Calculate CCD FOV and resolution

[rikyuu]

hi,

I've been reading about CCD astronomy in a couple of books, one of them "A practical guide to CCD astronomy" gives a very useful formula to calculate the resolution and FOV for a CCD (nb. for 'ARCTAN' use 'TAN to the minus one' on a calculator:

Resolution = ARCTAN (pixel size / focal length) ..... both in meters

.....this will give you the angle of sky that one pixel is seeing

FOV = ARCTAN (CCD size / focal length) ..... both in meters

The book suggests a good resolution of 1-4 arcsec although this was a good few years ago.

My Mintron has a pixel size of 8.6um x 8.3um and the scope a focal length of 1.2m. Hence the resolution is 1.48 arcsec

The CCD size is 7.4mm x 5.95mm (1/2 sensor) hence the FOV is 0.35 degrees. With a barlow I get about 0.175deg (10arcmin) FOV and 0.74arcsec resolution, which is good for M57 which is 1arcmin across. With the focal reducer, I get about 0.7deg FOV. So not a bad range.

I can imagine with a smaller CCD and long focal length SCT, the FOV would be quite small eg. about 0.17deg for a 1/3 chip on a 2m focal length.

Hope this is useful to someone.

Cheers

[SteveL]

An even easier way is to feed your OTA and camera into CCDCalc (Ron Wodawski)

http://www.newastro.com/book_new/camera_app.php

[rikyuu]

An even easier way is to feed your OTA and camera into CCDCalc (Ron Wodawski)

http://www.newastro.com/book_new/camera_app.php

just gave CCD calc a try, it almost gets it, but doesn't take into account the gap between the pixels (as it's an interline CCD) so calculates the CCD size slightly too small. It's usefull none the less.

Cheers

[mcrossley]

That looks about right, but why does the inter pixel gap matter? It is the pixel pitch that will determine the image scale and resolution. Putting in 8.6x8.3 as the pixel pitch and 752x582 as the resolution, gives a resolution of 1.48aspp horizontally and 1.43aspp vertically. CCD Calc displays these two values as mean resolution of 1.45aspp, the FOV calcs are correct though at 18.5 x 13.8 arcmins, and an effective chip size of 6.5 x 4.8mm.

Mark

[rikyuu]

I understand the FOV depends on the size of the CCD rather than the number of pixels. CCD calc seems to calculate the size automatically, based on the pixel size x number of pixels so it works out a bit smaller if the gap's not taken into account and hence a smaller FOV.

[SteveL]

Not sure thats strictly true.... my QHY8 Hyperstar images that I send to Astrometry come back with a calculated FOV that matches CCDCalc exactly.

[mcrossley]

I understand the FOV depends on the size of the CCD rather than the number of pixels. CCD calc seems to calculate the size automatically, based on the pixel size x number of pixels so it works out a bit smaller if the gap's not taken into account and hence a smaller FOV.

No, the pixel size quoted by the chip manufacturers is not really the size of the pixels, but the pitch between the centres of the pixels, there is a lot of silicon around each photo sensitive site, many chips have a micro lens array on top of the chip to minimise the 'dead' areas (and this causes its own problems). You will find CCD calc is performing the calcs correctly.

The quoted pixel area of the CCD is not the same as the image area, there are many (parts) of the rows and columns that are not used in the final image.

Mark

[rikyuu]

hmm, yes, but when I enter my pixel size and number of pixels, the calculated size is a lot smaller that that which Sony states.

[mcrossley]

What Sony quote is a "chip size" of 7.4 x 5.95mm, it does not state that this is the size of the optical area. I suspect this is the size of the silicon.

They do quote a optical diagonal of 8mm, so if you take SQRT( (6.4*6.4) + (4.8*4.8 ) ) = 8.0mm

[rikyuu]

ah, that might explain it, many thanks