R.I.P. Nikon D3100 - AstroPhotography

[squeaky]

Yep, it's gone.

I bought my scope from a shop that sold both scopes and cameras, so naturally I asked them for a recommendation for a camera specifically for astro-photography.

They said the Nikon D3100, and who was I to argue?

Hmm...

By spec it sits about mid way between the Canon 1100D and the 550D. And that's about all that's comparable! <sigh>

Prime Focus on my scope, according to the information I was given calculates to a magnification of x40 which gives me an FOV roughly equivalent to my 32mm lens. BUT... I must use a 2x Barlow in order to be able to get a focus - so that now gives me an FOV roughly equivalent to my 17mm EP.

Except... the camera obviously images a rectangle inside this circular view and so the FOV is tighter still. So I did a calibration shot and compared the view in Stellarium and it turned out that the FOV was equivalent to a my TEN mm EP. Sheesh! (On my scope that's a magnification of x150)

And that's the maximum FOV I can image!

So you can imagine that if I'm a tiny bit out on centring a star in my 9x50 finder scope - my target just is NOT going to be in my field of view on the camera. I could not for the life of me find a way to get the live view to show me anything fainter than Mag 4 stars - so with M13 as the target, for example - I couldn't see a darn thing. And live capture to a laptop is not possible with this camera. So the only way to find out where I was pointing was to take an exposure and look at the preview. Hmm... a few faint stars - and if there's no recognisable pattern there (such as a trail of stars, or a triangle, or something) then I had no option but to start quartering for my target. So I'd work round my position in a square search pattern, and if I still couldn't recognise a pattern against the Stellarium view I'd have to move the search out one more grid and go round again. This is a ten second exposure after a ten second delay to avoid scope shake after pressing the button on the camera plus camera image saving time. So almost a minute between each take; plus the time it took me to find a pattern - or, more to the point... TRY to find a pattern. Once found I could THEN star hop to the target.

Then it was always a case of taking several shots with little tweaks of guidance until the target was centred and stable with minimal drift.

Total set up time was seldom less than an hour and often nearer two!

That wasn't the only problem. If I used EP's to centre my target in the FOV and then eased out the EP and gently (VERY gently) eased in my camera the target STILL would not be in the field of view!

Because...?

Because I was hanging 630 grams of camera on a nose piece with a barlow in the optical train and so the camera's weight was centred 13cm outside the OTA and in fact was a grand total of 31 cm (yes that's really thirty one centimetres) off the central axis of my scope.

Not to mention possible misalignment because of so many bits in the optical train and the weight bending it down.

Trying to counter balance the weight on my dob mount through the gearing of the goto motors just didn't work.

THEN... the preview image on the LCD screen after I had taken a shot would be brilliant. Unfortunately I would later find out that the actual image saved was MUCH fainter and muddier! I couldn't for the life of me find a way to make the preview image on the LCD screen look like what I was going to be saving, nor could I find a way to get the saved image to look like the LCD preview.

Nor could I find help either. The Canon helpful stuff must outnumber the Nikon helpful stuff by about a million to one

Finally I decided to go look for an image of M13 that I might realistically expect from my scope with an equivalent Canon and, once I found one I copied the image and sent a comparison to the shop...

The left hand image was my best effort so far on my 12" scope and an equivalent to the 1100D/550D...

The right hand image was taken on a SIX inch scope with a Canon 350D

!!!!

Well that did it for me. I contacted the shop by phone and along with explaining all my troubles sent them an email holding those two comparison shots.

Two hours later I was in the shop handing over the Nikon for a full refund. No arguments - think I made my case

So... guess who's shopping around for a Canon 550D ?

[themos]

So the only way to find out where I was pointing was to take an exposure and look at the preview. Hmm... a few faint stars - and if there's no recognisable pattern there (such as a trail of stars, or a triangle, or something) then I had no option but to start quartering for my target.

You have another option: submit it to live.astrometry.net which will "solve" the image and tell you the sky coordinates of the center. For example

http://live.astrometry.net/status.php?job=alpha-201203-26590107

[jgs001]

I don't know these scopes, but I think, there's a cover on the end of the draw tube, that unscrews to reveal a T thread... if you attach an SLR directly to that, you shouldn't need a barlow. The problem with using a 2x barlow, you are quadrupling the amount of time needed to capture the same amount of light... your f/5 scope, becomes f/10. You're going to struggle, especially with AltAz tracking, to get a long enough exposure. You probably don't want to hear this, but you can't compare an image taken without any details of the kit involved and directly compare it... there are too many factors involved, length of exposure time per sub, how many subs, how good the skies were at the time, the telescope used, the mount used... etc.

The preview on the LCD screen, won't look like the image on the computer. The LCD screen shows you the jpg preview of the data, as if the camera had processed the data. However, you need a lot more than one to create an astrophoto, along with a string of darks, flats and bias...

The main reason, I think, a lot of people use Canon for AP has more to do with the software support for the camera, rather than the ability of the camera to produce an image.

FWIW, I think you've done a nice job on that M13... you've clipped the black point a bit in processing it, the sky looks rather too black, and you don't have the same amount of colour information as the Canon example you've linked to, but more exposures will help with that one.

[carlc]

Well I had no plans to get into AP but this thread has definitely killed it off.

I spent hours (days) on photography forums getting advice like - Canon and Nikon are both the same - it's what you prefer (blah blah).

I went into a shop, liked the feel of the D3100 over the 1100D (slightly better specs too).

I should have paid more attention to reading blogs/websites of wildlife photographers (which is what I like to do) - they (almost) exclusively use Canon. A mid length prime telephoto lens is just where I would expect to be sat in terms of camera glass - and it's just where Nikon is beaten hands down by Canon!

Sometimes you have to suck it and see (but it can get expensive) - I'm glad you could get a refund. I'll be looking to part ex (if I don't get another scope, eyepiece(s), filter(s) etc first).

Sorry - that came across as a real whine - what I really was trying to say is that posts like this help to prevent people making mistakes in equipment choices. So thank you for that.

[squeaky]

You have another option: submit it to live.astrometry.net which will "solve" the image and tell you the sky coordinates of the center. For example

http://live.astromet...201203-26590107

Hmm... thanks for that. Very useful.

Certainly when I'm viewing from home. Out in a field somewhere I won't have access.

[squeaky]

I don't know these scopes, but I think, there's a cover on the end of the draw tube, that unscrews to reveal a T thread... if you attach an SLR directly to that, you shouldn't need a barlow. The problem with using a 2x barlow, you are quadrupling the amount of time needed to capture the same amount of light... your f/5 scope, becomes f/10. You're going to struggle, especially with AltAz tracking, to get a long enough exposure.

That's the point here. There is no way that I know of, nor did the shop, nor did a VERY helpful guy at Telescope house, to get a focus OTHER than using a Barlow with this camera. The one possible other option didn't work.

You probably don't want to hear this, but you can't compare an image taken without any details of the kit involved and directly compare it... there are too many factors involved, length of exposure time per sub, how many subs, how good the skies were at the time, the telescope used, the mount used... etc.

Yes, I know this - so I did my best in terms of the comparison shots - but while I'm well aware that with longer exposures and a lot more subs I'd be able to improve on my image and get much closer to the standard of the second one - this wasn't really the only issue. FOV and getting a target was the real pain.

The preview on the LCD screen, won't look like the image on the computer. The LCD screen shows you the jpg preview of the data, as if the camera had processed the data. However, you need a lot more than one to create an astrophoto, along with a string of darks, flats and bias...

The main reason, I think, a lot of people use Canon for AP has more to do with the software support for the camera, rather than the ability of the camera to produce an image.

Yep, I've since been told that, as you say I'll never get an image that looks like the LCD view - though again it's that lack of support for the Nikon, and especially the fact that with this model you cannot use live capture or external image control software such as Backyard EOS. Other models, wa-a-ay more expensive, allow it - but not this one.

FWIW, I think you've done a nice job on that M13... you've clipped the black point a bit in processing it, the sky looks rather too black, and you don't have the same amount of colour information as the Canon example you've linked to, but more exposures will help with that one.

Yes, I've been learning more about processing since then and should be able to have a more realistic sky and also lift up those background stars now without blowing out the central core and keeping the vibrancy and saturation. Working with layers helps too. I had a play with that image johnh posted of the Ara globular and managed to do that quite well. Then used layers - one of my background lift, plus his original (because the core was better) and dropped the opacity of his shot until I could "just" see those enhanced background stars coming through. Then sorted out the sky. When I next have a decent number of subs of M13 (plus darks, bias, flats) and have something to work with I reckon I'll go for three distinct layers... one for the background star field, one for the outer regions of the cluster, and one for the core. Properly set for opacity etc it might just come out quite well.

And I STILL have MUCH to learn, Grasshopper. Honest

[squeaky]

Well I had no plans to get into AP but this thread has definitely killed it off.

I spent hours (days) on photography forums getting advice like - Canon and Nikon are both the same - it's what you prefer (blah blah).

I went into a shop, liked the feel of the D3100 over the 1100D (slightly better specs too).

I should have paid more attention to reading blogs/websites of wildlife photographers (which is what I like to do) - they (almost) exclusively use Canon. A mid length prime telephoto lens is just where I would expect to be sat in terms of camera glass - and it's just where Nikon is beaten hands down by Canon!

Sometimes you have to suck it and see (but it can get expensive) - I'm glad you could get a refund. I'll be looking to part ex (if I don't get another scope, eyepiece(s), filter(s) etc first).

Sorry - that came across as a real whine - what I really was trying to say is that posts like this help to prevent people making mistakes in equipment choices. So thank you for that.

I wouldn't let my experiences put you off

I'm new to these super duper auto guiding find anything super tracking scopes - so there's a lot for me to learn there. I've only just reached the stage where in my home viewing spot I can wheel the scope out of the shed, set it in place and find Polaris in broad daylight. This means that I can now get alignment done at dusk and in about five minutes flat - whereas when I first started this would take me an hour. IF I didn't trip over the power cable in the dark and have to start all over again

As a camera, per se, that Nikon is brilliant. If I want to take shots of my cat sleeping on the sofa. It just has some limitations when it comes to astro-photography and, once I solved one problem I'd find that this led me to the next stage where I'd find another problem (or two) that needed resolving. Again this is all on top of what might be called the newbies learning curve, so I don't expect to be able to point and shoot my replacement camera just like that - but I do expect it to be easier, and it's very clear that help and support for the Canon is widely available and very good.

Oh, and a tip for D3100 users - that Prime Focus plus Barlow limitation means that you MIGHT do better using the optical holder and an eyepiece instead. The recommended MINIMUM focal length is 15mm - so from my Celestron lens kit the 13mm "might" give a wider field of view - and you may just get away with using the 17mm. At least for setting up purposes. I never got as far as trying it.

[JamesF]

That's the point here. There is no way that I know of, nor did the shop, nor did a VERY helpful guy at Telescope house, to get a focus OTHER than using a Barlow with this camera. The one possible other option didn't work.

The problem there may be that you just can't get the camera sensor forward far enough to coincide with the image plane of the telescope. By adding a barlow you effectively move the

image plane further out towards the end of the focus tube, making it easier to reach. It's entirely possible this will happen regardless of which DSLR you use.

It may be possible that fitting a low-profile focuser would resolve the problem because it allows the camera to move forwards more. Hopefully someone else will have some ideas because the only other one I know of is to move the mirror up the tube closer to the secondary.

James

[JamesF]

Oh, hang on. You've got a flex tube scope? You *might* be able to get away without a barlow if you don't open the trusses out fully. There are compromises if you do that, such as the secondary potentially being too small, but perhaps it's worth a try?

James

[jgs001]

As James said, I would expect the focus problem with any dSLR, not just the Nikon. I've read mention about the T thread on the end of the draw tube. I've also heard others say about replacing the focuser with a low profile one, and even moving the primary mirror up the tube with the collimation bolts. Having never used a Newt I can't specifically comment from my own experiences.

Also... what James just said, rather than move the primary, lower the struts a little.

[squeaky]

The problem there may be that you just can't get the camera sensor forward far enough to coincide with the image plane of the telescope. By adding a barlow you effectively move the

image plane further out towards the end of the focus tube, making it easier to reach. It's entirely possible this will happen regardless of which DSLR you use.

It may be possible that fitting a low-profile focuser would resolve the problem because it allows the camera to move forwards more. Hopefully someone else will have some ideas because the only other one I know of is to move the mirror up the tube closer to the secondary.

James

They recommended that I get a short focus barlow, so I did, and even THAT wouldn't get a focus until I used a low-profile focuser - well, a shorter EP holder.

As for moving my mirror... hmmm... The scope is still under warranty so I'd be reluctant to do anything so intrusive.

Anyway - unless I see the same problems with my new camera, when I finally decide on which one though I'm trending to the 550D this is all ancient history for me now.

Might help others though!

[squeaky]

Oh, hang on. You've got a flex tube scope? You *might* be able to get away without a barlow if you don't open the trusses out fully. There are compromises if you do that, such as the secondary potentially being too small, but perhaps it's worth a try?

James

Blimey! THAT suggestion never came up before! Sounds like a brilliant idea

One I'll certainly bear in mind when I have a camera again. Dead simple and a whole lot cheaper than all the routes suggested by the shop and by TH.

[squeaky]

As James said, I would expect the focus problem with any dSLR, not just the Nikon. I've read mention about the T thread on the end of the draw tube. I've also heard others say about replacing the focuser with a low profile one, and even moving the primary mirror up the tube with the collimation bolts. Having never used a Newt I can't specifically comment from my own experiences.

Also... what James just said, rather than move the primary, lower the struts a little.

Thanks

If I lowered the struts a little I know I'd have to redo collimation - but would this also reduce the field of view?

[JamesF]

Something else to bear in mind should it come to it, I have a feeling that somewhere I've seen a post extension kit for the larger Skywatcher dobs to move the mirror up the tube by about an inch. I can't remember which scopes they were for and it may just be the solid tube ones, but if you're stuck it might be worth asking around.

James

[jgs001]

Sorry, I really don't know... If it did reduce the FOV a little, you'd still have a wider FOV than you do now with the barlow...

[JamesF]

If I lowered the struts a little I know I'd have to redo collimation - but would this also reduce the field of view?

It shouldn't reduce the field of view, but there's a possibility that you might see some vignetting because the secondary is too small for the light cone at that point. Without actually trying it or doing the maths it's hard to say though.

James

[JamesF]

Actually, let me qualify that a little better

It shouldn't noticeably reduce the field of view as long as you don't overdo it.

James

[squeaky]

I'm So-o-o-o glad I started this thread - it means I don't have to bookmark it.

James - I think you might be right about that mod kit. I think I saw something way back when while looking for a solution to a different problem, so I'll have to search again and find it.

I rather thought moving the struts in would limit the FOV but not as much as the Barlow - just thought I'd check and make sure. Right now my brain hurts

[JamesF]

If you can visualise how the scope works, the primary mirror creates a cone of light going back up the tube and the secondary "bends the point over" so it travels down the focuser. Ideally the secondary should be big enough that all of the cone hits it, so you don't lose captured light back up the tube. If you move the secondary closer to the mirror a little (and a little is probably all you need) then you "bend the cone" a bit further down, where it's wider, and the point of the cone is therefore further up the focus tube. If you overdo it though, light will start to escape past the secondary. Up to a point that may not matter because the camera sensor can't use the entire "lit" area at the image plane anyhow -- it should be larger than the camera sensor.

It's not an ideal solution I'd be the first to admit, and it does require a bit of experimentation and mucking about. It's the best I can think of without permanently modifying the scope though.

James

[squeaky]

Aye, I've got that mental image - which is why I queried just to check that I was right.

Anyway, for now - all this has to wait until I have a camera again.

Fascinating stuff though, innit?

[themos]

Hmm... thanks for that. Very useful.

Certainly when I'm viewing from home. Out in a field somewhere I won't have access.

The alternative is to load the software on your laptop and do a local "solve". That's what I do for my polar alignment method for equatorials.

[E621Keith]

Prime Focus on my scope, according to the information I was given calculates to a magnification of x40 which gives me an FOV roughly equivalent to my 32mm lens. BUT... I must use a 2x Barlow in order to be able to get a focus - so that now gives me an FOV roughly equivalent to my 17mm EP.

Except... the camera obviously images a rectangle inside this circular view and so the FOV is tighter still. So I did a calibration shot and compared the view in Stellarium and it turned out that the FOV was equivalent to a my TEN mm EP. Sheesh! (On my scope that's a magnification of x150)

And that's the maximum FOV I can image!

Nikon camera has a longer flange focal plane distance than Canon (46.5 vs 44mm) so it will required a little more in focus than the Canon. You scope is a dobsonian, so it was never designed for mounting DSLR. You may need a special low profile focuser to bring the DSLR into focus (or using that strut adjustment technique mentioned). I don't think changing to Canon will solve this problem because the difference in focal distance is only 2mm. Nikon DX format sensor is slightly bigger than Canon's APS-C, so you will get a slightly bigger FOV (very slight) on the Nikon.

So you can imagine that if I'm a tiny bit out on centring a star in my 9x50 finder scope - my target just is NOT going to be in my field of view on the camera. I could not for the life of me find a way to get the live view to show me anything fainter than Mag 4 stars - so with M13 as the target, for example - I couldn't see a darn thing. And live capture to a laptop is not possible with this camera. So the only way to find out where I was pointing was to take an exposure and look at the preview. Hmm... a few faint stars - and if there's no recognisable pattern there (such as a trail of stars, or a triangle, or something) then I had no option but to start quartering for my target. So I'd work round my position in a square search pattern, and if I still couldn't recognise a pattern against the Stellarium view I'd have to move the search out one more grid and go round again. This is a ten second exposure after a ten second delay to avoid scope shake after pressing the button on the camera plus camera image saving time. So almost a minute between each take; plus the time it took me to find a pattern - or, more to the point... TRY to find a pattern. Once found I could THEN star hop to the target.

Then it was always a case of taking several shots with little tweaks of guidance until the target was centred and stable with minimal drift.

Total set up time was seldom less than an hour and often nearer two!

That wasn't the only problem. If I used EP's to centre my target in the FOV and then eased out the EP and gently (VERY gently) eased in my camera the target STILL would not be in the field of view!

Because...?

Because I was hanging 630 grams of camera on a nose piece with a barlow in the optical train and so the camera's weight was centred 13cm outside the OTA and in fact was a grand total of 31 cm (yes that's really thirty one centimetres) off the central axis of my scope.

Not to mention possible misalignment because of so many bits in the optical train and the weight bending it down.

Trying to counter balance the weight on my dob mount through the gearing of the goto motors just didn't work.

This is a problem with your scope. Again, dobsonian was not designed for AP, it's a purely visual system, so it doesn't matter whether you hang a Canon or a Nikon you will get the flex, the tracking error, etc...

Live view is a nice feature but it's not necessary, it makes your live a lot easier if you have it, but it doesn't mean you can't image without it. My camera doesn't have live view and can't be teathered to a laptop, so what I do is take a exposure, check focus in the result image, then make a small incremental adjustment on the focuser, and repeat the process until I get the camera in focus.

You can solve to shutter vibration using a Nikon ML-L3 IR remote. They are about £20 and allow you to trigger the shutter without touching the camera.

THEN... the preview image on the LCD screen after I had taken a shot would be brilliant. Unfortunately I would later find out that the actual image saved was MUCH fainter and muddier! I couldn't for the life of me find a way to make the preview image on the LCD screen look like what I was going to be saving, nor could I find a way to get the saved image to look like the LCD preview.

Nor could I find help either. The Canon helpful stuff must outnumber the Nikon helpful stuff by about a million to one

Finally I decided to go look for an image of M13 that I might realistically expect from my scope with an equivalent Canon and, once I found one I copied the image and sent a comparison to the shop...

The left hand image was my best effort so far on my 12" scope and an equivalent to the 1100D/550D...

The right hand image was taken on a SIX inch scope with a Canon 350D

!!!!

That sounds like you did not set your exposure correctly. Liveview displays a good image because it raises to ISO to a silly level, but when you take the image the camera imaged using the setting given to it. What setting were you using? What was the ISO and exposure time? Usually DSO images needs to be expose for several minutes. This may sounds silly, but did you shoot your camera in Auto, P, S, or M mode? You need to use M (Manual Mode) for AP. Also did you shoot NEF (RAW) or JPEG? You can pull a lot of details out of a RAW image in post processing.

You should keep the Nikon D3100 and buy another telescope. The biggest problem you have is trying to image DSO with a dobsonian. It will be much better if you have a ED80 and HEQ5. While Canon have an advantage over Nikon for AP, they are all on the software and accessories side. The problem you described isn't related to the camera, but your scope and technique. Simply changing the camera won't solve your problem.

That Canon image with 6" scope was probably taken using a 6" (150p) on a GEM using long exposure, so it's only natural it's better than one taken on a barlowed 12" dob. Have a look at this link. That M13 was taken by a 8" with an ancient Nikon D70. Your D3100 is 3 generations newer, so your camera isn't at fault here.

http://smithplanet.c...r/m13/index.htm

[shaunster]

Like the post above me says, you probably got most things wrong and blamed the camera, you would in all likeliness have had bad results with a canon as well. You simply cannot compare images taken with different cameras, different scopes and on different nights and expect them to be the same. A lot of the final result is in the processing, gathering good data is just the first step on the road to a good image. Give 10 people the same scope and same camera and you will always see 10 different results.

[Keiran]

You could have tried a 2" nose piece with a low profile T ring? i dont see why you need to use a barlow. you can use extension tubes.

[Steve Ward]

Sounds like you've been having all sorts of fun ,

Glad you got a refund on the Nikon , will keep my eye out for a 550D.