#1  
Old 21-09-2014, 02:50 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Seeing? Optics, Focus?, ???

I could use some advice. I'm kind of stumped. I never quite ran into this problem when I used my KAF8300 CCD on my TEC140. But, my Trius is giving me fits (and not such good ones at that!).

As you can see from the attached screen shot of CCDInspector, I had reasonable seeing (I presume) given all the FWHM values in the 2s. I used FocusMax and have a large sample of V curves. To my eyes blue looks shockingly poor, but, I cannot focus any better. The FWHM values make me think the images should look better. The .jpgs are in order of R G B and are at 160%. 5 min exposures with the blues showing very good aspect ratios. Aligned and stacked in CCDStack.

Is this seeing, optics, poor focus (but why if poor focus? I even ran v curves just with the blue filter (Baader) and cannot do better)?

I know Greg Bradley has complained about "bloated" stars using the Trius on his TEC180. Perhaps this is something similar. If so, it seems to be due to blue for some reason.
Attached Thumbnails
Click for full-size image (CCDINS.jpg)
162.3 KB68 views
Click for full-size image (Mean Red.BBS-00000012.jpg)
110.8 KB125 views
Click for full-size image (Mean green NN.00000006.jpg)
94.9 KB125 views
Click for full-size image (Mean blue. BBS- 00000009.jpg)
78.6 KB128 views
Reply With Quote
  #2  
Old 21-09-2014, 03:16 PM
JohnH's Avatar
JohnH
Member # 159

JohnH is offline
 
Join Date: Feb 2005
Location: NSW
Posts: 1,226
I am no expert here but had this issue with my 127ED, which, granted, is not a true APO, I had to re-focus for each (parfocal Astronomics type II) filter. It became tedious so I ended up automating the process using offsets from multiple calibration runs. I also note that the BLUE wavelength is more subject to atmospheric distortions and thus, ideally, should be shot with the target at the highest point, also a corrector or reducer in the imaging train may make things worse.

I could never remove this effect completely, so I settled for focusing the R and G to match the B star sizes rather than optimise all, I then got the best L frames possible - after all that is where the image resolution comes from, RGB just add colour.

It seems a limit of the refractive optics that the spot size is smaller for a longer wavelengths, I am not sure if that is what is dominating overall though, guess that depends on the arc secs per pixel.
Reply With Quote
  #3  
Old 21-09-2014, 03:32 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
If you look at the CCDI data you'll see that, on average, the blue subs are less bright than the red and green subs (Background).

When the subs get stretched the blue ones need to be stretched further and the blue channel stars bloat giving a blue halo. It's the same reason that Hubble palette images have magenta star halos.

Cheers,
Rick.
Reply With Quote
  #4  
Old 21-09-2014, 04:50 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by RickS View Post
If you look at the CCDI data you'll see that, on average, the blue subs are less bright than the red and green subs (Background).

When the subs get stretched the blue ones need to be stretched further and the blue channel stars bloat giving a blue halo. It's the same reason that Hubble palette images have magenta star halos.

Cheers,
Rick.
Rick,

I appreciate that you pointed this out. No doubt this is part of the equation. However, it doesn't exactly explain why the blue looks so much worse than anything I previously got with my KAF8300 CCD camera. Nor would it explain quite bloated bright stars in luminance.

I'm beginning to think that it might be due to the fact that TEC colour correction goes from 436 nm - 1000 nm, but the Baader transmits to ca 380 nm. Together with the ca 40% QE of the Sony chip at 380 nm might this be the reason?

Many thanks,

Peter
Reply With Quote
  #5  
Old 21-09-2014, 05:05 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Probably a combination of factors, Peter, as you say. The relative dimness of the blue subs will be down to factors such as camera QE, filter characteristics, etc. Atmospheric scattering has a bigger effect at the blue end of the spectrum as well. I attempted to help answer the question you asked: why your blue subs seem poor by comparison to the red and green. You didn't previously mention bloated stars in Luminance so I didn't attempt to address that issue...

Cheers,
Rick.
Reply With Quote
  #6  
Old 21-09-2014, 05:42 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by RickS View Post
Probably a combination of factors, Peter, as you say. The relative dimness of the blue subs will be down to factors such as camera QE, filter characteristics, etc. Atmospheric scattering has a bigger effect at the blue end of the spectrum as well. I attempted to help answer the question you asked: why your blue subs seem poor by comparison to the red and green. You didn't previously mention bloated stars in Luminance so I didn't attempt to address that issue...

Cheers,
Rick.
You are correct! I did mention Greg's experience but forgot to say that it was with luminance. He privately pointed me to an image in luminance where the large halo effect is quite evident.

Thanks!!

Peter
Reply With Quote
  #7  
Old 21-09-2014, 07:14 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by PRejto View Post

I'm beginning to think that it might be due to the fact that TEC colour correction goes from 436 nm - 1000 nm, but the Baader transmits to ca 380 nm. Together with the ca 40% QE of the Sony chip at 380 nm might this be the reason?

Many thanks,

Peter
that would be my guess. The APO could be spraying out a fair bit of a halo in the violet and still be in spec. Since the FWHM are pretty consistent, it is not likely to be seeing or focus.

The Trius could have more violet sensitivity than the 8300 for a variety of reasons - window material, AR coatings etc. as well as the high QE. Might be worth trying a minus-violet filter or fringe killer in conjunction with LRGB, which is a bit of a pain.
Reply With Quote
  #8  
Old 21-09-2014, 07:52 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Peter,
I think you can safely discard focus or seeing as a cause, simply due to the FWHM results you listed, you can confirm this yourself by looking at the FWHMs for the three stacked channels, I'll bet they are very comparable.

The listing you posted did not show the altitude of the individual frames, but again you can easily assure yourself that it isn't a factor (it would show up in the FWHM anyways).

The bgd count doesn't reveal much if anything about the focus or seeing, just the convolution of camera QE, filter bandpass and light pollution/natural sky brightness where the scope was pointed.

What is more important is the stellar profile. In Maxim there is a graph window which displays the line profile of star for all three colors simultaneously (for color images), and also a graph of intensity versus radius for each channel (you can do these graphs in excel if you can export the data in tabular txt format, but that's a pain). From this you can quantify the amount of halo in each channel independent of stretch or black point level. Don't have Maxim? send me the raw stacked FITS x3 in floating point or 16bit, cropped around the area of interest.

Now the FWHM calculation is not necessarily sensitive to the low intensity halo, but the aggressive stretching we use in processing to bring out faint nebulosity takes no prisoners. So here are some hypothesis to try:
1) greater reflection/scatter from CCD microlenses in the blue
2) reflection/scatter from telescope surfaces
3) reflection/scatter from filter surfaces
4) limited correction of objective lens in the blue

To confirm/deny these:
1) Use a different camera which has a different chip (e.g. KAF, KAI or DSLR, OSC) the latter two nicely avoid changes in seeing or altitude. You may already have images from this scope with another camera?
2) Not being a refractor guy I am unsure here, but I do know the design and AR coatings need to be right in order to avoid ghosts which can show up as haloes. Do others have the same problem with different cameras?
3) Borrow a blue filter from a mate and see if it goes away?
4) I doubt it is this one since you would think the FWHM would be affected, only pick as right answer if everything else is excluded!

HTH,
EB
Reply With Quote
  #9  
Old 21-09-2014, 08:51 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Ray and Eric,

Thanks a lot for your feedback and ideas.

I've done a bit more research and found a spec for the TEC140. It's corrected from 436-1000 nm, but the Baader seems to transmit to 380 nm where the Trius is still at ca 40% QE.

I was all set to try a Hutech IDAS LP filter (cuts at ca 410 nm) to see if this helps. Then the clouds rolled in, of course....

If this works I wonder if there is any way I could just do the UV cut in blue and not the other channels. A shame because these Baaders without any LP filter generate the nicest/easiest colour I've ever had.

Eric, I will test my images in Maxim. I have the program but rarely use it these days. I appreciate the other tips as well. I'm hoping it's just the need for an IR cut. The other issues would be fairly insurmountable!

Peter
Reply With Quote
  #10  
Old 22-09-2014, 07:18 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
hi Peter.

figure 149, example 17 in the reference shows the spots diags for your scope design (multi-band is the top spot diagram) - it definitely has an extensive halo at the blue end (about 10x the (black) diffraction spot) and presumably would be even less well corrected in the far violet. http://www.telescope-optics.net/semi...o_examples.htm The reference notes that the straight triplet is best for visual and the designer could have chosen to pull in the violet for imaging, by sacrificing performance elsewhere. However, the general characteristic of a significant violet halo will probably still be there. The Trius has much higher blue/violet sensitivity than most Kodak chips, so is better able to detect any such halos.

if it is a violet halo due to residual CA in the APO, it would be in the lum image as well (as Greg has found), so a violet-cut filter would need to be applied there as well - probably best to just put one front of the camera and use it for all channels.

Eric, I think you can discount 1. and possibly 3. - halos do not show up in images taken with reflectors using the same camera (and I think using the same filters).

Last edited by Shiraz; 22-09-2014 at 08:48 AM.
Reply With Quote
  #11  
Old 22-09-2014, 08:47 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by Shiraz View Post
hi Peter. if it is a violet halo due to CA in the APO, it would be there in the lum image as well, so a UV cut filter would need to be applied there as well - probably best to just put in front of the camera and use it for all channels.

figure 149, example 17 in the reference shows the spots diags for your scope design (multi-band is the top spot diagram) - it definitely has an extensive halo at the blue end (about 10x the diffraction spot) and presumably would be even less well corrected in the violet. http://www.telescope-optics.net/semi...o_examples.htm The reference notes that the designer could have chosen to pull in the violet by sacrificing performance elsewhere. However, the general characteristic of a significant violet halo will probably still be there.
Ray,

Many thanks for the spot diagrams link. That is most interesting. Remember that I am imaging with two scopes simultaneously! The TEC180 for luminance with a CLS filter (strong UV cut!) and the TEC140 for RGB/narrow band. I put a similar post at Cloudy Nights and received some very interesting info there re the ICX694 on a TEC140 (not a Trius, a QSI). Using Astrodon blue he is getting decent results and put a link to a blue fits file. It is a big improvement over what I'm getting. The Astrodon seems to cut at 400 nm whereas the Baader cuts at ca 380. Enough of a difference? Perhaps. At least with the QSI it seems to be. Could camera electronics come into the equation, or cover glass material? Also, yes, Greg Bradly shows bloated stars in luminance with his Trius on the TEC180. If you are curious here is the CN post:

http://www.cloudynights.com/topic/47...ius/?p=6219293

Now, if I could just incorporate a CLS filter with just BLUE I might be quite happy. The CLS cuts at 450 nm.

Peter

Last edited by PRejto; 22-09-2014 at 09:17 AM.
Reply With Quote
  #12  
Old 22-09-2014, 10:10 AM
JohnH's Avatar
JohnH
Member # 159

JohnH is offline
 
Join Date: Feb 2005
Location: NSW
Posts: 1,226
Quote:
Originally Posted by RickS View Post
If you look at the CCDI data you'll see that, on average, the blue subs are less bright than the red and green subs (Background).

When the subs get stretched the blue ones need to be stretched further and the blue channel stars bloat giving a blue halo. It's the same reason that Hubble palette images have magenta star halos.

Cheers,
Rick.
So I should have been exposing my B frames for longer than the Rs and Gs...any general guidance on exposure ratios to use or is it too dependent on the filter/optic/ccd combination??
Reply With Quote
  #13  
Old 22-09-2014, 03:19 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
The TEC fluorite series are among the most exotic scopes available.
It is most certainly not the scope. I have used several cameras with my TEC180 and only seen this problem with a Trius. I have observed it a few times on extremely bright stars with a KAF8300 camera. Both have small wells. I get perfect stars with a FLI Proline 16803 with over 100,000 electron well depth versus 18,000 with a Trius.

I think the best strategy is to match optics/CCDs and the Trius works better on some scopes than others. Either that or the Trius has has something different from the SX694. Perhaps SX changed something like the coatings or the glass etc.

I put it down to small wells. I see the effect on my CDK17 just not as noticeable.

Why it does appear to be a problem with Mike's scope is strange although he tends to use short subs.

Ray also uses even shorter subs so perhaps that is the best approach - very short subs so the wells don't fill further out from the star's centre.

I can try it on my AP140 and see if there is any difference next time I am using it. I suspect no difference but we'll see.

I am wondering if a FLI Microline 694 would behave the same way. Its only a feeling but I suspect not. This is why the SX is the cheapest, not necessarily the best.

Greg.
Reply With Quote
  #14  
Old 22-09-2014, 08:31 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by PRejto View Post
Ray,

Many thanks for the spot diagrams link. That is most interesting. Remember that I am imaging with two scopes simultaneously! The TEC180 for luminance with a CLS filter (strong UV cut!) and the TEC140 for RGB/narrow band. I put a similar post at Cloudy Nights and received some very interesting info there re the ICX694 on a TEC140 (not a Trius, a QSI). Using Astrodon blue he is getting decent results and put a link to a blue fits file. It is a big improvement over what I'm getting. The Astrodon seems to cut at 400 nm whereas the Baader cuts at ca 380. Enough of a difference? Perhaps. At least with the QSI it seems to be. Could camera electronics come into the equation, or cover glass material? Also, yes, Greg Bradly shows bloated stars in luminance with his Trius on the TEC180. If you are curious here is the CN post:

http://www.cloudynights.com/topic/47...ius/?p=6219293

Now, if I could just incorporate a CLS filter with just BLUE I might be quite happy. The CLS cuts at 450 nm.

Peter
thanks for the link Peter. One of the posts suggested a Baader UV-IR cut filter, which might be a solution if the problem turns out to be violet CA. If the IDAS makes a significant difference, that could be the next thing to try, since a UV-IR cut filter could stay in place without harming the LRG data.

thinking about it, you guys are really pushing the boundaries by using the scopes outside of their design spectral region, with pixels 1/4 the area of more common ones and with chips that have high sensitivity in the violet region.

Last edited by Shiraz; 22-09-2014 at 08:52 PM.
Reply With Quote
  #15  
Old 22-09-2014, 09:37 PM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Ray,

You are correct, and personally I don't think it's working out too well for me. I looked again at the link to the photo that was put up over at CN. I now think it is just as poor as what I'm getting. I think I wanted to believe it was better!

Tonight I took a lot of data that I need to think about and perhaps graph. The Hutech IDAS filter certainly helped with a sense of better focus in blue, and reduced the bloat some. It's a step in the right direction, but not enough. I also took 3 sets of images working upward in 30 sec increments up to 5 minutes. There is pretty clear bloating/time in blue. I didn't run the same extensive test in red but there was also some bloat. Compared to blue it wasn't terribly significant. So, it seems there are at least 2 issues going on here. The violet spectrum is out of focus and is causing focus and halo looking issues, and even 5 min exposure of some stars will be far too much.

My luminance image using a CLS filter on the G2-8300 is tack sharp. I took some images with the blue Astronomik filter just to compare. I'd never done that on the TEC180. I noticed a tiny amount of star enlargement along with a tiny bit of soft looking focus. It appears that even this camera will suffer just a tiny bit without a UV block. Previously I had used this camera on my TEC140 but always with the Hutech IDAS filter which produced pretty tight stars in blue. I'm inevitably coming to the conclusion that this camera/scope matchup is not made in heaven. If the filter option fails it might be time for the camera to find a new owner with a reflecting system perhaps.

Peter
Reply With Quote
  #16  
Old 22-09-2014, 10:25 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Hi Peter,
So while a few more measurements will really confirm this, it looks like it is a combination of 4) + a slightly different 1) and 3).

The camera has to much QE for the exotic refractor hmmm.... I guess I'll ditch the camera and find a worse one...

But seriously have you considered a catadioptric? Although if a simple bandpass filter fixes the problem, well that is the best outcome for your wallet.

Best,
EB
Reply With Quote
  #17  
Old 22-09-2014, 11:05 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Peter, perhaps have a look at the Wratten 2E filter - it might possibly do what you want, but I have no idea where to get one. The standard filters are gelatin, but someone may make a glass one - quick search did not find any though, so will keep looking.

Agree with Eric - is probably worth persevering a bit with filters if there is still some possibility of getting a usable system - the idea of using the two scopes is a really good one.
Reply With Quote
  #18  
Old 23-09-2014, 07:09 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
Quote:
Originally Posted by Shiraz View Post
Peter, perhaps have a look at the Wratten 2E filter - it might possibly do what you want, but I have no idea where to get one. The standard filters are gelatin, but someone may make a glass one - quick search did not find any though, so will keep looking.

Agree with Eric - is probably worth persevering a bit with filters if there is still some possibility of getting a usable system - the idea of using the two scopes is a really good one.

Eric and Ray,

Ah, I was unclear. I very much want to get the Trius to work. Of course I'm not giving up yet. I'm waiting for a filter I ordered. If it doesn't work I don't really see how or why I should continue to struggle with this issue. Yes, it's a great camera but it reveals something not so great in this particular combination. That's all. I really hope to find a way around all of this!


Peter
Reply With Quote
  #19  
Old 24-09-2014, 07:53 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Rylstone, NSW, Australia
Posts: 1,397
I have continued experimenting with the blue filter and Hutech IDAS. I also was loaned an Astrodon blue which made for an interesting comparison with the Baader. The Baader cuts at 380 nm, and the Astrodon at 400 nm, thus one would expect the Astrodon to be slightly sharper. This was born out in testing though the difference is subtle. The Astrodon appears to concentrate the light a bit more but still showed quite a large halo effect. With the Hutech IDAS in place in front of both filters there is an improvement as the Hutech cuts at 410 nm. To my eyes the effect was better for the Baader; I'm not sure why but I saw it after several iterations.


I am going to be trying out the Baader Semi-Apo filter and depending on the result may purchase a Shott GG435 as recommended by Gerd at Astronomiks. That cut might exactly match the TEC140 lens specification. What remains to be seen is how overall colour balance works with such a filter in place. A look at the graph showing frequency response of various cameras easily shows why the effect I never noticed with my KAF8300 camera is now so obvious with the Trius.



The other issue that clouds all of this discussion is that of star size using the Trius. It's hard to separate the two issues so I ran a bit of an experiment* with the blue filter by gradually increasing exposure times in 30 second increments up to 5 minutes. The stars in question range from about 9 to 11th magnitude. I've plotted the results. Perhaps my test had too many variables; seeing, gradually lowering altitude, tracking (I was guiding), etc. Still, the average result points to a possible enlargement taking off around the 3.5 minute mark. The reason I don't put this on the doorstep of tracking is that the aspects in CCDInspector are still quite good at the 3-5 minute positions. Anyway, I put it out here for what it's worth. Personally I do think that my 5 minute exposures have probably been too long if I want to avoid even bigger and overly saturated stars. But, even 30 seconds gives saturated stars. I wondering if in using this camera one might need to take a whole range of exposures in order to simulate the effect of deeper wells?


Peter

* I focused with FocusMax before each of the three runs. This was with the Baader blue filter.
Attached Thumbnails
Click for full-size image (Baader-30 sec.jpg)
172.9 KB35 views
Click for full-size image (AD-30 sec.jpg)
174.2 KB28 views
Click for full-size image (Baader + IDAS-30 sec.jpg)
160.0 KB27 views
Click for full-size image (AD + IDAS-30 sec.jpg)
165.7 KB26 views
Click for full-size image (CCDInsp.jpg)
193.7 KB27 views
Click for full-size image (Size Graph.jpg)
147.2 KB31 views
Click for full-size image (materials_schott_GG435_1.gif)
18.6 KB28 views
Click for full-size image (Freq-cameras-02.jpg)
173.4 KB34 views
Reply With Quote
  #20  
Old 24-09-2014, 10:25 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Certainly looks like that is on the right track Peter - the Hutech made a huge difference.

The FWHM result is a bit surprising - I can think of no reason why increasing exposure should change star shape on a linear CCD (it will when the stars reach saturation, but CCD inspector throws them out). Maybe the increasing background level is upsetting the FWHM calculation. The contrast ratio is reducing with longer exposures (I think it should increase) which may suggest that the data has not been dark calibrated - that could explain the change in FWHM with increasing exposure. Perhaps rerun with dark calibration if it was not done - I think that the "Reduce" checkbox lets you do that if it has not already been done elsewhere. If you don't have darks for those exposures, maybe try something like the dynamic PSF in Pixinsight (do you use PI?).

you could use multiple exposures to gain a bit of dynamic range if saturated stars bother you - I think that Rick posted some details on this process (again in PI).

regards ray

Last edited by Shiraz; 24-09-2014 at 10:46 AM.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 08:31 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement