Celestron RASA 8 astrograph: fast imaging with the RASA 8

The Celestron RASA 8 is a dedicated astrograph.

Celestron’s RASA 8 is a pure astrograph, a photographic telescope that can only focus light onto a camera chip – there is no eyepiece. I’ve spent the past year imaging with the RASA 8 that belongs to Dr Paul Kummer, a retired computer physicist from the University of Manchester, who co-owns a set-up with me at our remote imaging observatory at Les Granges Astronomy Holidays in southern France. Throughout that time, I’ve come to know the RASA 8 very well.

At first glance, the RASA 8 resembles a Schmidt– Cassegrain telescope, but the primary mirror sends light not to a secondary mirror, but through a lens and onto the front-mounted camera. The instrument’s headline feature is its phenomenally fast photographic speed of f/2, which is where we shall begin.


Looking down the tube of the RASA 8, with a front-mounted ASI 2600MC camera, home-made dew-shield and semi-circular cable guide to avoid diffraction artefacts around stars.
Image: Olly Penrice.

Focal ratio can be extremely misleading, so let’s cut to the chase: the RASA 8’s aperture is 203mm but, since the camera sits in the front of the optical path, it has a considerable central obstruction leaving it with a true light-collecting area of 25,572 square millimetres. By comparison, a typical 85mm fast refractor of f/4.7 would have the same focal length, but a light collecting area of just 5,674 square millimetres. Quite simply, the RASA collects 4.5 times as much light as a generic fast refractor. Three hours in the RASA really does equate to 13.5 hours in the refractor.

Given this remarkable saving in exposure time, we might expect all discussion to end here, with the RASA 8 as the only short-focal-length imaging telescope anyone would ever want. However, its speed brings complications.

Finding focus

The light cone of an f/2 system is steeply angled, giving a very shallow depth of field. Because astronomical targets are at infinity, astrophotographers are not normally concerned about depth of field, but very fast f-ratios require disconcertingly precise focus, well-collimated optics and an imaging train free from tilt. A tilt between the chip and the incoming beam of light means that only a part of the image can be in focus.

The RASA 8 ready for action on an Avalon Linear Mount.

Celestron claim that the RASA should arrive in good collimation, but owners’ experiences suggest that it may or may not do so. Since it cannot hold an eyepiece, the only practicable way to collimate it is to use a small, cylindrical camera such as a guide camera, small enough to give access to the three pairs of antagonistic (push and pull) screws that adjust the front lens while you observe a defocused star on a computer screen.

A simple home-made jig to test a camera chip’s orthogonality. The beams here are Photoshop-generated to illustrate the principle.

We needed to collimate Paul’s example and the adjustments required were truly tiny. We settled, in the end, for a good but not strictly perfect collimation.

With the imaging camera in place, we turned to the issue of focusing, using Celestron’s dedicated focus motor. This was fairly straightforward to fit, but it is not precisely machined and so must be tightened up in small iterations to avoid stressing the focuser shaft. Be sure to follow the instructions carefully. Our first focus motor expired after a few outings but its warranty replacement has worked well.

First light revealed considerable tilt, giving us good star shapes in the centre but unacceptable elongation along the short sides of the image. Tilt often comes from the camera, so we constructed a test jig following principles helpfully laid down on the website of Starlight Xpress. Our simplified jig is shown in the photograph on the left. A lowpowered laser sends a reflection from the camera chip onto the base of the box. With a tilt-free chip the reflection will form a dot that turns on the spot as the camera is rotated.

Surprisingly, the instrument holds focus very well indeed, usually needing a refocus only after a meridian flip. Life is also simplified by the fact that heat from the camera and ventilation from its fan mean that a short, unheated dewshield is all that’s needed to combat corrector-plate dew. We found the best way to route the camera cables in front of the corrector was to shape a semi-circular guide. We were briefly perplexed by a light gradient in our images, but this was traced to a tiny LED on the camera lighting up the bottom of the cable guide.

After all these adjustments, we had acceptable but not perfect star shapes at the edges of the chip. Stacking a large number of sub-exposures improves them and, to be fair, Paul’s ASI 2600MC Pro camera has a diagonal a few millimetres larger than Celestron’s claimed circle. Imagers who like to ‘pixel peep’ might not like what they see, but I was happy with what we had.

Your choice of camera is limited to models that will fit within the central obstruction, so DSLRs and filter-wheels are ruled out. We found that ZWO’s ASI 2600 MC is a good match for the RASA 8 and the camera can be used with a multiple bandpass narrowband filter very effectively. The small pixels of modern CMOS cameras are well-suited to the 400mm focal length.


Despite a consciously gentle approach to processing here, the RASA 8 had no difficulty finding the ghostly outer extensions of an old favourite, M31. This four-pane mosaic, with roughly 80 three-minute subs per pane, was captured and pre-processed by Paul Kemmer, with postprocessing by Olly Penrice.

Rekindling a passion

At last we can now turn to the fully fettled telescope’s performance under the stars. Let there be no mistake: it is exceptional. Just sixty exposures of three minutes can give an exciting and polished result, even on faint targets like dark nebulae. Bump that up to ninety subexposures and the image processor is in heaven. The data are a joy to work with, deep and clean. They do, however, benefit enormously from processing with the software Star Xterminator, (the updated software of which is given a trial run in this this issue in Nik Szymanek’s Imaging masterclass on page 76). By processing the nebulous detail separately from the stars, the RASA 8’s depth of capture can be fully exploited and the stars kept under control when removed. In my view it would be a shame to process RASA data without using Star Xterminator or, perhaps, its rival, Starnet++.

The Shark and Anglerfish nebulae cruise above other dusty forms in Cepheus. This 15-pane mosaic has between 60 and 80 three-minute subs per pane, captured and pre-processed by Paul Kemmer, with postprocessing by Olly Penrice.

When I say that the RASA has rekindled my passion for wide-field imaging and mosaic-making, bear in mind that I live at a dark site with hundreds of clear nights each year. If I were fighting cloudy weather and light pollution, I would be even more persuaded by this telescope. It has another advantage, too: it works best when used to take large numbers of short exposures, meaning that the sigma-stacking routines will eliminate satellite and aircraft trails more effectively. One last last bonus: it seems that ‘dust bunnies’ – the shadows cast on the chip by dust particles close to it – are blurred out of existence in the RASA 8. Because they are only correcting vignetting, the imager’s flat fields should work for a very long time without being updated.

At less than £2,400, the RASA 8 is remarkably good value. It may not be perfect, it may not suit everyone, but it has most certainly won me over. I find it a constant pleasure to use.

At a glance

Type: Rowe–Ackermann Schmidt astrograph (RASA)

Aperture: 203mm (8 inches)

Focal length: 400mm

Focal ratio: f/2

Central obstruction: 93mm (46% of aperture diameter)

Resolution: 0.57 arcseconds

Image circle: 22mm

Weight of OTA: 7.7kg

Details: celestron.com

Price: £2,345 (OTA only)

Olly Penrice is an astrophotographer who runs Les Granges Astronomy Holidays, which you can find more about at sunstarfrance.com.