Why use a webcam?

[Astrosurf]

OK, this is embarrassing. I've  been imaging for some years now, but when I tried to explain atmospheric turbulence and use of a webcam for planets, I got stuck. I know that using a webcam for the bright planets is best due to the short frames. I understand why a webcam also useful for coping with seeing, as it can capture that split second where seeing is clear. But this is where I got stuck: surely there's atmospheric turbulence when shooting DSOs with a DSLR so why isn't there the same problem? Or is the turbulence more obvious with bright objects?

I blush while I write this!

Alexxx

[red dwalf]

don`t blush, i`d like to know also

[ollypenrice]

There is exactly the same problem but there's (almost) nowt you can do about it! So it wasn't a silly question.

Damian Peach can resolve details on the moons of Jupiter. No deep sky imagers resolve to that level. This is partly because an image of M42 at the same scale would be about a mile across. (Invented statistic but the point is correct. ) It is also because taking fraction-of-a-second images of DSOs would not record anything much on the chip. The only way to get faint stuff to show is to take long exposures and let the atmospheric turbulence do its worst.

What can the DSO imager do about this?

1) Locate their telescope in space. (You pay for this one, I'll buy the next...  )

2) Use adaptive optics. Fire a laser up to the sky parallel with the scope and find a multi million dollar way of distorting the optics in almost real time once the returning laser  beam (reflected from the high atmosphere) has informed a computer of what the distortions need to be to calibrate out the atmosphere. (This will have to be your round as well!  )

3) Use Active Optics. This is a system in which a fast tilting disc of glass is placed in front of the guide camera and the imaging camera. Instead of the guide camera finding shifts in the guide star and moving the mount, it finds shifts in the guide star and tilts the glass disc to put the beam back where it should be. It can re-tilt the glass disc way faster than it can move the mount (maybe 8 times per second as opposed to mount nudges every couple of seconds or so.) These are made by a couple of firms for amateurs and are effective on high resolution, long focal lengths setups - if you have a bright enough guidestar and the patience to make them do what is says on the tin. Whether this is really just a way of correcting mount errors super-quickly, or whether it can really take out any atmospheric effects, I don't know. I very much doubt that it has much power over turbulence since it makes whole-image corrections.

Fortunately the image scale used by DS imagers is very coarse compared with planetary imagers so we don't need the kind of resolution they need. Our pixels contain too much sky to allow us to resolve tiny details.

Olly

[Christopher Davenport]

You just made me blush thinking about this. 

I think the answer comes down to size of the object, brightness and the focal length required.

So to capture Jupiter very bright and small,  I need a focal length of around 3000mm, with a exposure of 10ms and +- 1000 Stack

Whereas to capture Orion, faint and large, I only need a focal length of 600mm, with a exposure of 180s +- 25 Stack

Having said that atmosphere will still be a problem on Orion, so long exposure and stacking must somehow be averaging the seeing conditions.

I am sure there is a voice of higher reason on this forum which will knows better. i.e Olly above who beat me to the post.

[IanL]

As Olly says, there is exactly the same problem with DSOs, but DSO "lucky imaging" isn't possible for amateurs at the moment. This is due to the fact that there is so little signal from a faint target that it would be swamped by noise in an exposure short enough to (potentially) peek through the seeing.  It is possible with lunar and planetary imaging because the targets are much brighter and therefore sufficient signal can be gathered in short exposures, i.e. processing a video in Registax or similar to average the best bits in to a final image.

It may become possible to do lucky imaging on DSOs eventually, e.g. here is a write up about L3CCD technology:

http://www.ast.cam.ac.uk/research/instrumentation.surveys.and.projects/lucky.imaging/l3ccd.technology

The basic concept is that the CCD sensor elements have a (low) probability of initiating a cascade effect and massively amplifying a very small signal in to something much more useful, overcoming the other sources of noise.  Combine that with lots of exposures and selecting the best information and it may (one day) be possible to use a similar technique to planetary/lunar imaging on DSOs.  So in theory you could shoot at the maximum resolving power determined by the aperture of the scope and the pixel scale of the entire system rather than being limited by the prevailing seeing conditions.

[michael.h.f.wilkinson]

As Olly said the main difference is the resolution used in most DSO imaging. In planetary, you sample the image at the limit imposed by your optics (the so-called Nyquist limit), whereas in most DSO imaging you do not go anywhere near that. Another issue is the ratio of photon noise to readout noise. Because planets are bright, the small amount of readout noise in each frame does not matter very much, and photon noise dominates the image (because there are many photons striking the chip). Even if thousands of frames are stacked, the readout noise is rarely an issue. By contrast, if I take 1/60s exposures of DSOs, the photon count is so low, that readout noise is much higher than the photon noise, so that if I stack thousands of images I end up with more readout noise than photon noise. This is not a situation you want.

[michael.h.f.wilkinson]

Must type faster

[Astrosurf]

You're brilliant guys, thank you!

[Christopher Davenport]

Great answers on here, especially on the adaptive optics bit.

I know that you can buy a adaptive optics filter wheel to put on before your mono cam.

http://www.firstlightoptics.com/starlight-xpress-accessories/starlight-xpress-sxv-ao-lf-large-format-active-optics-unit.html

Are these worth it in terms of image quality and if so why are they not all the rage?

[JamesF]

To further illustrate what Olly and others have said, it's not uncommon for planetary imagers to be working at ten or maybe even fifteen times the focal length used when DSO imaging.  A half-pixel shift in the image due to atmospheric conditions when DSO imaging, which might not even be visually detectable in the finished image, might be five pixels or even more (at the same pixel size) for a planetary imager using those much longer focal lengths.  That's quite a significant degradation of the image unless the frame rate is sufficiently fast that the total shift is spread over lots of frames.

James

[sharkmelley]

The effect of atmospheric turbulence is quite obvious in DSO imaging - it spreads the light from a pinpoint star over a small area and makes them appear as discs.  We are all familiar with the effect that the discs of bright stars are larger than those of dim stars - this is atmospheric turbulence at work.  Similar smearing is taking place over the whole image - though this is more noticeable at longer focal lengths.  In brighter parts of a DSO image, it is possible to improve image detail (i.e. reduce smearing) using wavelets/deconvolution just as you would for planetary imaging.

Mark

[Astrosurf]

Very informative, guys. Thanks!

[dph1nm]

It may become possible to do lucky imaging on DSOs eventually

Don't forget you will always need enough photons from your object in each exposure to be able to centroid the image so you can stack. Otherwise lucky imaging is not going to help much. Short exposures may show good seeing but the object position will move around slightly from exposure to exposure (which is why long exposures end up blurry). So you need to be able to take this out when stacking.

NigelM

[IanL]

Don't forget you will always need enough photons from your object in each exposure to be able to centroid the image so you can stack. Otherwise lucky imaging is not going to help much. Short exposures may show good seeing but the object position will move around slightly from exposure to exposure (which is why long exposures end up blurry). So you need to be able to take this out when stacking.

NigelM

There isn't enough information in the page I linked to determine how the 'stacking' would be done and whether it is similar to what we do now, or some different approach.  Clearly there would have to be some means of registering the frames otherwise you still get blur, but the basic point is that you end up with a very large gain with low readout noise for about 2% of the detected photons.  I.e. You end up with a lot of signal from very few photons, making short exposures possible and (presumably) capable of registration and integration.

[Mr TamiyaCowboy]

i think the best point for use of a webcam is this ......... 

COST !!!!! 

most simplest and cheapest way a newcomer can get into filming/photographing the sky is via camera.

the cheapest camera of all is the humble webcam, small cheap to produce and the need for basic

auto settings and a usb. you can pick up webcams these days for under £10. 

DSLR / Astro cam spec systems, not everyone can afford to throw out a couple hundred pounds on a astro cam

or even a DSLR , models are becoming cheaper so a newcomer can atleast save up some pennys.

DSLR are setting hungry its a new learning curve unlike a plug n play webcam. ISO/exposure/whitebalance

then dark frames light frames and those flat frames, not forgetting drizzle and bayer.

webcams are cheap, they are not so powerful in sensitivity as a DSLR hence more used for lunar and the planets.

but it all boils down to one thing COST , the price you pay for a decent starting planetry imaging kit.

[ollypenrice]

Great answers on here, especially on the adaptive optics bit.

I know that you can buy a adaptive optics filter wheel to put on before your mono cam.

http://www.firstlightoptics.com/starlight-xpress-accessories/starlight-xpress-sxv-ao-lf-large-format-active-optics-unit.html

Are these worth it in terms of image quality and if so why are they not all the rage?

This in an active, not an adaptive accessory. (An adaptive system might use a deformable mirror supported on adjustable pistons which, incredibly enough, adapt the mirror shape according to feedback from the laser.)

I don't know why the active units are not more popular. I've had them used here twice on long focal length instruments and they were impressive but a bit of a palaver to set up.

Olly

[Christopher Davenport]

This in an active, not an adaptive accessory. (An adaptive system might use a deformable mirror supported on adjustable pistons which, incredibly enough, adapt the mirror shape according to feedback from the laser.)

I don't know why the active units are not more popular. I've had them used here twice on long focal length instruments and they were impressive but a bit of a palaver to set up.

Olly

Aaah, I see the difference.

I can see how a adaptive system could cost a lot more. 

So active is more for guiding problems than seeing problems.

Makes sense now, why not more popular as is $$$ on top of a $$$ mount.

[Astrosurf]

Brilliant thread guys. Thank you!