SUITABLE CAMERA FOR PLANETARY IMAGING

[johnturley]

Have done a bit of planetary imaging recently through my 14in Newtonian using my Canon 6D full frame digital SLR using eyepiece projection, and taking an MVI file which I convert in PIPP, and then process in Registax and Adobe Lightroom. 

This method works fine for Lunar imaging (see attached image), but for planets in particular Mars, I end up with a very small image size.

Is there any way round this problem, or do I need to get a Astro type camera with a suitable sensor size, I really wanted to avoid having to have a laptop with associated trailing USB cables in my observatory shed in the dark.

John

Alpine Valley and Plato Processed.bmp

[AbsolutelyN]

A 6D is a great camera but better suited to deep sky images as it has a large pixel size. You'd be better with something like the ZWO 224MC which has a much smaller pixel size. You can then select a much smaller area of the frame and run at a much higher frame rate. Prime focus with a barlow will probably give better results. I'm sure others can answer much better. 

[Tommohawk]

Hi again!

The main issue with planets is that no matter what you do (reasonably!) with the optics, your actually image on your 6D sensor will appear tiny. My mono planetary camera has a sensor only 3x5mm approx. 

The disadvantage of such a tiny sensor is locating the target can be a right pain - BUT you can process uncompressed video at about 400fps. In fact once the target is located, typically you use only the centre of the sensor - the region of interest or ROI. This might be just a couple of mm across, maybe 400x300 pixels for Mars.

I dont think the 6D offers uncompressed video at all - you might want to check that though. Also check Magic Lantern - I've never used it but it might be useful. Ironically, the smaller 550D offers "video crop mode" which uses the centre 1/5 of the sensor and does uncompressed video.

If you cant find a way of getting uncompressed video, you might want to experiment with just multiple stills, but typically planetary imagers use tens of thousands of individual images - it would take a while and hammer your shutter count!

The 6d is obviously a fabulous camera, and great for DSOs, but I think its probably not the right tool for planetary work.

I don't know if there's a self-contained planetary camera available - maybe someone else will know. Like you I resisted the need for a laptop control, but gave in eventually. If you're going to do any serious DSO work you might want one anyway for guiding. 

Theres loads of conventional planetary cameras, ZWO have been popular lately, but lots of others. I'm experimenting with Omegon 385C at the moment.

 

  

[Nik271]

Hi John,

I am imaging with a Canon 250D in prime focus and  share your reluctance to move to a dedicated planetary camera since that will involve another gadget, a laptop and more cables to trip in the dark. As people have already pointed out there are many disadvantages to DSLR for planetary: large sensors, large pixels, lossy compression and being limited to 30fps in older models cameras.

However I like the fact that I can find the target directly in the optical viewfinder in the camera and get rough focus even before switching live view  And its easier to track of course, even manually.  There are several things I've learned with trial error and reading this forum:

Seeing, focusing and correct image scale (in this order I believe) are more important than the sensor: if you get a sharp good image on your screen this will beat the disadvantages. Even at 30fps you can get 4000 frames on Jupiter easily with these large pixels before the rotation gets too bad for  AS!3 to correct.

Now for image scale: As a rule of thumb in average seeing you should be imaging with focal ratio at 4x or 5x your pixel size in microns.

Your pixels are 6.5 microns so you should be aiming at a focal ratio about 4x6.5= 25. At these focal lengths Mars will be bigger

So my advice it to try camera in prime focus first.

Nikolay

 

Edited September 16, 2020 by Nik271
typos corrected

[Nik271]

PS:

I just noticed you wrote  that you use a 14 inch Newtonian. I presume this is F5 so maybe about 1800mm focal length?

Then F25 will make it 9000mm which is too much for the seeing in UK. Try with a 3x barlow at F15 first, already at that focal lengths Mars will be bigger and easier to focus.

Edited September 16, 2020 by Nik271
got my maths wrong :)

[johnturley]

1 hour ago, Nik271 said:

PS:

I just noticed you wrote  that you use a 14 inch Newtonian. I presume this is F5 so maybe about 1800mm focal length?

Then F25 will make it 9000mm which is too much for the seeing in UK. Try with a 3x barlow at F15 first, already at that focal lengths Mars will be bigger and easier to focus.

Thanks for your reply

My 14in Newtonian is f5 with a focal length of about 1800mm, the above photos of Jupiter and Mars were taken using eyepiece projection with a 9.7mm Plossl eyepiece which I would have thought would have given more amplification than a 2 or 3x Barlow, I do have an ES 2in 2x Barlow, and I'll give it a try with that to check. I could consider getting a TV 4x 2in Powermate which would give me f20, but it is quite expensive, and doubt as to whether the amplification would be greater than using eyepiece projection with the 9.7 mm Plossl. What really surprises me is that there appears no software that allows you to increase the image size of a planet, you can increase it when viewing on a computer, but not it appears the saved image.

My 14in Newtonian would not actually be very suitable for imaging with dedicated planetary camera with a laptop, due the positions the focusing mount, depending on the position and altitude of the planet, can end up in. My piggybacked Esprit 150 would be better, as the position of the focusing mount in this case is much less variable, but besides a new dedicated planetary camera, I would probably also need a new laptop with a faster processor.

John 

 

[Nik271]

58 minutes ago, johnturley said:

Thanks for your reply

My 14in Newtonian is f5 with a focal length of about 1800mm, the above photos of Jupiter and Mars were taken using eyepiece projection with a 9.7mm Plossl eyepiece which I would have thought would have given more amplification than a 2 or 3x Barlow, I do have an ES 2in 2x Barlow, and I'll give it a try with that to check. I could consider getting a TV 4x 2in Powermate which would give me f20, but it is quite expensive, and doubt as to whether the amplification would be greater than using eyepiece projection with the 9.7 mm Plossl. What really surprises me is that there appears no software that allows you to increase the image size of a planet, you can increase it when viewing on a computer, but not it appears the saved image.

My 14in Newtonian would not actually be very suitable for imaging with dedicated planetary camera with a laptop, due the positions the focusing mount, depending on the position and altitude of the planet, can end up in. My piggybacked Esprit 150 would be better, as the position of the focusing mount in this case is much less variable, but besides a new dedicated planetary camera, I would probably also need a new laptop with a faster processor.

John 

 

Yes, try with a 2x Barlow  - it will make a difference. What matters is the image scale which is measured in arcseconds per pixel. I use this calculator http://astronomy.tools/calculators/ccd

With 2x Barlow you are at 3600mm and with 6.5 micron pixels you will get .37'/pixel. Jupiter is about 40' so it will cover about 120 pixels, not bad. Most of the images that people post are crops of 480 by 480 pixels and so Jupiter will be 1/4 width of the image and look nice.

To give you an idea of the image size I attach  a processed, cropped image of Jupiter I took yesterday at prime focus and my native focal length 2700mm with my canon 250D which has 3.7micron pixels. The image has not been rescaled, only cropped to 480x480pixels. My image scale is .28'/pixel so this Jupiter is only 20% larger than what you will get with 2x barlow.

I think with a wide aperture you can use a 2.5x or  3x Barlow and still resolve much more detail than this image.

Nikolay

 

 

Edited September 16, 2020 by Nik271
typos!

[Nik271]

Oops I missed something very important:  you are taking videos, not still photos with the 6D.

The problem is that the video is only HD i.e. 1080 by 1920  while you sensor is 3600x5500 px

So the camera software is 'resampling' the image in effect combining several of you original pixels into one.

The conversion ruins the image scale: Now your sensor has been converted to fewer and so HUGE pixels, looks like 3 times larger. This means that in the image everything will be 3 times smaller than what I said earlier. No barlow can fix that big difference. (unless there are 6x barlows but they probably have aberrations)

Perhaps instead of video you can take 100 still photos, center and crop the planet in PIPP and stack these? Its too few frames but will give larger images than the video in prime focus.

    

[Tommohawk]

1 hour ago, johnturley said:

What really surprises me is that there appears no software that allows you to increase the image size of a planet, you can increase it when viewing on a computer, but not it appears the saved image.

I'm not sure if I understand you correctly, but to a great extent this is about how you display the image. You can certainly make the image appear bigger by cropping it and resizing it. Of course this is likely to lose resolution because you have to interpolate for the additional pixels. You really need to increase the raw image size by a combination of image amplification and ideally small pixels. My current planetary camera has 3.8 micron pixels so not teeny weeny, and I use a 5x powermate. 

 

1 hour ago, johnturley said:

My 14in Newtonian would not actually be very suitable for imaging with dedicated planetary camera with a laptop

I would bite your arm off for a go at planetary imaging with that scope!! You are right that you have to be careful with trailing cables especially because the camera position would move around considerably with a big newtonian, but for planetary work aperture is definitely helpful. The best images posted here are typically done with big aperture scope like 14" SCTs or big newtonians.

 

1 hour ago, johnturley said:

, I would probably also need a new laptop with a faster processor.

Yes, you need quite a lot of RAM and importantly USB3. Also quite a decent sized hard drive - some nights I get through 100+GB.

My current laptop I bought used for £250 - it has 500Gb, 8Gb RAM and USB3. 

 

 

[johnturley]

2 hours ago, Nik271 said:

Oops I missed something very important:  you are taking videos, not still photos with the 6D.

The problem is that the video is only HD i.e. 1080 by 1920  while you sensor is 3600x5500 px

So the camera software is 'resampling' the image in effect combining several of you original pixels into one.

The conversion ruins the image scale: Now your sensor has been converted to fewer and so HUGE pixels, looks like 3 times larger. This means that in the image everything will be 3 times smaller than what I said earlier. No barlow can fix that big difference. (unless there are 6x barlows but they probably have aberrations)

Perhaps instead of video you can take 100 still photos, center and crop the planet in PIPP and stack these? Its too few frames but will give larger images than the video in prime focus.

    

Yes, the Mars and Jupiter photos are taken from approximately 1 minute MVI video runs and have 25 frames per second.

I'll give it a go with still images, but I don't think the image size would be any different, attached are a photos of Venus and Mercury taken earlier this year, this time using eyepiece projection with a 12.5mm Plossl eyepiece, and which were taken from single images (not stacked), and processed in Adobe Lightroom. Mercury is very tiny. 

Edited September 16, 2020 by johnturley

[Tommohawk]

Just as a BTW there's a calculator for magnification with eyepiece projection here. You can then check how differrent a Barlow/Powermate might be

[johnturley]

44 minutes ago, Tommohawk said:

Just as a BTW there's a calculator for magnification with eyepiece projection here. You can then check how differrent a Barlow/Powermate might be

Thanks for the information Tommohawk. 

I forgot that the degree of amplification you get using eyepiece projection partly depends upon the distance between the eyepiece and the camera sensor. Where I currently have the eyepiece positioned, the distance is about 120mm, so according to the calculator, this gives f57 and an effective focal length of about 20,000 mm, so the image would be much smaller even with a 4x Barlow. With my existing eyepiece projection tube I can increase the distance to a maximum of about 140mm, which gives f67 or an efl of about 24,000mm, I think however I've tried this position before, and found that I couldn't reach focus.

John 

[Nik271]

 

3 hours ago, johnturley said:

Thanks for the information Tommohawk. 

I forgot that the degree of amplification you get using eyepiece projection partly depends upon the distance between the eyepiece and the camera sensor. Where I currently have the eyepiece positioned, the distance is about 120mm, so according to the calculator, this gives f57 and an effective focal length of about 20,000 mm, so the image would be much smaller even with a 4x Barlow. With my existing eyepiece projection tube I can increase the distance to a maximum of about 140mm, which gives f67 or an efl of about 24,000mm, I think however I've tried this position before, and found that I couldn't reach focus.

John 

 

These are massive focal ratios indeed, the images must be dim and hard to focus. Fundamentally eyepiece projection is no different from prime focus with a barlow, both serve to increase the size of the image on the sensor. The reason your images are too small is the change of resolution when recording video: from 20Mp sensor for photos to fullHD video which is a sad 2Mp. That's 10 fold loss of resolution so the image covers 10 times fewer pixels and is about 3 times smaller by diameter than what it could be.

You need to find a way to record video at 1-1 pixel resolution. Some people recommend recording the 5x zoomed live view  from the camera back screen using Backyard EOS but that will still involve a laptop. If you are set on a self-contained camera you could try the old 550D which records video from just a small central portion of the sensor, or the recent cameras which have the 4K video 'crop' mode like Canons M50 or 250D.

Nikolay

 

[johnturley]

51 minutes ago, Nik271 said:

 

 

These are massive focal ratios indeed, the images must be dim and hard to focus. Fundamentally eyepiece projection is no different from prime focus with a barlow, both serve to increase the size of the image on the sensor. The reason your images are too small is the change of resolution when recording video: from 20Mp sensor for photos to fullHD video which is a sad 2Mp. That's 10 fold loss of resolution so the image covers 10 times fewer pixels and is about 3 times smaller by diameter than what it could be.

 

Nikolay

 

Hi Nikolay

The image size is exactly the same regardless of whether I do a HD video, or take a single shot photo as with my images of Venus and Mercury.

The 'Live View' image is actually reasonably bright in the case of Venus, Jupiter, and Mars, but dim in the case of Saturn and Mercury, making focusing difficult.  In addition when I tried to take a video of Saturn, the MVI image appeared to be too dim for Registax to be able to find Align Points., and enhance the image.

John 

[Nik271]

Dear John,

You said:

> The image size is exactly the same regardless of whether I do a HD video, or take a single shot photo as with my images of Venus and Mercury.

 

The image probably looks the same on the screen because most view programs stretch it to fill all available space within certain limits of dots per inch screen resolution.

However when you crop the difference becomes apparent. I took the photo of Jupiter you posted (which is 1080x1920) and cropped it to 448x448:

                                          

 

For comparison I then rescaled your image as if was taken at the full resolution: 3600x5400 and again cropped at 448x448:

 

 

So if you can image at your full resolution you will certainly get larger planetary images (in terms of pixel size) and there is no need to go all the way to F57, something like F15-F20 will do. Most of the spectacular images that people post are with telescopes like a  14inch SCT at F20. A well-collimated 14 inch Newtonian working at F20 should produce very similar images. 

Nikolay

 

[johnturley]

5 minutes ago, Nik271 said:

 

                                          

 

For comparison I then rescaled your image as if was taken at the full resolution: 3600x5400 and again cropped at 448x448:

 

So if you can image at your full resolution you will certainly get larger planetary images (in terms of pixel size) and there is no need to go all the way to F57, something like F15-F20 will do. Most of the spectacular images that people post are with telescopes like a  14inch SCT at F20. A well-collimated 14 inch Newtonian working at F20 should produce very similar images. 

Nikolay

 

Nikolay

How do you crop to enlarge the image, I've asked this question several times before, but so far nobody has been able to provide an answer, and I was coming to the conclusion that it was not possible. 

To be honest I'm not convinced that using a Barlow would give better results than eyepiece projection, which is one reason why I'm a bit reluctant to spend £300+ on a 4x Powermate. 

John 

John 

[Nik271]

As far as I see there is no real difference between eyepiece projection and prime focus with barlow: both serve the purpose to project an image circle on the sensor.

So if you are happy with eyepiece projection just keep going.

The only intrinsic property of the image size is how many pixels does it have. All the rest is a function of what pixel per inch density the image is diplayed on the screen or printed on paper. Web browsers always try to fit in the largest possible space in the screen subject to the limitation of the web page and screen resolution.

So lets take my laptop: it has screen resolution at 267pixels per inch which is close to what a fine photo print resolution should be for close viewing.

If I look at the 448x448 image of Jupiter in windows photo viewer and choose 'view actual size' this will be about 1.5 inch by 1.5inch. Small but very crisp.

Normaly the photo viewers and web browsers dispaly at 100pixels per inch as it is good enough for general viewing purposes. This is why the photos above look much bigger on my browser, they seem to be about 3x3 inches in size, so the web browser has chosen to show them at perhaps 120pixels per inch.

 

You can certainly rescale you images, any good photo program has this function. I use Gimp where it is under the image tab. If an image is very sharp in its default size you can rescale it to larger number of pixels as long as you happy with the level of detail. Conversely a fuzzy looking image will look better if is downscaled. For big planets like Jupiter I'm happy if they cover about 100-200 pixels with my gear. 

There is also a method for producing a higher resolution images from many lower resolution ones called 'drizzling' This is done when you can take multiple images which are lower resolution than you telescope can provide and then use software to get a higher resolution image by exploiting small variations of the image data. It was pioneered for Hubble since its cameras are of lower resolution than the optics can deliver. I don't know much about this but there should be people on this forum who know how to use it.

Nikolay

[johnturley]

5 minutes ago, Nik271 said:

There is also a method for producing a higher resolution images from many lower resolution ones called 'drizzling' This is done when you can take multiple images which are lower resolution than you telescope can provide and then use software to get a higher resolution image by exploiting small variations of the image data. It was pioneered for Hubble since its cameras are of lower resolution than the optics can deliver. I don't know much about this but there should be people on this forum who know how to use it.

Thanks for the info.

There is drizzle function in Registax, which enables you to enlarge the image, however when I tried this the results were awful.

John 

[Tommohawk]

1 hour ago, johnturley said:

How do you crop to enlarge the image, I've asked this question several times before, but so far nobody has been able to provide an answer, and I was coming to the conclusion that it was not possible. 

In your image editor there should be a way to crop the image. You mentioned Lightroom - see this help page

1 hour ago, johnturley said:

To be honest I'm not convinced that using a Barlow would give better results than eyepiece projection, which is one reason why I'm a bit reluctant to spend £300+ on a 4x Powermate. 

I agree.... a powermate, whilst excellent, isn't your priority. You need to sort your image capture first - find a way of doing uncompressed video, stack multiple stills, or bite the bullet and go for a planetary camera.

[johnturley]

39 minutes ago, Tommohawk said:

In your image editor there should be a way to crop the image. You mentioned Lightroom - see this help page

I agree.... a powermate, whilst excellent, isn't your priority. You need to sort your image capture first - find a way of doing uncompressed video, stack multiple stills, or bite the bullet and go for a planetary camera.

Thanks for the info, I've had another look in Lightroom and had some success with the crop function (see attached image), it wasn't obvious, and this is the main thing I been trying to find out how to do - Many thanks!. With the amount of detail I have captured on Mars, I don't think it would look better if zoomed in any further. 

I don't think that there is a method of doing uncompressed videos with my Canon 6D, unless there is a 6D expert who can tell me otherwise, and stacking multiple stills would be too fiddly and time consuming for me, and wouldn't alter the image size anyway. 

[ShrewView]

Hi John

I think the answer to your original post title is, if you are looking to get the most out of the aperture you have, then a dedicated camera is the way to go. If you don't want to spend any further, then eyepiece projection and lots of video can certainly get some pleasant results. That's a good picture of mars you've got. Much better than i did with my 6D.

Back in 2016 i was using my 6D on my 8.5 inch newt to image the moon and planets because that was all i had. My results on the moon and even Jupiter were pleasing enough, but on smaller targets the 6ds pixels are too large to catch the detail even my 8 inch can capture. Don't get me wrong the 6d is great, and excels for the deep sky stuff where it can soak up faint light sources with longer exposures. Anyway, wanting to get close to the best i can from my scope, i bit the bullet and bought a zwo224. Yes there's a cable and laptop to contend with but the images are much better.

For the price of the powermate you considered you could buy a dedicated planetary camera and a cheaper barlow to connect it with and probably a filter to aid contrast, which would give you much more detailed images than your canon can.

I think I've got a way to go yet with my processing skills but the attached images show the difference between the two approaches with the same 8 inch scope. One in 2016 with the 6d and the other last night with the 224 and a couple of barlow elements stacked. 

If you are considering spending money to get the most out of your scope then i'd personally recommend swapping to the dedicated camera. Good luck with whatever you decide. 

 

Mars 2016 - Copy.tif Mars_2019 - Copy.tif

[johnturley]

I think that I'm inclined to concentrate on getting the best out of my existing system at least for now, I'm just put of by the problems associated with a trailing USB cable in my observatory shed in the dark, rather than the cost of a new planetary camera and laptop. Its a pity that, at least as far as I am aware, no wireless planetary cameras available at present, maybe these will come in time. 

John

[Tommohawk]

I just thought I'd post my best effort with my Canon 550D DSLR to show what can be achieved - IF you can use uncompressed video. This was done with a Skywatcher 200PD, so a budget scope, with a 5x powermate, 550D using video crop mode It's clearly not to the same standard as a dedicated planetary camera but it's fair I would say.

I looked at Magic Lantern software -  which apparently can be downloaded to your camera (I've never tried this) - but I dont this it will help. It appears that the 6D just wont do uncompressed or RAW movies at all - unless someone knows better?

Edited September 17, 2020 by Tommohawk

[Davey-T]

On 15/09/2020 at 16:29, johnturley said:

Is there any way round this problem, or do I need to get a Astro type camera with a suitable sensor size,

I'm pretty sure the 6D has crop mode 640x480 one to one pixel video, I use this with the 60Da to get a sensible image, can make it a bit tricky getting the planet centred on the sensor though.

Dave

[johnturley]

3 minutes ago, Tommohawk said:

I just thought I'd post my best effort with my Canon 550D DSLR to show what can be achieved - IF you can use uncompressed video. This was done with a Skywatcher 200PD, so a budget scope, with a 5x powermate, 550D using video crop mode It's clearly not to the same standard as a dedicated planetary camera but it's fair I would say.

I looked at Magic Lantern software -  which apparently can be downloaded to your camera (I've never tried this) - but I dont this it will help. It appears that the 6D just wont do uncompressed or RAW movies at all - unless someone knows better?

Superb shot, like you say I don't think that there is a method of doing uncompressed videos with my Canon 6D, unless there is a 6D expert who can tell me otherwise, and show me how to do it. 

John