It's amazing what can be achieved by pointing an ordinary web-cam  at the heavens ! - provided you use the right software. I've recently started playing with this exciting technique and am on the steep part of the learning curve. Here's what I have learnt so far.

WEBCAM

• Webcams come in two flavours - CCD cameras and CMOS cameras. Of the two, CCD is the more light-sensitive, CMOS cameras are not thought to be very good for astro work because of their generally lower sensitivity; however this is changing because the main development is on CMOS chips. It's difficult, without a close examination of the specification, to discover which a particular camera is ( see this table for example). Some people are now finding digital cameras very useful for the job, providing they have the facility for some sort of time exposure. 

• I have a Philips Vesta-Pro PCVC680K camera. This is a colour USB device producing 640 x 480 sized images. It employs a Sony  ICX098AK CCD sensor with a claimed sensitivity of <1 lux. The chip size is 4.6 x 4.0 mm and has an array of 659 x 494 pixels.  Straight 'out the box' it isn't suitable for astro work because it has a very wide angle, short focal length lens. This needs to be unscrewed and a means of attaching the camera to a telescope or suitable lens devised. Without its lens, the camera CCD is very sensitive to Infra-red light so an Infra Red filter is built in the front of the lens to reduce sensitivity in that region of the spectrum.

Here's the lens assembly with the red rubber ring removed. The I.R. filter (seen in the centre) is held in by the black plastic ring on the front of the lens housing and was removed by carefully removing the ring (it's welded on ). The garden shot is two combined - the right-hand side is a normal view through the lens with the I.R. filter in place and the left-hand was taken with the filter removed. The effect is dramatic in normal scenes because the chlorophyl in plants strongly reflects I.R. and hence distorts the colours. For astronomy use, you may need to fit an I.R. stop filter somewhere in the optical path. This is more likely to be needed with refractors because the lens will not focus I.R. at the same point as visible light, leading to fuzzy CCD images. Reflectors are less of a problem because mirrors focus all wavelengths at the same point.

• One feature of CCDs you need to be aware of is their very small field of view. This picture shows a direct comparison between the sensitive area of the  chip on my Vesta camera and a normal 35mm film. Aiming at a required target is not easy. Cleanliness of the chip's optical surface is critical - any tiny dust motes on this will show up on the images, looking like craters. It's wise to check before any imaging session starts (easy for me to say, I always forget).

CAMERA MODIFICATIONS                                                                                                                                              

There are at least two approaches to modifying the camera . A third method, which involves modifying the electronics (not for the fainthearted) has recently (28/9/01) been developed to allow long exposures with the Vesta series of camera. This exciting new development has been pioneered by Steve Chambers. Details can be found on his web site at this address and on the QCUIAG site. This modification will allow imaging of deep-sky objects previously only obtained by expensive cameras. More details of my modification are to be found on this page.

• Mod 1 - Keep the hardware in the original pear-shaped housing and make or buy an adaptor which screws into the existing thread. The adaptor has a conventional 1.25 in barrel to fit inside a telescope focuser tube. The camera is still available for conventional use. Some have modified the end of a 35mm film container and glued it to the camera. These containers happen to fit telescopes. This way, the camera cannot be used with its normal lens. If you don't want the lens, it can be dis-assembled and the threaded barrel used to make an adaptor.

• Mod 2 - Remove the circuit boards from the housing and build a specialised box with the appropriate adaptor on the front. This was my approach for my MK1 camera.

      Here is the camera after it had been opened up. It's necessary to  carefully lever the two halves apart with a thin screwdriver blade; taking great care because the circuit boards locate in slots inside. The boards will then pull out together with the cast metal lens housing. This is attached to the CCD board by two tiny bolts which need to be unscrewed. The CCD board plugs into the main board as does the cable. This site has excellent pix of the process.

Here's an general view of the boards as removed from the housing, and here's a view from the front showing the CCD and mounting holes. Note - I've unsoldered and removed the microphone and 'on' LED.

THE MK1 CAMERA                                                                                                                                                            

•    As a first attempt at getting the Vesta into a useful state for work as an astro-camera, a new box was built with an old M42 camera lens  mount bolted to the front. The M42 thread is found on old Pentax, Praktica, and Zenit cameras  and others of similar age. I removed the plate off the front of a defunct Zenit. It was held on by four screws. An alternative lens mount may be made by obtaining a rear M42 screw-on lens cover and cutting a suitable hole in in. This may then be glued/screwed to the camera body.

Here is the Mk1 camera without its cover. It's very crude being made from wood.  I found that I had to move  the circuit boards a lot nearer to the front than shown in the pic above to allow it to focus. This system allows me to screw on a range of 35mm camera lenses for direct imaging (I've used up to 800mm focal length) and, by means of a commercially available T-adaptor, to mount it on the telescope focuser.

This pic shows the camera on its equatorial mount together with the motor drives. The blue object at the front is a finder scope modified from my Cassegrain scope. Most people experience two difficulties with webcam ccd work - placing the desired object in the very narrow field of view and focusing the image. It seems to me that much frustrating trial and error work is required. An accurately aligned finder scope is a must. It's best to set it up in daylight on a distant target and note or mark the lens for the best focus at the same time. Very small adjustments of the focus will obviously affect the quality of the final image so it's best to take a few shots, adjust focus and repeat... I still have a lot of work to do on these aspects yet.

THE MKII CAMERA                                                                                                                                                            

• Two shortcomings soon became evident with the MKI camera - (a) it was very difficult to align on a target because the finderscope was not accurate enough and (b) the CCD chip was not centered well enough in the field of view. The MKII attempts to correct these problems.

    The picture show a view of the camera made on my flatbed scanner (hence the poor quality) and a drawing (click for larger version) of the assembly. It's built on the back of an old Zenit SLR. The CCD board is bolted onto a piece of PC board with a hole cut in it just large enough to accommodate the CCD chip. The control circuit board sticks out the back of the camera through a rectangular hole cut in the back of the camera. The assembly is held in place by the camera back.  I built a 45 deg. mirror to enable me to look down through the viewfinder. I can now look via the camera at the target object, centre it in the field, open the shutter and - there is the object on the computer monitor. The fact that the camera viewfinder has a larger field of view than the CCD is an advantage in finding the target.

   This camera can be used off-telescope with conventional lenses for wide-field work. Its real problem is the view through the finder of the camera screen is very dim.

This is the T-adapter screwed on to the eyepiece adapter ready to go between the 'scope and the 35mm camera.

THE MKIII CAMERA

   This incorporates a Vesta 675 board modified for long exposures, described here. in more detail.

  For one of the cameras, I've built a crude flip mirror which accepts a standard 0.965 " eye-piece in the top, and the Vesta camera, still in its original case at the end. This screws into a threaded tube made from a dis-assembled Vesta lens assembly glued into the end of the mirror box  . This is really the same approach as the camera above, without the extra weight of the 35mm camera body. I've seen a similar mod where the mirror assembly has been removed from a camera and incorporated into a new box. 

  The housing for the mirror was made from double-sided  plain printed circuit board, soldered up and the mirror is plastic, glued to a threaded rod and flipped up by a small lever which clips into detents in the two positions. OK, so the mirror is not top-quality front silvered stuff, but adequate for locating targets in the sky !. It's important when making such a unit to keep the distance between the CCD and the front surface as small as practical (hence the 0.965 eyepiece) otherwise the telescope focuser will not be able to accommodate the extra distance.

THE MK1V CAMERA

The above camera had a limitation in that it couldn't be used as well with normal camera lenses, so, time for a re-design again. What was needed was a mount which would take 35mm camera lenses and fit onto my telescope with a T- adaptor. Rather than severely alter the MK111 box, I built a new one; not with a flip-mirror, but with a fixed 45 deg. mirror with a small hole in it to enable the CCD to always see the target. Now the stars can be seen in the mirror by means of an eye-piece and the object selected can be dropped into the hole where it appears on the CCD. These pix will explain:

Seen in the side-view is the Vesta board at the rear, a vertical panel on which the CCD board is mounted, the 45 deg. mirror and the home-made 80mm eye-piece. In the front view, the M42 lens screw-mount is seen, behind it is the mirror with its 6 x 6mm hole and just visible through it is the Vesta CCD chip.

   The body is made of soldered-up printed circuit board (see these pix). How did I cut the hole in the mirror ? - ah, it's made of plastic, so it was an easy job. Plastic, high-quality mirrors are available, possibly  from car accessories. Care has to be taken to get the distances from the lens-mount to the CCD and the eye-piece correct ; more critical when using camera lenses.

  This system produces a good wide-field bright view.

WHERE TO PUT THE CAMERA

• All my work so far has been carried out by simply mounting the web-camera, without its lens or an eye-piece, on the 'scope at the eye-piece draw-tube. This is known as prime-focus imaging and is the simplest method. A disadvantage is that magnification is fixed by the focal length of the primary and is generally not large enough for planetary imaging. The various ways of mounting are shown in this diagram below (click for larger version). Not shown is increasing image size by the use of a Barlow lens or decreasing it by using a focal reducer. 

CABLE LENGTH WITH USB

• USB cables cannot be of any length, one reason is that they power the USB device down the cable and some gadgets can require up to 500 mA of current. The cable supplied with the camera is 2 metres long (or short, depending on your viewpoint). Extension cables can be bought which have an active repeater at the far end to make up the losses. These are available in up to 5 metre lengths and it is claimed that up to five of these can be 'daisy-chained' (I use two with no problems). There is no point in mentioning suppliers, circumstances change too quickly but here is a pic of one as an example.

      The plug end fits in the computer, the repeater socket box mates with the plug on the camera lead.

IMAGE PROCESSING

• What makes Webcams capable of astronomical use ? it isn't their light sensitivity ; they're not particularly sensitive. no, it's the techniques that have been developed by amateur astronomers to make the most of the images obtained.  The general pattern is to obtain as many images as possible, either as a AVI movie or many Bitmaps or Fits, align and stack them and work on them with image enhancement software to tease out detail not visible on the uncorrected images.  The pix below illustrate what can be achieved. The left ones are frames straight from the web-cam; the right are the result of stacking and processing 20 such frames. They are, perhaps, slightly over-processed and also note that the processing has emphasised the dust motes on the CCD surface - always a problem.

Before and after

The registration and stacking aren't carried out manually - there are specialist programs for that (see below). Note, the programs won't put back detail which isn't somewhere in the images. This raises a question - which image is more 'valid' ? the unprocessed one as seen by the CCD or the image after processing ?.

I haven't mentioned 'dark frames' - blank frames with the 'lens-hood' on taken during the observing session and used to subtract any hot pixels from the images, or a 'flat-field' frame  used to correct for differences in sensitivity across the CCD.

For a series of excellent tutorials on how to connect and use a web-cam with the telescope, I can do no better than direct you to this fine web-site by Jan Timmermans.

A list of links to the more useful image manipulation programs is on this page.

CLICK TO GO TO  THE VESTA S-C PAGE for the long-exposure modifications.

Member of  Quickcam and Unconventional Imaging Astronomy Group.