Combining Images from different focal lengths?

[WolfieGlos]

Due to limitations in my area with buildings and a pesky streetlight partially blocking the south-east to southern sky, it makes imaging targets in areas such as Orion, Sagittarius, etc, very difficult and time limiting. I ran a test a few nights ago and with the current conditions I can capture no more than 45 minutes on the Lagoon Nebula...which will only get shorter as it rises earlier (and is darker later). In fact, as it starts to rise earlier, it will be totally obscured by said streetlight even quicker. Currently I'd be looking at a multi-year project to try and justify imaging this target! Never mind then picking up M20 to make a mosaic.....

So what I'm looking to try and do is to gather photons faster. Upgrading to a faster scope is out of the question, given I've only recently done so! My setup is currently a modded Canon 800D + Starfield 102 on a HEQ5. I guide using the ZWO miniguider bundle, all operated via Stellarium and NINA.

Now I have the Starfield, I have a semi-retired Skywatcher 72ED, and also a stock Canon 77D, so I was considering whether to mount the 72ED on top of the Starfield. I would need to run a dovetail bar across the top of the SF (I have a spare that fits, it just needs a hole drilled), so in theory this could be done. I've tried to show it below in my superb mock-up in paint...

Both cameras have the same sensor, but the scopes are different focal lengths (570 vs 357 with reducers). Working out the weights, the OTA's with cameras and the guider comes to 7.8kg (without flatteners), so I should be within the HEQ5's limit of 11kg.

I imagine I would need to stack images from both separately (at least due to flats), but in theory I could then combine the stacks? Is this achievable? Are there any issues combining images taken at different focal lengths with the same camera sensor?

Also, can NINA operate two cameras at once, and control dithering?

Any help appreciated 🙂

45 minutes on M8 with l-enhance filter, AstroDeNoisePY and then mild stretches in Siril. Played about with wavelets as a test to sharpen. Outer shells just visible.

[Elp]

Your differences in FL should sort the main issue when mounting equipment parallel (you can get the same target into the fov and crop down). Though in the end I found it easier to mount two setups onto their own mounts if you have the means to, this also gives you flexibility to image different things as well as have room to breathe if you encounter technical problems, one can be imaging whilst the other is being sorted.

Stacking could be an issue due to pixel scales and fwhm of the stars between the two stacks may have differences, I haven't got it to work between different setups as the softwares refuse to play ball, usually layer them manually in PS or GIMP. It should be possible though, someone with more experience doing this can say ye or nay.

Edited June 17, 2023 by Elp

[ONIKKINEN]

You can stack all of the data to one image with Siril (many others too probably.

What you'll need to do is calibrate the datasets with their own calibration frames first and then move on with stacking. Choose a reference frame (best image of either set) and load that as the first image in Siril. This will make it the reference frame and during registration the image scale from this image is automatically applied to any other image you load. For example you choose a 102ed image as the reference frame, the 72ed data is then upscaled to the 102ed data. Or vice versa of course. Then just stack all of the images together.

However even if you can stack the images does not mean it all goes flawlessly without weighting the subs. The bigger scope will provide loads more signal to noise ratio than the smaller one, so stacking both datasets together without any weighting will result in an image that may be worse than just the bigger scope data alone. Which is why you should choose the "weighted fwhm" option in weighting for stacking. This will calculate weights based on star counts and star sizes, both being slightly worse in the 72ed will give that data a lower weight overall but probably still contribute to the image somewhat.

I think you can run 2 instances of NINA in the case when you do dual camera imaging, although have only read about that briefly so wont throw any more guesses on how that goes really.

[woldsman]

You could investigate RegiStar to combine images taken with different scopes: https://aurigaimaging.com/

[WolfieGlos]

21 hours ago, Elp said:

Your differences in FL should sort the main issue when mounting equipment parallel (you can get the same target into the fov and crop down). Though in the end I found it easier to mount two setups onto their own mounts if you have the means to, this also gives you flexibility to image different things as well as have room to breathe if you encounter technical problems, one can be imaging whilst the other is being sorted.

Stacking could be an issue due to pixel scales and fwhm of the stars between the two stacks may have differences, I haven't got it to work between different setups as the softwares refuse to play ball, usually layer them manually in PS or GIMP. It should be possible though, someone with more experience doing this can say ye or nay.

Thanks Elp. I didn't consider that, I do have a Star Adventurer which I used to run the 72ED on, but it needed the guider or else subs were limited in length (best I ever achieved unguided was 20 seconds with that FL). So potentially I could invest in another guidescope bundle and do this.

With regards to different stacks, I didn't mention this originally but I did consider (if possible) to stack them both separately, then stack the stacks and then run starnet to remove stars. I would then add back in the stars from the 102ED stack and crop to suit. The stars on the 72ED are terrible, so losing them wouldn't be the end of the world, I suppose a bit like some do with RGB stars on narrowband images.

21 hours ago, ONIKKINEN said:

You can stack all of the data to one image with Siril (many others too probably.

What you'll need to do is calibrate the datasets with their own calibration frames first and then move on with stacking. Choose a reference frame (best image of either set) and load that as the first image in Siril. This will make it the reference frame and during registration the image scale from this image is automatically applied to any other image you load. For example you choose a 102ed image as the reference frame, the 72ed data is then upscaled to the 102ed data. Or vice versa of course. Then just stack all of the images together.

However even if you can stack the images does not mean it all goes flawlessly without weighting the subs. The bigger scope will provide loads more signal to noise ratio than the smaller one, so stacking both datasets together without any weighting will result in an image that may be worse than just the bigger scope data alone. Which is why you should choose the "weighted fwhm" option in weighting for stacking. This will calculate weights based on star counts and star sizes, both being slightly worse in the 72ed will give that data a lower weight overall but probably still contribute to the image somewhat.

I think you can run 2 instances of NINA in the case when you do dual camera imaging, although have only read about that briefly so wont throw any more guesses on how that goes really.

Thanks Onikkinen. I always scan my subs (i.e. for trails) before committing them to Siril's stacking and I currently only use scripts for ease, but sometimes I do go back and check the graph, remove bad subs and then restack, so I'd have to get more involved with manually stacking. Which is fine, I just need to understand the whole process a bit better 🙂 I've done it a few times following Siril's guide on the Comet earlier this year.

In fact, I have just found some older data of M51 (that I thought I'd lost) that was taken on the 72ED, so I could try combining it with data I've recently captured on the 102ED.....so I might give this a go later this week and see what happens.

With regards to NINA, I think you're right. I was trying to see if one instance could operate two cameras, but a quick google search has come up with the same thing you mention; running two instances of NINA. If the stacking seems to work, I'll investigate this and see what it can do. 

13 hours ago, woldsman said:

You could investigate RegiStar to combine images taken with different scopes: https://aurigaimaging.com/

Thanks woldsman. Not heard of it before, but if all else fails I'll take a look at that software.

[Elp]

12 hours ago, WolfieGlos said:

stack the stacks

This is how I do it. You can stack multiple nights worth I think if the setup isn't altered in any way, but I don't risk the flats causing an issue so tend to stack each nights worth separately with their own calibration files (other than darks with the cooled camera, uncooled I have to take darks every time).