chipsizes.jpg [ 27.87 KiB | Viewed 3373 times ]
Altair 183M on Optiphot.jpg [ 478.1 KiB | Viewed 3373 times ]
I have no commercial connection with them, but strongly recommend that anyone looking for a
microscope camera should investigate the 183M and 183C monochrome and colour cameras sold by
Altair Astro. For high sensitivity and low noise the cameras sold to well-funded science labs are now usually backthinned CMOS type. With their own software, these 'scientific CMOS' cameras cost many thousands of dollars. I have discovered that the astronomical community consists of a large number of
so-called amateurs who do superb work but have modest means, and Altair supplies them with cheaper but very high quality backthinned CMOS cameras with free software, all of which are USB connected.
I mention the 183M and 183C because they are 20 Mp CS mount cameras which need only a 5mm spacer ring ( which Altair supply) to fit on a standard microscope c-mount. They might be a little more expensive than a used DSLR but they are vibration-free ( even with fan cooling) and the software is very powerful, allowing live dust subtraction, use of an out-of-focus image to correct image shading caused by DIC optics or incorrect illumination. And unlike some DSLRs they generate live video without any distracting data display. It is remarkable how many imaging problems that bother us in microscopy have already been solved in the astronomy field. For example, the problem of cameras being primarily silicon detectors and giving poor images with the unfocussed 800 nm infra-red light which is not extinguished properly by polaroid filters is well known to the astro guys and Altair supply a range of IR filters specially for this problem.
I am attaching a couple of images. One shows my Altair mono camera on a Nikon Optiphot B, which has a trinoc that rotates, allowing total removal of the prism from the camera path: ideal for photography. The Nikon has DIC optics. The other is a diagram I drew for teaching, which shows the information content in pixels (assuming Nyquist sampling without information loss) in a typical microscope eyepiece field ( the circle) compared with various camera chip sizes. It is clear that 20 Mp is really necessary if you do not want to end up sampling only a tiny postage stamp in the middle of your microscope field, or else losing most of the detail.
I spent many years as a biological researcher trying to collect information with old video-standard cameras ( shown by the tiny green rectangle in the diagram). 3 to 5 Mp is certainly better and now available cheaply in many USB cameras, but I use it only for video. For public demonstrations, these 20Mp cameras with live video that can be zoomed and panned by software to show regions of interest has been a terrific boon to me.