M81 Bode's Nebula

[IvanT]

My first attempt at this one. 10 x 210 sec subs, guided in RA only as I've a really annoying problem with EQMOD & PHD pulse guiding that I don't understand. It periodically takes a huge jump and sounds like a gear slipping.

I did take 5 darks but I think there was some light leaking in through the camera finder so couldn't use them. I took 10 flats but when applied they seemed to make the (what I assume to be) vignetting worse.

So this is just the 10 subs stacked, no processing.

Am I correct about the vignetting?

With the 35 mins of detail in this, should I be expecting to see more or do I just need to add a lot more data?

The tracking seems ok and the star stayed almost directly in the PHD cross hairs for the full 35 mins, so am I correct in saying that tracking in DEC only becomes essential if polar alignment is bad?

Thanks

[PhoenixRising]

I'm no expert, but to me its looking very good. I think you need more data, but for 35 minutes, that image is looking ace.

You need to be adding 20+ flats and darks to your images, although you'd have to ask someone more knowledgable about the flat problem you're having. How are you taing them?

Although my own processing has been frustrating, I noticed the difference in using flats as soon as I tried them (first time this week.)

[IvanT]

Thanks

You need to be adding 20+ flats and darks to your images, although you'd have to ask someone more knowledgable about the flat problem you're having. How are you taing them?

I posted some detail on my flats here, in summary:

This is a test flat using Av 1/200, ISO 800, light panel on the end of the scope with my modded 1000d (LP filter & coma corrector in place). I found that 1/200 was the closest setting to get the exposure histogram in the middle of the graph.

The problem it seems is my flats are too flat what method do you use?

I'm trying to get all the niggles & questions ironed out before this winter.

[JohnnyD333]

Hey Ivan,

What ISO are you using for the subs? For the amount of subs you took you should see quite a bit of detail. Phoenix is right about taking many flats. Darks are very important although Im not sure why you camera is leaking light. If you can get the darks a higher ISO is an option which will provide more detail, as long as you have the darks to get rid of the noise. I attached a single sub of 5 minutes at ISO 1600 with my Nikon, I added 5 flats and 3 darks. I think once you get you guiding fixed and can experiment, you should try different settings, all cameras have a sweet spot and you will certainly find it. Good luck:)

[JohnnyD333]

A good way to take flats is the white t-shirt method. Point your scope at an even light source and stretch a white shirt over your scope. Olly mentioned this method and it works pretty well.

[IvanT]

What ISO are you using for the subs

Sorry, it was at ISO 800. Thanks for posting the comparison, I must admit, I was expecting more from 35 mins even without the calibration frames.

Sent from my iPhone using Tapatalk

[roundycat]

you shouldn't really be using less than ten darks. You can get away more safely with the same number of flats as they are of a much higher exposure level so intrinsic noise is less of a problem. The brightish centre in your picture is a hot spot. If you were using a camera lens you would see it recede as you stop down. A vignette is really sharply cut off corners, very obvious.

If the flats are undercompensating try shooting them at a different exposure level. The software used for flat fielding can have its own effect too, are they normalised?, over what area?, median combine or average?

Dennis

[IvanT]

I was simply loading lights, flats, darks etc into DSS a letting it sort it out. The combination method was set to average I think. The shots were all taken at prime focus with a skywatcher coma corrector & light pollution filter attached.

Could the A3 EL Panel I'm using be simply too bright and need toned down somehow?

Could you elaborate on the hot spot please? I don't think it's related to the recent modding I performed on the camera as this type of problem was in images I took before the mod. Is a hot spot something that is common to the 1000d?

[roundycat]

The hot spot is a lens effect, hence it will disappear as you stop down and the aperture becomes smaller. Due to the cos^4 law I think. It has nothing at all to do with the camera.

The panel may need some paper as a neutral density filter. Best flat exposures would be 1-3s.

Dennis

[IvanT]

Ok, so as I'm shooting at prime focus does that mean that this can only be vignetting?

I'll look into neutral density filtering the panel.

Thanks

[roundycat]

You could shoot with a telescope and not be able to adjust the aperture as there is no diaphragm. Use a camera lens and you can adjust the aperture, what ordinary photographers call the 'f' number, strictly the 'f' ratio.

In either case if the lens or scope connects direct to the camera you are at prime focus. The native focal length of the lens is operative. Aperture or 'f' ratio in camera lenses has nothing to do with being at prime focus.

None of this has anything to do (directly) with vignetting. A vignette is a very obvious effect mainly towards the corners of the picture. It is almost always down to the lens having insufficient covering power for the size of the negative or chip area. You hardly ever see this in astrophotography as people tend to use relatively small camera chips that do not reach out to the edges of the lens' covering power.

What you do see a lot of is people referring to the vignette when there is no such thing present. Mostly what you see is a hot spot, another lens effect, always at its worst at full aperture. In the first picture you posted what you see is a hot spot, pure and simple. A good master flat will cure it instantly.

Dennis

[IvanT]

Thanks for that Dennis, that explains it very well. I guess it comes down to not always believing what you read, one mans vignetting is another mans hot spot. I'll mess with different settings with/without diffusing the panel and reprocess the image to see if I can find my counter hot spot sweet spot :-)

Could I ask a big favour of you please, would you upload a typical flat that you'd use to give me an idea of brightness. The ones I'm getting look quite blue idividually and having just opened the master flat produced by DSS I can see very apparent banding.

Sent from my iPhone using Tapatalk

[roundycat]

There is another saying that springs to mind, viz; one man's meat is another man's poison. That's how it is with flats, mine are not like anyone elses and what passes for a good exposure to make my flats work might not work for you. To get the best flat you have to make a master and then use it. If it is no good make another with a different exposure.

The attached shows a red master that I did recently for my STL11000 and FSQ106 combination. The STL shows some fixed pattern noise, seen as a ripple effect across the centre of the frame. This is obviously present in all the light frames as well as the dust spots, vignettes (yes! in this case it IS a vignette in the right hand corners) and other muck.

Flat fielding the light here shows the frame to be considerably cleaner (I think there was a reason for the dust spot not going in this example) but all the frames have been stretched alarmingly to show the problems. The individual flats were made to have an average adu value of 12,000. The histogram is from Photoshop and actually shows how little use it really is. Your camera shoots adu's not levels.

Edit: ignore the gradient to the left. That is not a function of flat fielding.

Dennis

[IvanT]

Ok, this has really been a flats baptism of fire. The light panel seemed to be giving off a blueish colour so I stretched a white T-shirt over it and redid the flats.

There is still signs of a hot spot in there but I think it's an improvement on the original. Don't think I'm quite there yet but thanks for all the helpful hints folks.

[roundycat]

the latest one is a bit light in the background but looking quite flat. There is a gradient from the left but nothing to do with the flats. I played around a bit but found some odd colour gradients that I didn't bother with. (bedtime!). I should have done some noise reduction.

I use a Neumann EL Panel and it is also bluish but this is down to what you expect, I think. You expect it to be white and compared to tungsten light white the ELP is very blue. I measured mine and found it very neutral.

Dennis

[spanky]

Ok, this has really been a flats baptism of fire. The light panel seemed to be giving off a blueish colour so I stretched a white T-shirt over it and redid the flats.

There is still signs of a hot spot in there but I think it's an improvement on the original. Don't think I'm quite there yet but thanks for all the helpful hints folks.

I hope you don't mind, but I just did a quick pass with Gradient XTerminator. It's excellent, you can get it from here: GradientXTerminator

[adamsp123]

I found for my 1000D the fridge was a good place to take darks, the main compartment is around 5C and the ice box 0C so I used that to get my darks library going, at ISO800 BTW. With those 2 temperatures I found that they were good for most nights of imaging although I did do a load at 10C one evening.

For flats I used a clean white T shirt over the scope, normally at dawn and set the camera to Av mode and fire away!

[IvanT]

I hope you don't mind, but I just did a quick pass with Gradient XTerminator. It's excellent, you can get it from here: GradientXTerminator

Wow, that has done a fantastic job, thanks for the link. I'll be getting that!

[IvanT]

I played around a bit

Can I ask what you did/used? You seem to have managed to extract a lot more colour than I was able to.

The gradient you mentioned is most likely down to my heavy handed (aka non-skilled) use of PS.

[IvanT]

I found for my 1000D the fridge was a good place to take darks

I've heard of others using the fridge for darks but I must admit to not understanding how it can work without more accurate temp matching, i.e. the temp the camera was exposed to on the night.

Maybe I'm wrong, but wouldn't 20 darks taken at say 5C in the fridge introduce noise due to the difference in hot pixels in the fridge dark compared to the hot pixels in the sub on the night, unless of course it was 5C outside?

Don't get me wrong, I'd be all for anything that would remove the time spent at the end of the night taking dark frames. Or maybe I'm getting to hung up on calibration.

As for the flats, what settings do you typically use, i.e.

auto WB,

shutter speed 1/200

ISO setting 800

any other settings you change when taking your flats?

[roundycat]

I took most of the gradient out using the Gradient tool in PS to make a selection. A few simple iterations of Curves to neutralise the background colour, introduce some more contrast and then darken the background. A couple of small increases of saturation and a limited amount of High Pass to enhance detail in M81. M82 didn't like it so I took out the High Pass over M82 with a splodge of layer mask.

That's as much as I can remember (it was yesterday!), but if you can improve a jpeg slightly it shows there is more there in the original TIFF.

The gradient is unlikely to have been put there by you. It is nigh on impossible to introduce a gradient with any of the tools in PS. What you will do is to exaggerate it quite a lot by normal contrast stretching. Once it is obvious it needs looking after.

Some will have you believe that poor flats or just normal flat fielding will introduce a gradient. I have yet to see it. I think this is down to people sometimes grabbing at straws and then the story goes round - 'gradients are caused by bad flats!'. Almost always it is an external light source such as the street lights, an industrial estate or the Moon that does it. Depending on the angle of your camera to the mounting plate that gradient could be at almost any angle and may even only affect one corner of the frame.

Dennis

[adamsp123]

Personally I don't worry about dead accurate temperature matching, if you start a run at 8C and over time the temp drops to 4C then IMO 5C darks are fine.

Now a 10C difference might be noticable, but in our climate I reckon 99% of works could be covered with Dark librarys of 10, 5, 0, -5C....

[IvanT]

but in our climate I reckon 99% of works could be covered with Dark librarys of 10, 5, 0, -5C....

Cool (excuse the pun)

So over a longish period of time you take darks of say 1 min, 1.5 mins all the way to your max likely exposure duration and you take them at varying temperatures using the fridge to set the temp point, then using the temp info from the RAW files stack with the appropriate set?

[adamsp123]

I worked out a set of exposure times I would typically use after a bit of trial and error, then I set up the camera connected to a netbook (outside the fridge!) and using the EOS software I ran a series of darks, 30 subs each, for a combination of sub lengths and 5C plus 0C.

Looking at my darks library I did them for 1,2,4,6 and 8 mins.

If I ran a set of subs of different length I would pick the nearest one to the sub time.

[IvanT]

Just checked my fridge, there is no temperature control on it, it self regulates, sounds like magic to me, probably the same guy that turns the light out when you close the door. The light is out BTW, Mr Schrödinger assures me it is.

Nice idea though. I think I'm just inventing excuses to buy myself a shiny new cooled CCD astro camera