Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

QHY183M Reflection Problems

  • Please log in to reply
28 replies to this topic

#1 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 15 March 2019 - 12:10 PM

I thought this wasn't supposed to be a problem like it is with the ASI1600? Did some initial testing with my new QHY183M and Chroma LRGBSHO filters. Here are the results:

 

Merak 2.37 mag A1V (not as bright as Alnitak at 1.74 mag)

Gain 5 offset 10 (ASI equivalent to gain 50 and offset 100)

Tec140 @ f7

90s exposure LRGB

120s exposure SHO

Chroma LRGBSHO filters (3nm SHO). There are two versions of the L and B filters. The ones labeled "new" have a new antireflective coating scheme from Chroma.

There is an Empty Slot frame that was taken with no filter as a control

 

Ignore the weirdness on the left side of the diffraction pattern with the old filters. I smudged the flattener during filter change and later cleaned it and the weirdness went away. Am I expecting too much on a bright target like this?

 

Any thoughts?

 

 

 

 

Attached Thumbnails

  • QHYFilterTest.jpg

  • rockstarbill likes this

#2 Konihlav

Konihlav

    Surveyor 1

  • -----
  • Posts: 1622
  • Joined: 05 Mar 2009
  • Loc: Germany

Posted 15 March 2019 - 04:10 PM

what you should try:

- use other telescope

- use the same optical train, but do include an extra UV/IR cut filter (for the blue part of spectra) like the Baader one who cuts from 420nm+ see item # 2459210A

 

this looks to me like a "problem" in the optical path as you see this reflection issue only in Blue channel (and as well as Lum and no filter).

 

the IMX183 shall not definitively have the same kind of ar-microlenses-coating-issue as the PANA 34230 chip has (in ASI1600 or QHY163).

 

there's a known problem (for many years, also on my blog http://blog.astrofotky.cz/pavelpech/) with refractors and e.g. Astrodon filters since they allow 380nm wavelengths to pass through and that is OK for reflecting optics, but refractors typically have problems with UV light).

Put an UV cut filter in path.

 

these are the facts (apart from my dislike of QHY and Chroma :-) ).



#3 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22923
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 15 March 2019 - 04:40 PM

How possible it is for the microlenses to diffract would depend on what wavelengths of light the AR coating on the cover glass is designed to filter out. Multi-layer AR coatings operate on the wave-canceling properties of light. Different thickness coatings will reflect different wavelengths of light. Cancellation of reflections is not perfect. Depending on exactly how broad the blue bandpass is, it may be possible for the microlenses to diffract if the deep end of the blues is not properly canceled by the AR coating. Early cutoff of the blue filter (i.e. before 400nm) might help with this. I am not familiar enough with Chroma's bandpasses to know where their blue filters cut off...but it is something to look into. 



#4 Konihlav

Konihlav

    Surveyor 1

  • -----
  • Posts: 1622
  • Joined: 05 Mar 2009
  • Loc: Germany

Posted 15 March 2019 - 05:52 PM

one more thing I forgot to mention is that these new CMOS chips (as well as some of the CCDs like ICX694) have a really very, very high QE in the UV / Blue wavelengths!

That's why many people with the same refractor (TEC140) do not observed any problem with bloated stars in Blue channel (and L) if they were using KAF-8300 (or KAF-16200 nowadays) CCD cameras with very poor QE in the Blue channel... Replacing KAF for ICX or now IMX183 with extremly high QE in that spectra immediately shows this issue.

I am almost (99%) certain that if you add a UV cut filter to the optical path, it will solve your problem. And it is the only and best thing for you to do now.



#5 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 15 March 2019 - 06:02 PM

what you should try:

- use other telescope

- use the same optical train, but do include an extra UV/IR cut filter (for the blue part of spectra) like the Baader one who cuts from 420nm+ see item # 2459210A

 

this looks to me like a "problem" in the optical path as you see this reflection issue only in Blue channel (and as well as Lum and no filter).

 

the IMX183 shall not definitively have the same kind of ar-microlenses-coating-issue as the PANA 34230 chip has (in ASI1600 or QHY163).

 

there's a known problem (for many years, also on my blog http://blog.astrofotky.cz/pavelpech/) with refractors and e.g. Astrodon filters since they allow 380nm wavelengths to pass through and that is OK for reflecting optics, but refractors typically have problems with UV light).

Put an UV cut filter in path.

 

these are the facts (apart from my dislike of QHY and Chroma :-) ).

I don't have another scope (yet), and unfortunately I don't have anywhere to mount another filter in front of the others. The only other filter in my possession at the moment is an IDAS LPS D2. I can test it out on its own, but don't know what it would add to the data I have now. I've not heard of this additional UV filter being necessary. Is this common practice?



#6 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 15 March 2019 - 06:14 PM

How possible it is for the microlenses to diffract would depend on what wavelengths of light the AR coating on the cover glass is designed to filter out. Multi-layer AR coatings operate on the wave-canceling properties of light. Different thickness coatings will reflect different wavelengths of light. Cancellation of reflections is not perfect. Depending on exactly how broad the blue bandpass is, it may be possible for the microlenses to diffract if the deep end of the blues is not properly canceled by the AR coating. Early cutoff of the blue filter (i.e. before 400nm) might help with this. I am not familiar enough with Chroma's bandpasses to know where their blue filters cut off...but it is something to look into. 

Here are the transmission graphs for the L and Blue. L is about 365 to 700, and Blue is about 390 to 510. The old version of the blue goes down to 365. It was raised to 390 based on customer feedback. There is no difference in diffractions between the new and old filters.

 

The green frame is actually brighter than the blue, with a higher total illumination and same peak brightness. I'm surprised it doesn't also show the problem if this is purely a microlens reflection issue. 

Attached Thumbnails

  • Screen Shot 2019-03-15 at 18.36.09.jpg
  • Screen Shot 2019-03-15 at 18.36.26.jpg


#7 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 15 March 2019 - 06:21 PM

one more thing I forgot to mention is that these new CMOS chips (as well as some of the CCDs like ICX694) have a really very, very high QE in the UV / Blue wavelengths!

That's why many people with the same refractor (TEC140) do not observed any problem with bloated stars in Blue channel (and L) if they were using KAF-8300 (or KAF-16200 nowadays) CCD cameras with very poor QE in the Blue channel... Replacing KAF for ICX or now IMX183 with extremly high QE in that spectra immediately shows this issue.

I am almost (99%) certain that if you add a UV cut filter to the optical path, it will solve your problem. And it is the only and best thing for you to do now.

Funny you mention the 16200. That is what started this whole mess. I had halos with Lum and Blue with my 16200, along with several other folks (as it turns out, also with 16200 chips). Again, no problems in Red or Green or any of the narrowband. The new filters show the same properties, but so does an image with no filter (image on far right). There is another thread on the whole Chroma saga linked below.

 

Here is my 16200 data. Same target but 45s because they are at f5 and not f7. New and old versions of Lum and Blue, plus an image with no filter. Again, the green and red did not show these halos. So confused. 

 

Shouldn't a cut off of 390 block out most of the troubling UV?

Attached Thumbnails

  • AtikFilterTest.jpg

Edited by Monkeybird747, 15 March 2019 - 06:25 PM.


#8 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22923
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 15 March 2019 - 07:09 PM

Here are the transmission graphs for the L and Blue. L is about 365 to 700, and Blue is about 390 to 510. The old version of the blue goes down to 365. It was raised to 390 based on customer feedback. There is no difference in diffractions between the new and old filters.

 

The green frame is actually brighter than the blue, with a higher total illumination and same peak brightness. I'm surprised it doesn't also show the problem if this is purely a microlens reflection issue. 

Ok...I think that is probably the issue then. The L is very, very wide, and stretches into the NUV. Down to 360nm, which is definitely violet and UV. The Blue stretches below 400nm. I would not be surprised if the AR coating on the 183 cover glass did not account for those wavelengths. A lot of other filter sets will cut off right AT 400nm, or even before 400nm, to filter out some LP wavelengths. The L filter in particular here...its pass goes pretty deep into the near UV for an L filter. I actually started experimenting with the Astronomik Ln filters, they have L1, L2 and L3, and the differences between them are their cutoffs...each higher number has a tighter cutoff into the visible spectrum to cut out NUV, NIR and some LP. I use the Astronomik L3 now for my guide camera.

 

Anyway...these bandpasses could be the problem. If you can find blue filters that cut off before 400nm, say 410 or even 415 or so, that might produce better results. 


  • danielguo likes this

#9 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22923
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 15 March 2019 - 08:24 PM

So, looking at your original examples again. I am even more convinced now that the UV spectrum is the problem here. Your old blue had a worse problem than the new blue. The old blue, you said went down to 365nm, and had stronger reflections. The new blue goes down to around 390, and doesn't have as strong of reflections. That seems to indicate to me that it is UV/violet/deep blue light beyond 400nm (maybe even beyond 410-415nm) that is reflecting.



#10 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 15 March 2019 - 10:58 PM

The best I can tell from the graphs the Astrodon Gen II E blue filter has an almost identical cut on point as the Chroma version. However the Chroma Lum is quite a bit wider. I'm not aware of any LRGB sets that cut on at the low 400s. And with the inability to add additional filter earlier in the optical patch, I could be stuck.

 

Someone is doing some testing with their FLI 16200 and astrodons with similar scope and same reducer to see if the problem can be duplicated. Of particular interest is the no-filter test. That should help determine if the protect window coating is of some interest. 

 

I'm also testing with the LPS-D2 filter. It has a cut on of around 415 or so. Again, hard to tell from the graph but it is above 400. I think some clouds passed during the exposure so I will rerun the frame before posting. Initial results looked like the diffraction patter was reduced, but not eliminated. I'll post an update this evening. 

 

The QHY183 is mounted up and I'm in test mode. If anyone is curious about the camera and wants to see a certain test I'm up for it.



#11 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22923
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 15 March 2019 - 11:12 PM

The best I can tell from the graphs the Astrodon Gen II E blue filter has an almost identical cut on point as the Chroma version. However the Chroma Lum is quite a bit wider. I'm not aware of any LRGB sets that cut on at the low 400s. And with the inability to add additional filter earlier in the optical patch, I could be stuck.

 

Someone is doing some testing with their FLI 16200 and astrodons with similar scope and same reducer to see if the problem can be duplicated. Of particular interest is the no-filter test. That should help determine if the protect window coating is of some interest. 

 

I'm also testing with the LPS-D2 filter. It has a cut on of around 415 or so. Again, hard to tell from the graph but it is above 400. I think some clouds passed during the exposure so I will rerun the frame before posting. Initial results looked like the diffraction patter was reduced, but not eliminated. I'll post an update this evening. 

 

The QHY183 is mounted up and I'm in test mode. If anyone is curious about the camera and wants to see a certain test I'm up for it.

Your blue plot cuts about 10-15nm deeper into the UV than my Gen2 E-series ADons do. The reflections on the new blues you have are definitely fainter than the old blues, so it seems that having less UV light improved the issue.

 

Have you looked at the Astronomik Deep Sky RGB filters?

 

https://www.astronom...our-filter.html

 

They seem to cut at 420nm. You could also pick up the Astronomik L3, which also cuts out before 400nm. See of those do the trick.


  • danielguo likes this

#12 Konihlav

Konihlav

    Surveyor 1

  • -----
  • Posts: 1622
  • Joined: 05 Mar 2009
  • Loc: Germany

Posted 16 March 2019 - 06:19 AM

as we both say, add a UV cut from 420nm + and you are done... This is needed because of the very, very high QE of the IMX183 in the UV part of the spectra, it has insanely high QE in UV compared to shallow QE in UV of KAF-8300 or 16200 chips... (no additional talk about filters involved).

 

:-)

 

let us know that this fixed the problem as we are almost certain about it...



#13 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 16 March 2019 - 02:18 PM

The chroma graph and astrodon graph have a little different graph scale. The literature says astrodon cut on at 395, and Chroma 390. So reasonably close. What could be important though is the transmission rate at that wavelength. The astrodon is only 50% up to about 400, while the chroma is over 90% as soon as it cuts on at 390, and does not taper like the astrodons.

 

I did not realize the Astronomiks cut on that high. Worth a look. Coincidentally, I had a set of those unopened that I returned when I decided to get the Chroma filters...

 

 

Your blue plot cuts about 10-15nm deeper into the UV than my Gen2 E-series ADons do. The reflections on the new blues you have are definitely fainter than the old blues, so it seems that having less UV light improved the issue.

 

Have you looked at the Astronomik Deep Sky RGB filters?

 

https://www.astronom...our-filter.html

 

They seem to cut at 420nm. You could also pick up the Astronomik L3, which also cuts out before 400nm. See of those do the trick.

 

Attached Thumbnails

  • astrodonespectra.jpg


#14 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 16 March 2019 - 02:29 PM

as we both say, add a UV cut from 420nm + and you are done... This is needed because of the very, very high QE of the IMX183 in the UV part of the spectra, it has insanely high QE in UV compared to shallow QE in UV of KAF-8300 or 16200 chips... (no additional talk about filters involved).

 

:-)

 

let us know that this fixed the problem as we are almost certain about it...

If I had a place to stick a UV filter I would try it. But I just don't have a place for it without a complete redesign of the image train and a few pricey custom adapters. However the UV aspect of the discussion is interesting.

The testing with the higher cut on LPS D2 filter was an improvement, but did not completely eliminate the problem. It cuts on around 415 or so.

The odd part about the QE discussion is I'm also having halo problems with the 16200 chip, which does not have high QE. Those pictures are posted above. The green and red and narrowband on that chip don't show the problem.

I'm surprised that someone hasn't take this camera and pointed it at Alnitak with a set of Baader filters or some other filters with sub-400 low cut on and started crying foul at the diffraction patter that is not supposed to be present on this sensor. Could it be different protect glass coating than the ASI version?


Edited by Monkeybird747, 16 March 2019 - 02:30 PM.


#15 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 16 March 2019 - 02:33 PM

Jon, have you pointed your ASI183MM with the FSQ at a mag 2-3 star with LRGB at any point in your testing? I need more data points before I launch into another frenzied equipment shuffle.



#16 rockstarbill

rockstarbill

    Fly Me to the Moon

  • *****
  • Posts: 5660
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 16 March 2019 - 03:06 PM

I dont know if Jon did or not, but I found an image on Astrobin that is interesting. 183MM-Pro and Astronomik Deep Sky filters. No issue on Alnitak that I can see. 

 

https://www.astrobin.../full/374339/B/



#17 AIP

AIP

    Vostok 1

  • -----
  • Posts: 135
  • Joined: 14 Feb 2019
  • Loc: Madrid, Spain

Posted 16 March 2019 - 04:01 PM

 

I'm surprised that someone hasn't take this camera and pointed it at Alnitak with a set of Baader filters or some other filters with sub-400 low cut on and started crying foul at the diffraction patter that is not supposed to be present on this sensor. 

Alnitak.

FSQ106EDX f/5 + Atik 16200m + Baader LRGB 2"
L: 15x600 bin1 and RGB: 12x300s bin2 each channel

 

high quality link:

https://farm8.static...21a34d33d_o.png

 

33521362798_78618a1354_b.jpg


  • rockstarbill and Monkeybird747 like this

#18 AIP

AIP

    Vostok 1

  • -----
  • Posts: 135
  • Joined: 14 Feb 2019
  • Loc: Madrid, Spain

Posted 16 March 2019 - 04:10 PM

Funny you mention the 16200. That is what started this whole mess. I had halos with Lum and Blue with my 16200, along with several other folks (as it turns out, also with 16200 chips). Again, no problems in Red or Green or any of the narrowband. The new filters show the same properties, but so does an image with no filter (image on far right). There is another thread on the whole Chroma saga linked below.

 

Here is my 16200 data. Same target but 45s because they are at f5 and not f7. New and old versions of Lum and Blue, plus an image with no filter. Again, the green and red did not show these halos. So confused. 

 

Shouldn't a cut off of 390 block out most of the troubling UV?

With what filters are taken? Chroma?



#19 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 17 March 2019 - 12:38 AM

Alnitak.

FSQ106EDX f/5 + Atik 16200m + Baader LRGB 2"
L: 15x600 bin1 and RGB: 12x300s bin2 each channel

 

high quality link:

https://farm8.static...21a34d33d_o.png

 

33521362798_78618a1354_b.jpg

Thank you AIP, this is a great data point.


  • AIP likes this

#20 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 17 March 2019 - 12:39 AM

With what filters are taken? Chroma?

Yes, Chroma LRGB and 3nm SHO.


  • AIP likes this

#21 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 17 March 2019 - 12:42 AM

I dont know if Jon did or not, but I found an image on Astrobin that is interesting. 183MM-Pro and Astronomik Deep Sky filters. No issue on Alnitak that I can see. 

 

https://www.astrobin.../full/374339/B/

Thanks Bill, another good data point. I'm researching if the 2" mounted versions of the Astronomik Deep Sky LRGB and L3 filters are available in 3mm substrate. The website is ambiguous on this when it comes to mounted versions. Unmounted versions seem to be available in 3mm.



#22 rockstarbill

rockstarbill

    Fly Me to the Moon

  • *****
  • Posts: 5660
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 March 2019 - 12:44 AM

Thanks Bill, another good data point. I'm researching if the 2" mounted versions of the Astronomik Deep Sky LRGB and L3 filters are available in 3mm substrate. The website is ambiguous on this when it comes to mounted versions. Unmounted versions seem to be available in 3mm.

The best data point would be the consideration of the correction of the TEC140. The manual calls it to be corrected from 436-1000. That could be at play here, although based on the other thread where an FSQ showed similar problems... I am not so sure. 


Edited by rockstarbill, 17 March 2019 - 12:44 AM.


#23 Monkeybird747

Monkeybird747

    Messenger

  • *****
  • topic starter
  • Posts: 405
  • Joined: 31 Aug 2017
  • Loc: Louisville, KY

Posted 17 March 2019 - 12:50 AM

The best data point would be the consideration of the correction of the TEC140. The manual calls it to be corrected from 436-1000. That could be at play here, although based on the other thread where an FSQ showed similar problems... I am not so sure. 

Yes, thanks for pointing that out. It also mentions being optimized for visual in terms of color correction. And as a comparison I believe you mentioned your GTX is corrected for 400 - 706, and your 16200 chip does not show the halos with that scope. I'd love to hear from someone on whether or not the published color correction wavelengths could be tied somehow to these halos and/or UV light transmission. I'm way out of my element when we start going down that road.



#24 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22923
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 17 March 2019 - 01:17 AM

Yes, thanks for pointing that out. It also mentions being optimized for visual in terms of color correction. And as a comparison I believe you mentioned your GTX is corrected for 400 - 706, and your 16200 chip does not show the halos with that scope. I'd love to hear from someone on whether or not the published color correction wavelengths could be tied somehow to these halos and/or UV light transmission. I'm way out of my element when we start going down that road.

I don't believe it is the scope at all. Microlens diffraction occurs within the sensor itself, only. The microlens array diffracts, that diffracted light bounces off the back of the sensor cover glass and back down to the sensor. The scope is not involved in that at all. Even if a scope is not well-corrected at UV wavelengths, they can still pass them. That is all that the scope needs to do here...pass UV wavelengths down to 360nm.

 

I believe the issue is simply that the IMX183 AR coating on the cover glass is simply not canceling out UV wavelengths. AR coating has to target specific wavelengths of light to cancel out. It is a multi-layer coating (anywhere from 2 to hundreds, depending on how advanced it is). The more layers, the more distinct wavelengths of light can be targeted to be canceled out. It is rather doubtful that UV rays from 399 down to 360 are going to be covered by the AR coating on the sensor. I would actually be surprised if it had more than just a few layers, and was targeting wavelengths around "red", "green" and "blue" only, thus being an imperfect cancellation of even the visible spectrum.

 

Since your filters are allowing UV light as deep as 360nm to pass...if the AR coating is indeed NOT canceling those wavelengths, then any microlens diffraction will shine brightly at those wavelengths.


Edited by Jon Rista, 17 March 2019 - 01:18 AM.


#25 AIP

AIP

    Vostok 1

  • -----
  • Posts: 135
  • Joined: 14 Feb 2019
  • Loc: Madrid, Spain

Posted 18 March 2019 - 10:08 AM

Alnitak.

FSQ106EDX f/5 + Atik 16200m + Baader LRGB 2"
L: 15x600 bin1 and RGB: 12x300s bin2 each channel

 

high quality link:

https://farm8.static...21a34d33d_o.png

 

33521362798_78618a1354_b.jpg

well...

 

the same telescope, the same filters and Atik 460ex (IXC694)

 

only microlenses reflections are visble in blue channel, very faint btw

 

31842783815_778851a60a_b.jpg


Edited by AIP, 18 March 2019 - 10:24 AM.



CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics