M33

[Rodd]

 

TOA 130 at F7.7 and STT-8300. The image consists of 5 hours each of RGB in 10 min subs and 6 hours of lum in 20 minute subs.  I was inspired by Gnomus's amazing M33 made with rather limited data. I do not feel I have done it justice and am really debating on whether it is worth collecting the Ha for this image.  Will it benefit from the Ha, or will the Ha add to the confusion.  Opinions welcomed.

 

 

[Allinthehead]

There's some great detail in that Rodd. Wonderful 

[Rodd]

1 minute ago, Allinthehead said:

There's some great detail in that Rodd. Wonderful 

Thanks Richard.  I am on the fence about the Ha.  It seems a bit chaotic to me.  Never satisfied I guess.  Here's a final tweek until I get the Ha (I guess I am committed after all)

[tooth_dr]

Very nice Rodd. I think until you actually attempt to image this yourself do you realise how difficult it is! This is very well resolved. 

[Hayduke27]

Beautiful image!  I love M33, this does it justice!

 

[Rodd]

3 hours ago, tooth_dr said:

Very nice Rodd. I think until you actually attempt to image this yourself do you realise how difficult it is! This is very well resolved. 

Thanks Tooth Dr.  

Rodd

[Rodd]

3 hours ago, Hayduke27 said:

Beautiful image!  I love M33, this does it justice!

 

Thanks Hayduke.  

Rodd

[wimvb]

Great image, Rodd.

The only comment I have concerns the brown and blueish patches on the left in the galaxy. I haven't seen this in other images of this target. Otoh, there's a very nice small reflection nebula just above these patches, which you brought out very well.

[ollypenrice]

I really like most of this image but what's going on with the brown patch lower left in the disk? Just to its right there's a similar shaped magenta patch suggestive of artefact. 

Ha does bring out some nice features if not overdone.

Olly

Edit. Sorry, I missed Wim's similar observation.

[Rodd]

7 hours ago, wimvb said:

Great image, Rodd.

The only comment I have concerns the brown and blueish patches on the left in the galaxy. I haven't seen this in other images of this target. Otoh, there's a very nice small reflection nebula just above these patches, which you brought out very well.

Thanks Wim.   See my discussion below regarding "strange" things in this image

4 hours ago, ollypenrice said:

I really like most of this image but what's going on with the brown patch lower left in the disk? Just to its right there's a similar shaped magenta patch suggestive of artefact. 

Ha does bring out some nice features if not overdone.

Olly

Edit. Sorry, I missed Wim's similar observation.

I am glad you asked!! Attached is blue sub and a lum sub.  Both the blue and lum subs all have this V shaped patch.  Also attached is a master blue flat that shoes a similar V shaped blotch--but the V shaped blotch on the subs and the flat does not line up.  And they are of different dimensions.  The V shape blotch on the flat actually removes a similar blotch on the subs--just like it removed the blotch at the top.  But the V shaped blotch on the subs is oinly really visible on Lum subs (and blue after meridian flip).  Also attached is a blue sub after Meridian flip that has the same V shaped blotch as is on the flats. Any Idea what is going on?

1) The V shaped patch does not rotate with the camera

2) V shape visible on subs does not appear on flats

3) V shape visible on flats removes a much fainter V shape from the subs (not very visible on subs--but after meridian flip it can be seen faintly)

master Blue Flat

Lum sub:  v shaped blotch seems to be same as in flat

Blue sub--V shape definitely not the same as on flat--flat does not correct

Blue sub after Meridian flip showing same patch as on flat

 

Please Help!!!

[Rodd]

So--upon further inspection--The V that will not calibrate out is only on the Blue subs.  But it does not appear on Blue flats taken the same night.

 

There bare 2 issues--1 is the V shaped blotch on blue subs that does not appear on any other filter or in any other flats.

The other issue is the V shaped blotch that appears on all subs and on all flats IS removed on all filters except for the blue subs where it is not wholly removed.

Here is an uncalibrated blue sub, and the same blue sub calibrated.  Note that all smudges are gone (except the mysterious V shape)--but in the calibrated sub you can see the V shape that is on the flat--almost like the flat did the opposite of what it was supposed to do--but only for this area--the smudge at top IS gone).  The whitish v shape next to the dark V shape in the calibrated sub is the V shape that was on the flat magnified somehow.

Rodd

Uncalibrated blue sub

Calibrated blue sub

[DaveS]

I'd have a go at getting some HII data to bring out some of the star-forming regions.

I found out the hard way just what a thankless task it is to image this under LP. In fact I've put this on hold until I move to a darker location.

[Rodd]

28 minutes ago, DaveS said:

I'd have a go at getting some HII data to bring out some of the star-forming regions.

I found out the hard way just what a thankless task it is to image this under LP. In fact I've put this on hold until I move to a darker location.

I had a rare week of exemplary weather.  M33 was high in sky.  For once my broadband data looked decent.  EXCAPT FOR THAT BLUE SUB V.  Still don't know what would be in sub but not in flat--moisture maybe?  Seems an odd shape for frost or dew.

Rodd

[ollypenrice]

A maddening problem and I'm sorry you have it! Pre and post flip is irrelevant unless the flip dislodged the guilty particle(s). (When you do a flip the entire light path is inverted so the same flats apply.)

I wonder if something wandered into the light path between the imaging run and the the blue flat? If so, just try a different flat (lum?) on your blue set.

You know the look of the artefact so try different flats/no flats till you get the best result per channel. There's a formula out there on the net for calculating the distance of the offending particle from the chip but I have never used it. Is the particle on a filter or on the chip window? And is it mobile? I don't know.

I do know that I'd attack this one in Photoshop with some hope of fixing it.

Olly

[Davey-T]

Very strange Rodd, looks almost insect like, maybe a very very small fly got in somehow, would explain it appearing, moving and disappearing ?

Dave

[Rodd]

1 hour ago, ollypenrice said:

A maddening problem and I'm sorry you have it! Pre and post flip is irrelevant unless the flip dislodged the guilty particle(s). (When you do a flip the entire light path is inverted so the same flats apply.)

I wonder if something wandered into the light path between the imaging run and the the blue flat? If so, just try a different flat (lum?) on your blue set.

You know the look of the artefact so try different flats/no flats till you get the best result per channel. There's a formula out there on the net for calculating the distance of the offending particle from the chip but I have never used it. Is the particle on a filter or on the chip window? And is it mobile? I don't know.

I do know that I'd attack this one in Photoshop with some hope of fixing it.

Olly

Definitely strange.  I was able to fix the flat not removing the normal smudge V from the blue channel (clone stamped it out on the master flat--seemed to work).  But the wider V that is only on the blue subs can't be removed at all.  The only thing I know that flats won't remove is dew or frost issues.  Since its only on blue subs I don't think its that.  I tried using different flats--but none fixed it--and they created other issues.Don't know.  I am contemplating redoing the blues tonight.  Meantime--heres the image with Ha added.  I re[processed it from scratch.  Not totally satisfied with the Ha leakage around the nebulas.  I'll have to work on that. Not sure how.

[Rodd]

47 minutes ago, Davey-T said:

Very strange Rodd, looks almost insect like, maybe a very very small fly got in somehow, would explain it appearing, moving and disappearing ?

Dave

I was thinking the same thing.  What ever it was stayed in place for the 7 hours or so that I was using the blue filter.  I would like to know what ever it is (hopefully was) why won't flats remove it?  and why didn't the flat remove the normal V shaped smudge on the blue subs--it almost made it worse.  Look at the HaLRGB image above, and note the blue artifact is no longer next to the red/brown artifact created by the V.  I had to clone stamp that out of the master flat (or clone stamp it in actually).  I hope I don't have a problem with something!

Rodd

[Rodd]

Here is a blend between the last one and the new one with Ha.  I think I like this one better.

 

 

[ollypenrice]

I've honestly seen a bug moving in real time around my flats, changing place from sub to sub, and been unable to find any trace of it whatever on the hardware once dismantled. I find it hard to believe that this was possible but it was.

Olly

[Davey-T]

Let's hope it wasn't breeding in there

Dave

[Rodd]

Last one lest I succumb to the marathon.....not going to happen.  but this one resolves the core a bit. Should I redo the blues, or leave well enough alone and conquer some other target?

[wimvb]

You've got a super image, just too bad it has this flat issue. You could try using the blue master flat to create a mask that only leaves the v-shape open. Then use curvestransform in PI to gently change the colour balance/intensity. You should be able to minimise it some extent. Unfortunately, the provided blue flat didn't have the same geometry as the final image, so I couldn't test this.

I would definitely open the camera/filter wheel and clean the optics with a rocket blower.

I downloaded your latest version and had a go at it in PixInsight. Only quick n' dirty; but it shows that with some care, you can clean this up in processing.

• Extract R channel and blur. Then use this as reference to create a range mask that only includes the galaxy.
• Applied the range mask protecting the background.
• Used colour saturation to lower the colour saturation of the discoloured areas.
• SCNR to the whole image, since the background looked a bit green.

[Rodd]

2 minutes ago, wimvb said:

You've got a super image, just too bad it has this flat issue. You could try using the blue master flat to create a mask that only leaves the v-shape open. Then use curvestransform in PI to gently change the colour balance/intensity. You should be able to minimise it some extent. Unfortunately, the provided blue flat didn't have the same geometry as the final image, so I couldn't test this.

I would definitely open the camera/filter wheel and clean the optics with a rocket blower.

I downloaded your latest version and had a go at it in PixInsight. Only quick n' dirty; but it shows that with some care, you can clean this up in processing.

• Extract R channel and blur. Then use this as reference to create a range mask that only includes the galaxy.
• Applied the range mask protecting the background.
• Used colour saturation to lower the colour saturation of the discoloured areas.
• SCNR to the whole image, since the background looked a bit green.

Thanks Wim--probably could do it without redoing then huh?  I will see if I get the same V shape in blue subs at least.  Besides, M1 (my next project) doesn't rise until around midnight so.  Have to do something!  I will have a look over your suggestions and see if I can figure it out probably take me longer than redoing them   I just hope I don't have some kind of permanent issue.  Still confused as to what it could be.  

Rodd

[Rodd]

20 minutes ago, wimvb said:

You've got a super image, just too bad it has this flat issue. You could try using the blue master flat to create a mask that only leaves the v-shape open. Then use curvestransform in PI to gently change the colour balance/intensity. You should be able to minimise it some extent. Unfortunately, the provided blue flat didn't have the same geometry as the final image, so I couldn't test this.

I would definitely open the camera/filter wheel and clean the optics with a rocket blower.

I downloaded your latest version and had a go at it in PixInsight. Only quick n' dirty; but it shows that with some care, you can clean this up in processing.

• Extract R channel and blur. Then use this as reference to create a range mask that only includes the galaxy.
• Applied the range mask protecting the background.
• Used colour saturation to lower the colour saturation of the discoloured areas.
• SCNR to the whole image, since the background looked a bit green.

Oh yeah--one of the V shapes that caused the trouble is not on the flats.  Strange.

Rodd

[wimvb]

My guess is that it's a fibre of some sort that's stuck on glass close to the sensor. That's why I recommend removing the camera for closer inspection.

4 hours ago, ollypenrice said:

...

There's a formula out there on the net for calculating the distance of the offending particle from the chip but I have never used it. Is the particle on a filter or on the chip window?

...

 

Olly

I found this one (but haven't confirmed it):

distance = pixelsize x f-ratio x shadow diameter in pixels

where pixelsize is in microns, and shadow diameter is the width of the dust particle (ideally a small round speck), not the lenth.

This would mean that if you have a pixelsize of 5 microns, an f-ratio of 6 and the width of the dust particle's shadow is 100 pixels, the particle is 5*6*100 = 3000 microns (3 mm) from the sensor.

If the cover glass of the camera (5 micrometer pixels, f/6 system) is 10 mm from the sensor, then a dust particle on the glass would show up as a 330 pixels wide blob in the image.