For the first time since Dave Allmon's discovery in 1999, that the B/W Connectix webcam could be made to do long exposures, we are making new discoveries and innovations that enable other video cameras to see deep sky objects.
Indeed, the next breakthrough was made by Steve Chambers in August 2001 with the modification of a low cost CMOS based webcam, the Trust SpaceC@m, enabling it to take long exposures. This camera still has considerable potential, which is being explored.
Steve has also produced in September 2001, a working long exposure modification of the Phillips Vesta family of CCD based webcams
Other QCUIAG members are working at the theoretical level on long exposure modifications to video cameras and Jon Grove has produced in September 2001, a long exposure modification of a 0.02 lux surveillance video board camera.
In June 2002, Robert Piggott produced a long exposure hardware modification for the Logitech QuiickCam VC (Parallel port version)
Results are being accumulated, and further refinements to the modifications are already being discussed.
Software developments by and in association with QCUIAG members since the summer of 2000 mean that both traditional and new approaches can be made towards deep sky imaging with the advent of new long exposure camera modifications. It seems likely that approaches combining medium length exposures and off-chip frame integration may become the norm.
If a true long exposure is possible, then, deep sky objects can be imaged by even a single exposure. The length of the exposure will depend on the amount of dark noise that is accumulated during the exposure relative to the signal derived from the object being imaged. Environmental cooling of the interior of the camera, within the temperature operating parameters of the device may enable slightly longer exposures to be made by reducing the thermal noise. Peltier cooling of the light sensitive chip will greatly reduce the thermal noise and will enable much longer exposures, whilst requiring even more involved hardware modification. There is already some accumulated experience on this in relation to the B/W QuickCam.
The burden of this approach is that very good alignment and tracking are required in order to avoid star trailing in the aquired image.
This approach is only new in terms of the scale of the operation rather than in its basic concept. It has long been the practice of astro-imagers to sum or more likely, average a number of exposures. However, the number of exposures involved are typically small.
The development of software such as AstroVideo by COAA and Vega by Colin Bowness allows for off-chip integration of extremely large numbers of video frames. This allows the very faint signals in parts of each frame to be summed in incredibly deep 32 bit Fits files. The results have been astounding since the summer of 2000, with some deep sky objects being clearly revealed by the summing of thousands of video frames from cameras taking 1/50s exposures at most. If this philosophy is applied to video frames resulting from exposures of several seconds, the sensitivity of the cameras will increase hundreds of times in relation to the fainter parts of deep sky objects. This derives from the fact that 1/50s exposures of traditional video cameras are not long enough for the accumulation of sufficent photo electrons from very faint objects to reach critical quantites required to digitise the signal. This sets real limits to the faintness of objects that can be seen by video. Nomatter how many frames you sum, you will never detect an object that produces too few photoelectrons in 1/50s. A longer exposure of the video frame will rectify this.
Software such as Astrovideo will enable the accurate registration of images summed over huge numbers of captured frames, effectively synthesising very long exposures from many shorter (but still much longer than 1/50s) exposures, whilst eliminating star trails. This software can now support the long exposure modifications.
This is demonstrably true with the results from the first tests of Steve Chambers' modification of a CMOS based camera. This is an inherently insensitive device, and yet when given exposures of 4-5 seconds and when large numbers of these exposures are summed by AstroVideo, deep sky objects such as M13 and M57 are easily imaged. Based on these initial encouraging results, the effects of environmental and peltier cooling in conjunction with the summation of even larger number of frames, coupled with optimising the exposure time, will determine the usefulness of CMOS based devices for astronomical imaging.
Please read this WARNING.
Steve Chambers's modification of the CMOS based Trust Spacec@m 200 webcam (Steve Wainwright and Derek Francis)
Steve Chambers's modification of the Vesta 675 webcam (Steve Chambers)
Jon Grove's long exposure modification of low light Video Surveillance cameras (Jon Grove)
Jon Grove's website (Jon Grove)
Robin Leadbeater's website with Jon Grove modification (Robin Leadbeater)
Steve Chambers' website
Steve Chambers USB capture website
Dave Molyneaux website (Dave Molyneaux)
Robert Piggott's website for the Logitech QuickCam VC long exposure hardware modification