Bubble Nebula

[Herzy]

I got 4 hours on the Bubble Nebula the other night and It is really ugly. It is way worse then I would've expected from 4 hours considering I could see it in my raw frames. The bubble is so close to the edge of the frame that there is no point in continuing and getting more data because I will have to crop the stack to get rid of the artifacts and it will cut into the nebula. The moon wasn't out and it was pretty dark that night so I don't understand how 4 hours could yield such bad results. Is this object just really faint or something? Or am I just not processing it well?

I know it's ugly, but it's the best I could do with this data. I had to overdo noise reduction to get rid of the noise and I had to back off on the gradient removal because it kept removing nebulosity. If someone would care to try out the stack just to see if it is my processing that is lacking here is the stacked file.

https://www.dropbox.com/s/geo1ffr4d3k7i36/Bubble Nebula - 4hr (Reject).tiff?dl=0

If you have the chance please try out this stack so I can know if it's my processing that is lacking or if it is my data that is lacking.

 

[Tim]

How long were the individual subexposures?

The results look as I would expect for a DSLR if the subs are short. At least you got M52 in there as well

[MARS1960]

I am going to be imaging this tomorrow, if my attempt looks anything like yours i will be very happy,

What exposure, iso, subs, equipment did you use?

[Herzy]

6 minutes ago, MARS1960 said:

I am going to be imaging this tomorrow, if my attempt looks anything like yours i will be very happy,

What exposure, iso, subs, equipment did you use?

 

8 minutes ago, Tim said:

How long were the individual subexposures?

The results look as I would expect for a DSLR if the subs are short. At least you got M52 in there as well

240 x 60s @ ISO 800. Is it just excessive read noise from low ISO? I'm confused as to how shorter exposures could affect the noise.

Thanks MARS. I mean, it's not bad. I just expected more from 4 hours.

Equipment is an AVX mount, C6-N, and Nikon D3200 (not modified).

[MARS1960]

I think that's half my problem, i keep expecting more but i'm only using an unmodded Canon 60D.

I'm 2hrs into the Iris nebula at the moment with my AT72ED using ISO 1600, hoping for 80 x 180s subs, i will see what it yields tomorrow but i'm not hoping for much, it's the first clear night for a while but the seeing here is awful, with lots of misty haze and sky glow.

[Herzy]

1 minute ago, MARS1960 said:

I think that's half my problem, i keep expecting more but i'm only using an unmodded Canon 60D.

I'm 2hrs into the Iris nebula at the moment with my AT72ED using ISO 1600, hoping for 80 x 180s subs, i will see what it yields tomorrow but i'm not hoping for much, it's the first clear night for a while but the seeing here is awful, with lots of misty haze and sky glow.

Well, once your finished post in here. I'm looking forward to seeing how others do with this same target. Everything on astrobin that I can find is like 18+ hours so I can't relate it to what I'm doing.

[MARS1960]

14 minutes ago, Herzy said:

Well, once your finished post in here. I'm looking forward to seeing how others do with this same target. Everything on astrobin that I can find is like 18+ hours so I can't relate it to what I'm doing.

Will do. I assume you used your 90mm MAK, if so i will use my SW 100ED without focal reducer, that should give me a similar if not slightly wider FOV.

[wimvb]

Herzy,

I think that the problem with your image is the short exposure. Try increasing that as much as you can. Generally lower iso settings produce less noise, but also less signal. More subs does nothing to increase signal. It can only decrease noise, but not lower than the read noise. To get better subs, you need more signal = longer exposures.

[Tim]

+1 ^^^

[Astrosurf]

Is your camera modded? If not, the camera's filter will block the Hydrogen Alpha (Ha) red photons. As this neb is an emission type, Ha is where most of the photons come from. You did well to capture what you did. It's a nice image. Nice star colour.

Alexxx

[Herzy]

8 hours ago, wimvb said:

Herzy,

I think that the problem with your image is the short exposure. Try increasing that as much as you can. Generally lower iso settings produce less noise, but also less signal. More subs does nothing to increase signal. It can only decrease noise, but not lower than the read noise. To get better subs, you need more signal = longer exposures.

My mount can't handle more then 60s, but I could try out guiding.

So more integration = less noise and longer exposure length = more signal?

If I understand what you're saying I was doing perfectly fine with removing the noise, but the signal isn't strong enough because I'm not capturing the necessary photons?. 

I get the logic, but it still confuses me.

If it's the same integration, in the end, the same amount of signal would've been captured. "Signal" is just a stream of photons. I'm still capturing every photon, just in smaller amounts per picture.

[Herzy]

2 hours ago, Astrosurf said:

Is your camera modded? If not, the camera's filter will block the Hydrogen Alpha (Ha) red photons. As this neb is an emission type, Ha is where most of the photons come from. You did well to capture what you did. It's a nice image. Nice star colour.

Alexxx

Unfortunately it's not modded. That's an upgrade I hope to get soon.

[wimvb]

2 hours ago, Herzy said:

My mount can't handle more then 60s, but I could try out guiding.

So more integration = less noise and longer exposure length = more signal?

If I understand what you're saying I was doing perfectly fine with removing the noise, but the signal isn't strong enough because I'm not capturing the necessary photons?. 

I get the logic, but it still confuses me.

If it's the same integration, in the end, the same amount of signal would've been captured. "Signal" is just a stream of photons. I'm still capturing every photon, just in smaller amounts per picture.

Yes

Put it this way. Suppose your target is so weak that you only detect 1 photon (= 1 electron) per minute. At the same time, your read noise is also 1 electron per frame. Each frame will then contain ON AVERAGE 1 signal electron and 1 noise electron. There's no way to distinguish the two, S/N = 1. No matter how many subs you take, each has a S/N = 1, and there may or may not be a signal electron in a sub.

With a 5 minute exposure, you get 5 signal electrons and 1 noise electron per frame S/N = 5. Each sub has guaranteed signal that is real.

When you stack you can either combine the average or median pixel value of all your subs.

Suppose you use median. This means that from all the subs, one is taken to represent the target in the final image. This is done for each individual pixel. But each pixel in the final image is from just one of your subs. So each pixel has received the same exposure as that one sub. Taking the median is just a way of selecting the pixel that (probably) describes the target best.

If you use average as combination, you have the same situation, only that the average value of all subs is taken. The average of all your exposures is still the exposure of every single sub. The only difference is that also the noise has averaged out: less noise.

The advantage of having less noise is that you can stretch the image harder until you hit the same noise level again as with fewer subs.

BTW, I left out signal noise and dark current noise, because that doesn't alter the argument, it only clouds the main argument.

Only if read noise is 0, would that be true. This is the advantage of modern low read noise, high sensitivity CMOS chips. You can get away with shorter exposures and more subs.

But in the end, you are still limited by the low number of signal photons hitting your sensor

Hope this clarifies some more.

Having an unmodded camera means that it'll be more difficult to get the photons from Ha targets. The camera will pick up some, but many are blocked by the camera filter, while at the same time sensitivity is low for these and longer wavelengths. How much Ha an unmodded camera picks up is very dependent on make and model.

With an unmodded DSLR, the targets you can get are very limited. You may try, but results will vary. During cold winter months, your camera will produce less noise, and you may get better results. As always: experiment

[Herzy]

1 hour ago, wimvb said:

Yes

Put it this way. Suppose your target is so weak that you only detect 1 photon (= 1 electron) per minute. At the same time, your read noise is also 1 electron per frame. Each frame will then contain ON AVERAGE 1 signal electron and 1 noise electron. There's no way to distinguish the two, S/N = 1. No matter how many subs you take, each has a S/N = 1, and there may or may not be a signal electron in a sub.

With a 5 minute exposure, you get 5 signal electrons and 1 noise electron per frame S/N = 5. Each sub has guaranteed signal that is real.

When you stack you can either combine the average or median pixel value of all your subs.

Suppose you use median. This means that from all the subs, one is taken to represent the target in the final image. This is done for each individual pixel. But each pixel in the final image is from just one of your subs. So each pixel has received the same exposure as that one sub. Taking the median is just a way of selecting the pixel that (probably) describes the target best.

If you use average as combination, you have the same situation, only that the average value of all subs is taken. The average of all your exposures is still the exposure of every single sub. The only difference is that also the noise has averaged out: less noise.

The advantage of having less noise is that you can stretch the image harder until you hit the same noise level again as with fewer subs.

BTW, I left out signal noise and dark current noise, because that doesn't alter the argument, it only clouds the main argument.

Only if read noise is 0, would that be true. This is the advantage of modern low read noise, high sensitivity CMOS chips. You can get away with shorter exposures and more subs.

But in the end, you are still limited by the low number of signal photons hitting your sensor

Hope this clarifies some more.

Having an unmodded camera means that it'll be more difficult to get the photons from Ha targets. The camera will pick up some, but many are blocked by the camera filter, while at the same time sensitivity is low for these and longer wavelengths. How much Ha an unmodded camera picks up is very dependent on make and model.

With an unmodded DSLR, the targets you can get are very limited. You may try, but results will vary. During cold winter months, your camera will produce less noise, and you may get better results. As always: experiment

Thanks. This is making much more sense now.

You seem very knowledgable in the subject so I will ask you what I asked some others in a different thread.

I have a few options. 

1) Buy a modified canon camera - this would help on Ha emission nebula, but not much else.

2) Begin guiding - I don't have a laptop so I would be guiding off my desktop computer.

3) Buy a short FL refractor - I'm currently imaging with a big clunky Newtonian that catches wind and had a long FL. 

All of these options have their advantages. Which one do you think would be my best bet for improving my images? I'm leaning towards guiding with what you said earlier about signal vs noise because they would allow me to jump from 1 minute to 10 minutes.

Let me know what you think..