camera but also poorly machined adapters. The value of the tilt should be seen in contest and a value up to 20% neglectable. The tilt percentage shown refers to the minimum size of the FWHM that is generally around 3". A 20% of a 3" FWHM represent an elongation of only 0.60" so it is neglectable. 
  • Collimation: it measures the distance of the optical axis to the center of the center of the image. Even a few arcseconds mis-collimation can affect the image quality.


Considerations about the tilt

Every optical train (objective, corrector, adapters, sensor) will have some amount of tilt caused by one or multiple elements of the train. Some cameras have a slightly tilted sensor in their housing and this could be added to the tilt of the telescope. It is not easy to discern between the tilt of the sensor and the tilt of the telescope. Also, some telescopes are more sensitive to tilted sensors than others such as telescopes with a short focal ratio and especially Newtons. That makes it impossible to calculate a specific value of the tilt of the camera and offset it for all telescopes.

To understand what is introducing more tilt between the camera and the telescope, it is possible to conduct a test by taking a set of shots at 4 different angles, rotating the camera in the focuser. If the direction of the tilt remains almost constant, it means that it rotates with the camera so the sensor is tilted and it will mostly contribute to the overall tilt of the optical train.

 CCD inspector

These four curvature images have been obtained capturing a series of shots taken with the same telescope (SharpStar 150 HNT f/2.8) rotating the camera (Canon 6D) of 90° for every picture. The tilt value varies between 11% and 27% and the direction remains almost the same (150°). That means that the tilt introduced by the camera is higher than the tilt of the telescope. It is noticeable how curvature remains constant while the tilt varies.


 CCD Inspector tilt

 The telescope tested is the SharpStar 76EDPH with corrector and reducer combined with the Canon 6D. The tilt value varies between 36% and 15% and once again the direction of the tilt is almost constant (~60°). This is a confirmation that the tilt is mostly introduced by the camera.


If the tilt is only caused by the camera, it will remain constant. However, the overall tilt is a combination of more elements tilted. There will be a specific angle of the camera where the tilt of the sensor will match the inclination of the tilt introduced by the telescope, causing the overall tilt to be minimized.

As a solution, the tilt can be corrected by using a tilter that nowadays many cameras incorporate. 

Tilt on ASI cameras


CCD inspector is a professional software that can give you a full picture of the extra axial aberration of your telescope. It is an essential software to understand how to correct optical misalignment, focal plane tilts and collimation. Also, CCD Inspector dictates the maximum size of the sensor for the telescope used so the image will be fully corrected across the entire field. Do not pay too much attention to the tilt as it can be corrected while the curvature is caused by the optical design that cannot be changed. However, if the optical train uses a field or coma corrector, the curvature will increase if it has not been placed at the correct distance to the sensor. 


10 MICRON GM1000 HPS $16,499.00
10 MICRON GM2000 HPS II COMBI $22,999.00
10 MICRON GM3000 HPS $27,999.00
10 MICRON GM4000 HPS II $43,299.00
10 MICRON AZ1000 HPS $15,999.00
10 MICRON AZ2000 HPS $22,999.00
10 MICRON AZ3000 HPS $27,999.00
10 MICRON AZ4000 HPS $42,499.00
10 MICRON AZ5000 DDS $999,999.00

More Posts

Next Post

1 comment

  • Gion: December 21, 2020

    I have myself gone through rigorous collimation exercises with my Sharpstar 15028 HNT until I found out that at f2.8, the slight tilt introduced with the imaging train can cause sever offset of the secondary axis and in conjunction with the corrector lead not only to focus plane issues and shadows but also aberrations. Only after using CCDInspector, I was able to determine my overall system tilt and correct using a tilt-shift unit. After all my journey, this article sums up my experience and I really love how well you separated the different problems. Really great post!

    One thing I found a bit confusing at first, also from the description on the ZWO website about tilt analysis, is that the tilt moves along with the camera. But if you think about it, in relation to the image plate, the farthest portion of the tilted sensor will always remain in the same location no matter how you rotate the camera, so that makes a lot of sense. Also, it very much depends on where you are able to rotate the camera. If you can only rotate at the focuser, for example, your whole imaging train is considered the camera tilt as written in your article. This is something we need to take into respect. If you are able to rotate the camera only, then the static portion (in respect to the image dimensions) is only the camera sensor tilt.

Leave a comment

All blog comments are checked prior to publishing

Welcome Newcomer