A Large, Fast Astrograph: Planewave CDK14 With a Custom Imaging Train

The author’s Planewave CDK14 configured as an f/4.76 system with off-axis guiding. Credit: Rouzbeh Bidshahri

I have been very pleased with the performance of my Planewave CDK14 for imaging with its long focal length (f/7.2). See my review for AGT here. But given its very large 70mm circle, I felt it was a waste to not record all the photons that had travelled lightyears, ready to be recorded. A large sensor camera like the QHY411 would work, but with the required filters and accessories it would cost well over $60,000 – that was not an option!

The solution was to use a dedicated focal reducer with an affordable full frame camera, the QHY600M (reviewed for AGT here). With so many large targets during nebula season, a faster system would be better suited for narrowband imaging of these dim objects.

It took over a year of design and testing with three different CDK telescopes, but after several iterations my configuration now allows the Planewave CDK14 to operate as a widefield astrograph, using a full frame camera, filter wheel, and off axis guider. A similar configuration also worked well with my CDK12.5 using an APS-C sensor. The OAG, in particular, was an unknown as it had been suggested that this configuration wouldn’t work with the reducer in place.


The goal was to design a system that could operate as a large fast astrograph (a telescope designed specifically for imaging) that could also deliver high resolution images.

In its “fast” configuration, the system is now producing excellent results with the following features:

  • Fast focal ratio: f/4.76
  • Medium field of view: approximately 1.5 degrees
  • High resolution imaging with image scale of 0.46 arcseconds/pixel
  • Mono full frame camera with filter wheel support
  • Off-axis guiding
  • Precision focus and instrument rotation 
  • Can be converted back to native f/7.2 in a few minutes in the field

The Focal Reducer

Planewave offers a large focal reducer for this telescope; I reviewed it for AGT here.

The 0.66x reduction factor yields and effective f/ratio of f/4.76, which is relatively fast without being so extreme that it makes collimation, tilt, and focus overly sensitive.

The reduced f/ratio means the image is about twice as bright with a higher signal to noise ratio, allowing shorter sub exposures. While this can be synthesized digitally with binning, there are other advantages when done optically, as we will see.

This first test image was produced from only six hours of LRGB data from a Bortle 6 backyard (the pink H-alpha regions were later enhanced with H-alpha data).

First light image of M81 and M82, produced with only six hours of LRGB data. Note the faint IFN (Integrated Flux Nebula) wisps in the background. Credit: Rouzbeh Bidshahri

Field of View

The focal reducer provides a larger field of view. Using a relatively large full frame camera at the CDK’s native 7.2 f/ratio, the field of view (FOV) was less than 1 degree (diagonally). With the focal reducer, it is now about 1.5 degrees, more than doubling the area covered. This field of view is almost identical to the using the QHY411 mentioned above with the native focal length, but at far less cost.

Test image of M106 illustrating the FOV at CDK14’s native f/7.2 (top), and in the fast f/4.76 configuration (bottom). Note the vast increase in FOV! Credit: Rouzbeh Bidshahri

Resolution Loss?

Why use a long focal length telescope if you want to reduce its focal length?

It boils down to image scale, that is, how much sky each pixel will cover. With the small modern CMOS pixel, the native f/7.2 image scale of 0.3 arcseconds/pixel is too fine. To avoid oversampling, most imagers bin the pixels 2x to achieve a more reasonable scale of 0.6 arcseconds/pixel.

With the reducer, the image scale is high resolution at 0.46 arcseconds/pixel, which is still more resolution than most locations on Earth allow due to atmospheric blurring, regardless of the telescope size.

Small telescopes can offer a large FOV, but they lack resolution due to the small aperture. Alternatively, long focal length telescopes, like an SCT, can offer a high image scale (though not necessarily high resolution) but they lack the large FOV.  The CDK “Astrograph” achieves a high image scale, true high resolution thanks to its large aperture, and a relatively large FOV. All these render a large 62-million-pixel image with fine details that can be printed as a large poster, if desired.

A close-up of the same M82 galaxy first light image demonstrating both the large FOV and high resolution. Credit: Rouzbeh Bidshahri

Off Axis Guiding Capability

Guiding becomes more critical when imaging at high resolution with narrowband filters. Even with a low noise camera like the QHY600M, the sub exposures need to be 600 seconds long with 3nm filters.

While some very high-end mounts can achieve excellent tracking, most will not be able to guarantee perfect tracking at these high image scales for such long periods.

In its standard setup, the CDK14 and its reducer have 44mm (1.72 inches) of back focus (from the reducer flange to camera sensor). That’s not enough space for a monochrome camera, filter wheel, and an off-axis guider (OAG).

The dimensions of the assembly have tight tolerances as the reducer, off axis guider and sensor need to be precisely positioned. Here, the assembly is being tested for being perfectly square (orthogonal) while rotating. Credit: Rouzbeh Bidshahri

However, the custom imaging train I built for the CDK14 was designed to accommodate an OAG. This configuration uses the low-profile Optec Gemini rotating focuser (reviewed for AGT here) and custom adapters that allow a large OAG to be placed in front of the focal reducer, as shown below.

Schematic of the modified CDK14 imaging train. Credit: Rouzbeh Bidshahri

Tests on multiple targets have revealed excellent off axis stars and guiding without blocking any of the incoming light to the main imaging sensor.

With multiple stars detected by the guide camera and very good autoguiding performance, the total guiding error is a very low 0.29 arcseconds. This is typical on nights with good seeing. Credit: Rouzbeh Bidshahri

Image Quality

Imagers usually prefer to avoid extra elements like focal reducers that may degrade the quality of the image.

Since there were no vignetting or spot diagrams available for the reducer, I conducted my own tests to determine if there was any detectable image degradation in this configuration.

Star size (FWHM): A typical single exposure, shown below, shows a measured median star size (FWHM) of 4 pixels. With the scale being 0.46 pixels/arcsecond, this means a typical star size is 1.84 arcseconds, just as sharp as the native configuration. The system is limited by seeing blur rather than a lack of image scale. 

There isn’t much increase as we move off axis, with the corners approaching 4.4 pixels, a 10% increase which is acceptable.

A representation of the camera sensor with the sizes of stars (in pixels) across the sensor. Stars in the center are 3.9 pixels across, while the corners are only slightly larger at 4.2 to 4.4. Credit: Rouzbeh Bidshahri

Corner Stars: A visual inspection of a single exposure shows the stars in the corners are still reasonably round. The last few millimeters in the extreme corners exhibit slightly deformed stars when viewed at a 1:1 scale. The stars are reproduced very well over the remaining field of view.

Mosaic showing a closeup of the stars in the center of the frame and in the corners. This is a single sub. Each square is 10% of the total image size. Credit: Rouzbeh Bidshahri

Vignetting: A flatfield analysis plot below shows the extreme corners (black) have a 40% drop in illumination. This is fully corrected by flatfield calibration and is not evident in the final calibrated images.

Illumination of the sensor plotted from maximum brightness (white) to minimum brightness (black). The OAG prism is not blocking any of the incoming light. Credit: Rouzbeh Bidshahri

Field curvature: Testing the field curvature with CCD Inspector shows the curvature at the edges to be 25%, which is a great result that’s only 15% more than the native configuration. Looking at the visual representation below, we can see that the overall field is flat.

Simulation of the curvature of the field using CCD Inspector to analyze imaged stars. Top: The CDK14 in “Astrograph” mode. Bottom: The author’s triplet APO with flattener/reducer and the same camera. The CDK14 with the custom imaging train is much flatter. Credit: Rouzbeh Bidshahri

Dual Focal Length 

This custom configuration can be switched back to the native focal length configuration in minutes.

The procedure involves loosening three set screws, releasing the reducer from the OAG, and replacing the reducer with a preassembled part that maintains the focus position. This assembly is made with off-the-shelf extensions and adapters with a dovetail connection that simply plugs into the OAG with the camera on the other end. The OAG settings don’t need to be altered, and the guider doesn’t need to be recalibrated.

The system can be restored to its native f/7.2 mode by replacing the reducer with this single assembly. Credit: Rouzbeh Bidshahri

Final Thoughts

It has been a long journey, but I am now very satisfied with the design and performance of the CDK14 in the current “Astrograph” mode. The integration of an OAG has made guiding robust, allowing for very tight stars and sharp images.

The fast f/4.76 ratio is very pleasant to image with, and the large FOV allows me to capture many more targets from nebulae to large galaxies.

Gazing at the resulting large, high-resolution images at full scale really gives one a sense of the vastness of the cosmos. While it’s very easy to switch back to the native configuration, at this point I have no reason to!

More high-resolution examples can be seen in the author’s gallery:

A narrowband image captured with the customized CDK14 imaging train. Top: Image illustrating the FOV. Bottom: Closeup of the same image showing the full resolution. Credit: Rouzbeh Bidshahri


About Rouzbeh Bidshahri

Rouzbeh Bidshahri is a mechanical engineer with a lifelong passion for astrophotography. He has tested dozens of telescopes ranging from 3 to 20 inches in aperture and has spent several years optimizing systems for very high-resolution planetary imaging in the sub 0.1 arcsecond/pixel range. He has contributed to several institutions such as ALPO (The Association of Lunar and Planetary Observers). His main area of interest has been designing and operating larger setups, and he is currently focusing on high resolution, long exposure photography for both broadband and narrowband deep sky imaging.

Related posts