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

QHY600M first light with TEC 160FL and TEC FRC

  • Please log in to reply
44 replies to this topic

#1 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 29 November 2021 - 05:25 PM

I finally took test images of M31 last night with my new QHY600M camera. 

 

I used gain of 0, offset of 20, temperature at -10C, TEC 160FL, TEC 0.9x FRC, 1008mm focal length, f/6.3, Lum filter. I took two sets of each of the following:

 

1) 10 minutes, bin 1x1

2) 5 minutes, bin 1x1

3) 5 minutes, bin 2x2

4) 2.5 minutes, bin 2x2

 

I couldn't believe this huge sensor was able to squeeze in pretty much the whole Andromeda Galaxy. This is quite a high resolution image of M31 at 1008mm focal length. There were minimal star saturation at 10 minutes bin 1x1. The well capacity for read mode #0 and gain of 0 is about 80K electrons!!!! The FWHM of pretty much all images were less than 2 arcsecs. I've never seen such great guiding and it was the first time using multi-star guiding with PHD2. The stars viewed through OAG's port were elongated due to the prism located at quite a bit "off-axis" but that didn't affect PHD2 at all. PHD2 RMS errors for both axes were something like 0.24" - 0.28" and that was my best.

 

I uploaded the best of each and un-calibrated light subs at:

 

https://www.dropbox....F_20_W.FIT?dl=0

https://www.dropbox....F_20_W.FIT?dl=0

https://www.dropbox....F_20_W.FIT?dl=0

https://www.dropbox....F_20_W.FIT?dl=0

 

Master flats in bin 1x1 and 2x2:

 

https://www.dropbox....set_20.fit?dl=0

https://www.dropbox....set_20.fit?dl=0

 

Notice double v-shaped lines in Master flats. I believe this was caused by reflection from OAG's prism but didn't affect calibrating with lights.

 

Calibrated, registered, Linear FIT and stack of 8 light subs with different exposure times (150s, 300s, 600s) and binning (1x1 and 2x2) totaling only 45 minutes:

 

https://www.dropbox....ration.fit?dl=0

 

The title of the files will tell you the details.

 

I absolutely had no issues calibrating the lights and flats. I used Master bias instead of dark flats to calibrate the flats and it worked perfectly. The sub-exposure times of flat subs were less than 0.1 second. I didn't need to use output pedestal even though one calibrated image had a minimum of 2 ADU!!!! The minimum ADU for calibrated bin 1x1 is much smaller than bin 2x2 subs.

 

Notice very strange vertical object at x, y location 3690, 1140 of all of my images. Zoom this location. It's likely reflection from OAG's prism. Hopefully raising the prism will help. If you stretch the Master Flats heavily, you will barely see prism shadow at top and it's directly above the reflection.

 

I see some stars astigmatism in the corners. This is probably due to spacing issues which I have to fix. There are elongated stars at the corners probably due to vignetting and spacing. I kind of expected vignetting because focal reducers naturally makes illuminated image circle smaller.

 

Overall, I'm pretty satisfied.

 

Peter

 

M31_Lum_Compressed.jpg


Edited by Peter in Reno, 30 November 2021 - 10:57 AM.

  • StephenW, tboss70, rwstanley and 8 others like this

#2 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 08:52 AM

Peter,

 

Interesting.  It's funny, I emailed you this morning before I saw this and mentioned that I was interested in the extended FW mode. 

 

I buzzed around your image and I couldnt find any blown stars.  I found one spot on the image that was 65534 ADU and that was the core of M32.  Otherwise all of the brightest stars came in somewhere below that.  (Not sure if this chip saturates at the normal 16bit of 65535.  In statistics the max pixel value is indeed that 65534 that I found in M32.  My 6200 clipped at 65535.  I wonder what causes this insignificant discrepancy. 

 

Once I get my camera I'll take an equivalent exposure of M45 that I took with my 6200 to see how the extended mode compares.  If the additional ~30k electron FW is indeed real it should make a huge difference.    I'm using this with a 130mm scope at 875mm FL.  The FOV is wonderful and as you mentioned with your scope, the image scale is very nice. 

 

Looking at the rest of the image it does appear that you have some sort of tilt (top to bottom) in your image train.  I know you have tested the chip out extensively, and it is flat... so I wonder if it's in the rest of the train or focuser or what?? 

 

Regarding the "x" shaped stars.  I saw the same thing with my QTCC, but with the larger image circle of the FF I get nice and round.  These tiny pixels really push the limits of optics.  I'd be curious how your stars look at the native FL of the scope. 

 

Thanks for sharing. 



#3 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 09:08 AM

I got your PM. I really like QHY600M. I actually like their driver as well. I never had issues installing drivers even though it's a little quirky. Yeah, I was surprised that the image didn't saturate especially at 10 minutes with Lum filter. This is good and will help to reduce number of subs captured to save hard disk space. I think 10 minutes for LRGB will be my typical sub exposure times for DSO imaging. 

 

The tilt looks very minimal that I'm not sure I want to tweak further but I just reduced the imaging train by 0.5mm and see what happens. Roland usually say to reduce spacing for large sensors to help correct stars at the corners. Rockstarbill gave me a good suggestion to change spacing until plate solve matches close to 1008mm. My plate solve of M31 is 990mm.

 

Peter



#4 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 09:18 AM

While Roland  generally says to reduce the spacing for large sensors I ended up finding the opposite.  Spec spacing for my field flattener  and reducer is measured 103.9mm (with filters).  For the flattener I get perfect round stars at 112mm spacing and with the QTCC I could never get it dialed in but found that the best result was at 106mm spacing. 

 

I asked Roland about this, and he hypothesized that it was due to the filters.  Here is an excerpt:  "These 130 field flatteners are designed with edge rays diverging in order to fit as much light as possible thru the limited filter apertures. It may not be well known to astro-imagers that normal converging light rays are heavily vignetted when passing thru limited aperture filters (50mm diameter filters will not pass all the light into the corners of a 35mm full aperture chip if the light rays are converging). Because of this diverging geometry, any additional glass, such as filters and cover glasses, have an outsize effect on the field curvature. Therefore it will require somewhat longer distance (and slightly larger Barlow effect) in order to compensate for this additional glass compared to a glass-free optical path."

 

So that said, you might find the same relationship with your system as I did.  I think most of the previous large chip analysis was likely done with the 16803 chip and those 9um pixels can hide these kinds of issues. 

 

FWIW and to Roland's comment, I plate solved my image at 875mm FL.  (Spec is 873mm).  So if the relationship stands... for you to get a longer plate solved FL you would need to increase spacing. 


Edited by ChrisWhite, 01 December 2021 - 09:20 AM.

  • Peter in Reno likes this

#5 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 09:45 AM

Thanks for the valuable information. Looks like I have some work to do regarding to spacing. If I have to increase as much as 10mm, I would need longer screws for my imaging train. I don't use M54 threads, I use bolt-on or dovetail connections. The spacers I use is in between QHY CFW3 FW and OAG.

 

In my case, I'm using TEC 0.9x FRC which could behave differently and go opposite direction. Did you have to increase spacing for Quad TCC as well?

 

Thanks,

Peter


Edited by Peter in Reno, 01 December 2021 - 10:25 AM.


#6 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 09:46 AM

Thanks for the valuable information. Looks like I have some work to do regarding to spacing. If I have to increase as much as 10mm, I would need longer screws for my imaging train. I don't use M54 threads, I use bolt-on or dovetail connections. The spacers I use is in between QHY CFW3 FW and OAG.

 

In my case, I'm using TEC 0.9x FRC which could behave differently and go opposite direction. Did you have to increase spacing for Quad TCC as well?

 

Thanks,

Peter

Yes, I mentioned that.  Spec was 103.9mm, and I found that 106mm was the best. (I ultimately determined that the QTCC couldnt get perfect stars with the IMX455.)


  • Peter in Reno likes this

#7 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 09:51 AM

Additionally, if you are placing your spacing in between the OAG and the FW that could be why you are getting that funky artifact.  With my ZWO setup the OAG prism is only 43mm from the main imaging chip.  As I switch to QHY it will be 56mm.  I wonder if I will need to raise the prism a little.  For you, if raising the prism resolves the artifact, but impacts guiding, you might need to switch your spacing methods to get that OAG closer to your imaging sensor.  Just spit-balling here...


Edited by ChrisWhite, 01 December 2021 - 09:53 AM.

  • Peter in Reno likes this

#8 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 09:56 AM

One more trick for spacing that Roland has published, is that if you move the focuser OUT and corner stars get better than you need to REDUCE spacing.  Conversely, if you move the focuser IN and corner stars get better you need to INCREASE the spacing. 

 

With my steppers (Optec FT Motors) I am only moving the focuser 100 or 200 steps at most to see what impact this has on corner stars.  I'm also using voyager single star focus method so I can be sure that the focus point for the center of the frame is repeatable every time.  (Local field isn't be repeatable in my experience)


Edited by ChrisWhite, 01 December 2021 - 09:57 AM.

  • Peter in Reno and BenKolt like this

#9 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 10:07 AM

I wish it is summer now because it's a P.I.T.A. to deal with this in cold weather. coldday.gif

 

Peter 


  • lynnelkriver likes this

#10 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 11:21 AM

Do you know whether vignetting cause the stars to look bad? If so, it may be difficult or next to impossible to fix it.

 

Peter



#11 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 11:52 AM

Chris,

 

I've decided to increase spacing by 1mm. If you're wrong, you owe me three beers 🍺 🍺 🍺 . If you're right, you owe me one beer 🍺 . lol.gif

 

Peter


Edited by Peter in Reno, 01 December 2021 - 12:49 PM.

  • rockstarbill likes this

#12 lynnelkriver

lynnelkriver

    Viking 1

  • -----
  • Posts: 619
  • Joined: 14 May 2007
  • Loc: Elk River, Mn.

Posted 01 December 2021 - 12:33 PM

I got your PM. I really like QHY600M. I actually like their driver as well. I never had issues installing drivers even though it's a little quirky. Yeah, I was surprised that the image didn't saturate especially at 10 minutes with Lum filter. This is good and will help to reduce number of subs captured to save hard disk space. I think 10 minutes for LRGB will be my typical sub exposure times for DSO imaging. 

 

The tilt looks very minimal that I'm not sure I want to tweak further but I just reduced the imaging train by 0.5mm and see what happens. Roland usually say to reduce spacing for large sensors to help correct stars at the corners. Rockstarbill gave me a good suggestion to change spacing until plate solve matches close to 1008mm. My plate solve of M31 is 990mm.

 

Peter

Peter:  I'm going through the same process with my TEC140FL with QTCC and QHY268m trying to work out the spacing and tilt and I'm in the fine tuning stages like you.  I've been using the ASTAP image (CCD) inspector tools for looking at tilt and curvature (off axis abberation).  I'm curious what plate solver you use that gives you the focal length of the image (Rockstarbill suggestion)?  Thanks!  Scott



#13 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 12:44 PM

Any plate solver works. I use PixInsight to solve images. It gives me actual focal length results. 

 

Peter


Edited by Peter in Reno, 01 December 2021 - 12:47 PM.

  • lynnelkriver likes this

#14 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 12:52 PM

Do you know whether vignetting cause the stars to look bad? If so, it may be difficult or next to impossible to fix it.

Peter


I don't think so. My stars look great with the field flattener. Just required extra spacing. I think with reducers it's worse as the corrected image circle is reduced.

#15 Jared

Jared

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7,177
  • Joined: 11 Oct 2005
  • Loc: Piedmont, California, U.S.

Posted 01 December 2021 - 12:53 PM

Any idea what the spurious light bands are just to the left of center near the top of the frame? I'd be more worried about those than slight astigmatism in the corners... Could you have a very slight light leak? Or is that some sort of specular reflection off the edge of the guide prism? I hate tracking down those sorts of things.

 

Regardless, I'm glad your QHY600 appears to be doing a really nice job for you. That's a really satisfying first light image.


  • Peter in Reno likes this

#16 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 01:00 PM

Chris,

I've decided to increase spacing by 1mm. If you're wrong, you owe me three beers 🍺 🍺 🍺 . If you're right, you owe me one beer 🍺 . lol.gif

Peter


It's a deal, but if increasing spacing helps I suspect it will be more than 1mm.

I also suspect that if you can get a flatter field your tilt issue will look a little better.
  • Peter in Reno likes this

#17 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 01:42 PM

Any idea what the spurious light bands are just to the left of center near the top of the frame? I'd be more worried about those than slight astigmatism in the corners... Could you have a very slight light leak? Or is that some sort of specular reflection off the edge of the guide prism? I hate tracking down those sorts of things.

 

Regardless, I'm glad your QHY600 appears to be doing a really nice job for you. That's a really satisfying first light image.

That's most likely reflection from OAG's prism. If you stretch my Master Flats heavily, you should be able to see the prism shadow at top and a little bit left-center and it's directly above the reflection. I raised the prism and hopefully that will fix it. The prism shadow was easily corrected through calibration. 

 

Peter


Edited by Peter in Reno, 01 December 2021 - 01:46 PM.

  • Jared likes this

#18 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 01:45 PM

It's a deal, but if increasing spacing helps I suspect it will be more than 1mm.

I also suspect that if you can get a flatter field your tilt issue will look a little better.

I agree. It's a starting point. If the stars look better or focal length increases, then I know the right direction to go to.

 

Peter


  • ChrisWhite likes this

#19 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 02:09 PM

Additionally, if you are placing your spacing in between the OAG and the FW that could be why you are getting that funky artifact.  With my ZWO setup the OAG prism is only 43mm from the main imaging chip.  As I switch to QHY it will be 56mm.  I wonder if I will need to raise the prism a little.  For you, if raising the prism resolves the artifact, but impacts guiding, you might need to switch your spacing methods to get that OAG closer to your imaging sensor.  Just spit-balling here...

I've thought about what you said. The advantage of my method is it helps minimize distance between main camera's sensor and OAG's prism so that the guide camera is as close to the prism as well. Good thing PHD2 doesn't care about shapes of guide stars.

 

That's a pretty good amount of difference of back focus between ZWO and QHY cameras. Hope you won't have back focus issue.

 

Peter



#20 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 02:18 PM

Click on "Peter's Galleries" link in my signature to view updated pictures of my new camera and Optec Sagitta OAG. Click on "Astronomy" thumbnail, then click on "Current Equipment" thumbnail. 

 

Peter



#21 lynnelkriver

lynnelkriver

    Viking 1

  • -----
  • Posts: 619
  • Joined: 14 May 2007
  • Loc: Elk River, Mn.

Posted 01 December 2021 - 03:39 PM

Any plate solver works. I use PixInsight to solve images. It gives me actual focal length results. 

 

Peter

Thanks Peter for mentioning Pixinsight.  I did not know that capability was there!  I love your set-up and image looks great!  I plate solved some of my images in Pixinsight with various back spacings and thought I'd share the results.  The AP spec with QTCC for TEC140 is 99.1 +/-1 mm without the 18.3mm spacer.  Some time ago Roland from AP mentioned I should try adding 1mm-2mm as a starting spot with the QHY268M (APS-C sized sensor).  I used NINA for autofocusing each time I changed spacing and used ASTAP to look at tilt and off-axis abberation and got the following results (along with Pixinsight focal distance):

 

Back focus spacing  ASTAP off-axis abberation  ASTAP Tilt  Pixinsight Focal Distance

 

100.4 mm                          0.57                                   42%                    699.13 mm

101.4 mm                          0.44                                     9%                    698.23 mm

102.4 mm                          1.28                                   17%                    697.34 mm

 

Based on the off-axis abberation results it seems for my situation adding somewhere around 2mm is the cats meow.  Now for fine tuning as I'm guessing with QTCC its pretty sensitive to spacing as Chris has mentioned compared to a flattener.

 

The TEC140FL is 980 mm fl and with the QTCC at 0.72X= 705.6 mm calculated focal length.  I'm guessing there is a +/- some accuracy for the scope/QTCC/camera.  What I find interesting is that as I added to the back focus spacing the Pixinsight focal distance decreases.  Hope this is of interest.  Thanks!  Scott


  • Peter in Reno likes this

#22 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 03:50 PM

I've thought about what you said. The advantage of my method is it helps minimize distance between main camera's sensor and OAG's prism so that the guide camera is as close to the prism as well. Good thing PHD2 doesn't care about shapes of guide stars.

 

That's a pretty good amount of difference of back focus between ZWO and QHY cameras. Hope you won't have back focus issue.

 

Peter

The disadvantage though, is that because the prism is farther away from the main camera, it will have a larger impact on the light path for the same amount of insertion.  Getting it closer to the main camera sensor will allow you to insert it farther into the light path without the risk of causing diffraction artifacts in your images... if that is indeed what is causing the issue. 

 

While PHD can calculate a centroid even on poorly shaped stars, it does do a better job with rounder tighter stars.  That said, there is a lot more forgiveness using the multistar method as centroid variability (like seeing variability) seems to have a lesser impact on performance. 

 

Regarding the BF distance.. I have plenty to work with.  I'm using 20mm of spacers to get to 112mm (and that is with a Gerd in the mix).  So wont be a problem for me. 



#23 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • topic starter
  • Posts: 12,499
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 01 December 2021 - 04:55 PM

Chris,

 

Bottom line, are you saying that you may be concerned about guide stars being too far off-axis? If so, that's one of the disadvantage of full frame sensors or larger but I think it's a bit minor.

 

Multi-star guiding seems to help quite a bit because it was my best guiding ever.

 

Peter


  • rockstarbill likes this

#24 rockstarbill

rockstarbill

    Voyager 1

  • *****
  • Posts: 10,531
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 01 December 2021 - 05:17 PM

I've thought about what you said. The advantage of my method is it helps minimize distance between main camera's sensor and OAG's prism so that the guide camera is as close to the prism as well. Good thing PHD2 doesn't care about shapes of guide stars.

That's a pretty good amount of difference of back focus between ZWO and QHY cameras. Hope you won't have back focus issue.

Peter

I bet once the spacing issue is resolved you'll have much rounder guide stars.

Edited by rockstarbill, 01 December 2021 - 05:17 PM.


#25 ChrisWhite

ChrisWhite

    Cosmos

  • *****
  • Posts: 7,780
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 01 December 2021 - 05:37 PM

Chris,

Bottom line, are you saying that you may be concerned about guide stars being too far off-axis? If so, that's one of the disadvantage of full frame sensors or larger but I think it's a bit minor.

Multi-star guiding seems to help quite a bit because it was my best guiding ever.

Peter


That was part of it, but minor and secondary to resolving the giant diffraction artifact you have.


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