DSLR vs Webcam for Planetary Imaging

[cgarry]

I would like to try my hand at planetary/lunar imaging

with my NexStar 6SE. A friend has offered me his old DSLR, a Nikon D70S for this purpose. I have a T-adapter and a T-ring so I can connect it to the my scope.

My other option is to buy a webcam, possibly a Celestron Neximage.

So which is the best option? From reading the forum it seems that most people take the webcam route. Is that because it is a cheaper option, or because it produces better results?

Chris

[Breakintheclouds]

The webcam is cheaper and usually also produces better results. It's partly because the sensor in a webcam is much smaller, and this helps you get a bigger image of small objects like Jupiter compared to a DSLR, with its big sensor (which is more suited to wide images). However, it's mostly because by taking video rather than still images, you can go through the video later and find the frames where the air turbulence provided fleeting moments of stillness. All the best frames can be combined using software such as Registax to get a sharp image.

With a dSLR you take a single image and the chances are that at the exact instant you take the picture, the air will be turbulent. That's not to say you won't ever get good images with a dSLR, but you maximize your chances of getting good pictures with a webcam. Plus, it's a lot cheaper!

[brantuk]

Breakin makes some good points there. I would also add that you get a lot of light from planets so a rapid series of short exposures will be quite well defined. A webcam lends itself nicely to this.

Dso's on the other hand are very dim so you need a series of very long exposures to capture all the photons (typically 5 mins per exposure). Of course this is hard to do with a webcam unless it's purpose modified.

So it's really a question of "the appropriate tool for the right job".

If you go the webcam route then take a look at the Phillips SPC900NC (and the 880 also). These are the current favoured webcams and a lot cheaper than the Neximage (which has the same chip as the 900).

[beamish]

I'd read this before spending cash on a Neximager

[cgarry]

Thanks guys. I had assumed that I could take video with the DSLR, but looking at the D70S's manual I see that it does not support video, just 3 fps continuous shooting. So for various reasons it seems obvious the webcam route is the one to take for planetary imaging.

I had pretty much written off DSO imaging with the 6SE because of its long focal length and its alt-az mount.

[cgarry]

Thanks for the link beamish, just the info I need.

[cgarry]

I have ordered a Phillips SPC900NC, complete with a telescope adapter. Much cheaper than the Celestron Neximage.

[ollypenrice]

The webcam is cheaper and usually also produces better results. It's partly because the sensor in a webcam is much smaller, and this helps you get a bigger image of small objects like Jupiter compared to a DSLR, with its big sensor (which is more suited to wide images). However, it's mostly because by taking video rather than still images, you can go through the video later and find the frames where the air turbulence provided fleeting moments of stillness. All the best frames can be combined using software such as Registax to get a sharp image.

With a dSLR you take a single image and the chances are that at the exact instant you take the picture, the air will be turbulent. That's not to say you won't ever get good images with a dSLR, but you maximize your chances of getting good pictures with a webcam. Plus, it's a lot cheaper!

No, the image is surely no larger in a webcam than in a DSLR. The image scale is determined by the focal length. It is just that there will be a lot of 'unused' space around the object on the large DSLR chip.

The secret of video imaging is that you beat the vagaries of the seeing. You take maybe 1,500 images, select a reference image out of that lot, then ask the software, probably Registax, to find the 'best fit' equivalents from the video run and average them. When you average a very large number of rather poor images you get one rather good one!

Olly

[Breakintheclouds]

No, the image is surely no larger in a webcam than in a DSLR. The image scale is determined by the focal length. It is just that there will be a lot of 'unused' space around the object on the large DSLR chip.

I meant it in the sense that a given object will fill the frame more on a webcam than a dSLR. You're right that in terms of pixels it might be the same. But there's a strong (perhaps irrational) feeling that the webcam gives a 'bigger' object because of how the frame is filled.

[ollypenrice]

I meant it in the sense that a given object will fill the frame more on a webcam than a dSLR. You're right that in terms of pixels it might be the same. But there's a strong (perhaps irrational) feeling that the webcam gives a 'bigger' object because of how the frame is filled.

I see what you mean but in deep sky imaging we'd all like the 11000 chip at £££££££ simply because it is so darned BIG!

Olly

[bus_ter]

The sensor issue is mostly about pixel size. Pixels on a webcam sensor are very small, so the small image projected has a much higher resolution than on a dslr sensor. The larger pixels on a dslr are much more sensitive to light (because each larger pixel catches more photons). However the planets are plenty bright enough so that's not such a benefit.

Modern dslrs all record video too now. I've tried recording video with my D5000 and stacking, but it's no good because the resolution is too low. Also the dslr fps is less than a webcam, and when you put a dslr into video mode it's resolution decreases so the effective pixel sizes are even larger.

[Luke]

re: pixel size, a quick web search seems to say:

SPC900NC: 5.6 x 5.6 microns

450D: 5.15 x 5.15 microns

1000D: 5.68 x 5.68 microns

D70S: 7.8 x 7.8 microns

I'm surprised that my 450D's pixels are smaller than the webcam's pixels. That empty space really is deceptive!

[gthomas]

The pixel size is indeed very similar. The much smaller chip in the webcam I find makes it very tricky to position your planet but the results are ususally pretty good if you get the processing right.

However I've just tried BackYardEOS which is one of a couple of programs I've found that can use the DSLR liveview function to capture an avi even if the camera doesn't natively support video recording.

Result?

Webcam -> 640x480 image of about 900Kb

DSLR -> 848x568 image of about 1.4Mb

Ignoring the differences in processing these two images are pretty comparable I think. Given decent viewing and proper processing the DSLR can easily give the webcam a run for it's money.

Webcam

Jupiter_640x480_Webcam.bmp

DSLR

JUPITER_848x568_DSLR.bmp