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

Fixing Tilt Problems

  • Please log in to reply
44 replies to this topic

#1 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 10:27 AM

After many years of imaging with only very minor tilt problems, I seem to have gotten myself into a major issue with one of my systems. I have a TV127is scope coupled with an ASI071 OSC camera. A while back I took the camera apart to replace the dessicant plugs. I didn't understand that there were tilt push pull screws that needed to be all in the right place. When I put the system back together I had really bad tilt because I suspect that I rotated the tilt plate on the front of the camera. 

 

Previously I sent some other images to Televue and their reading was that this was a combination of tilt as well as the scope being pretty much at it's limit with an APS-C chip. They recommended getting the large field corrector which I have. I can't put that in the train, though until I get the tilt at least a bit under control Right now it's pretty awful. 

 

So, I completely reset the screws so that the plate was flush to the camera body and took some images last night in my back yard. I could not reduce the tilt numbers in any significant way no matter how I adjusted the tilt plate. I tried each screw in turn but that didn't do much other than make things a little better or a little worse. The center of the images look fine both by inspection and by measurement using MaximDL and CCD Inspector. 

 

So, while I don't expect perfection - an APS-C chip is about as big as you can expect to be useable with that scope, I did expect better results.

 

That leads me to ask a few questions.

 

1. What is your workflow to determine the direction of the tilt and which screw(s) to turn? 

 

2. Do you loosen all three of the pull screws and then just adjust the push screws that you think are the problem? Or do you do something else?

 

3.  Should I debayer the images before analyzing them? I'm not sure when measuring that the matrix doesn't distort the numbers.

 

4. What if any tools do you use when analyzing tilt?

 

5. Does anyone use a tilt adjuster instead? Which one?

 

Any help that I can get would be greatly appreciated. Right now, unless I'm imaging something really small in the center of my field, the results are just terrible. 

 

CCDI_Results.PNG

 

 

This is what the last run looked like. It's a bit worse than some of the others. I need some guidance around workflow more than anything else. 



#2 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,856
  • Joined: 27 Oct 2014

Posted 22 September 2021 - 10:39 AM

That CCDInspector plot is not tilt.  Tilt increases relatively smoothly in one direction (not necessarily horizontal or vertical).  That goes up and back.  It's some kind of weird curvature.

 

I can't see how adjusting the usual 3 point tilt plate would help.  If you can get one side better, the other will be worse.


Edited by bobzeq25, 22 September 2021 - 10:41 AM.

  • psandelle, ks__observer and OldManSky like this

#3 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 10:52 AM

I wondered about that. The scope was sitting in a closet (in the garage) for over a year since I last used it. It was not touched in any way. I just can't accept that somehow the optics deteriorated, but it is 11 years old now. All that the tilt adjustments seemed to do was move the tilt numbers around. I could get them somewhat better. But nowhere near acceptable across the frame. 


  • bobzeq25 likes this

#4 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 12,133
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 22 September 2021 - 11:55 AM

Take a look at this thread.

 

https://www.cloudyni...ilt-adjustment/

 

Look at post numbers 6 and 7 but read the whole thread. It's a lot easier to build the collimation rig and adjust for tilt indoors. 

 

I think this rig is a good investment for current and future cameras.

 

Peter


Edited by Peter in Reno, 22 September 2021 - 11:59 AM.

  • psandelle and rockstarbill like this

#5 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 12:13 PM

Interesting thread. I have to look more carefully at it to see if there are build instructions. I can figure out how to get a laser and some wood but I'm not sure how to size the bearing housing and the "thing" that holds the laser. My DIY skills (if I ever had any) are long gone at this point. Thanks for posting it.

 

Rgrds-Ross


  • psandelle likes this

#6 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 12,133
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 22 September 2021 - 12:19 PM

When you go to my web site about the collimation rig, click on 1st thumbnail and then click at bottom of image to reveal the parts list at left side of image. The links to parts are provided. 

 

https://peternagy.sm...ollimation-Rig/

 

Peter


  • psandelle and rockstarbill like this

#7 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 12:26 PM

Thanks, got it. That bearing is really expensive but cheaper than spending another night trying to figure out if tilt is the issue along with as Bob mentions something odd with the optics.  Once I get the tilt right then I've got the large field corrector which should cure the rest of the issues. 

 

I hate to ask, but how big are the plywood pieces? I've got some scraps in the garage that I can use if I don't need really big ones. 

 

Rgrds-Ross



#8 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 12,133
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 22 September 2021 - 12:39 PM

My base plywood is 9" x 17". It's not critical. 

 

Cheaper bearings may work as well. I hope they won't wobble too much while rotating the camera/bearings. Mine has a slight wobble so you have to train your eyes to compensate for the wobbling.

 

The holes to mount the bearings are 5/8" so I don't think you'll easily find these 5/8" bolts at local hardware store. Mcmaster.com have all necessary parts. Maybe cheaper bearings have smaller mounting holes. I recommend flat head (countersunk) bolts so they flush with bottom of plywood. 

 

Peter



#9 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 01:40 PM

I ordered the parts including the bolts! I'll go out to Home Depot later and see if they have some scrap plywood I can buy. I really don't want to buy a 4x8 sheet for this project. Thanks for all of this. I'm still thinking that there is some optical problem with the scope. I suspect that 1/2" will do just as well. 

 

I'm going to have another go at Televue about the situation. I sent the scope to Televue about 3 years ago. They tested it and said it was properly collimated. They star tested it if memory serves which may not suffice if one is using a large chipped camera. 

 

Earlier this week I put together the large field flattener that they sell onto the system and the results were not good. That led me to think that I needed to figure out if there was tilt and get rid of it first before trying to get the corrector working. So that's why last night I tried just to get the tilt out of the system. 

 

I read the Starlight Xpress article about using CCDInspector and I followed their exact protocol (by accident?). I slewed to near the zenith, then focused the telescope carefully using @focus3 in the SKYX. Then I took 5 10 second exposures. The graphic I showed earlier was from combining them into an aspect ratio display (may not have been obvious). Then I would make an adjustment, put the camera back on and repeat the cycle. 

 

Rgrds-Ross



#10 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 7,159
  • Joined: 28 Feb 2015
  • Loc: Northeast USA

Posted 22 September 2021 - 01:51 PM

It's still possible you have tilt. Ccdi analysis is helpful but it can get... confused...

I have some tips on addressing tilt but I'm on my phone right now and my thumbs are not well trained.

In the meantime please upload some raw files to take a look at. Seeing the stars is the only way to give a good "educated" guess as to what is going on.

#11 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 02:04 PM

I think that using the aspect analysis is wrong. I just took a look at the curvature graph  - in part based on what was in the other thread link that was posted - and it does not look bad at all. There's only a bit of tilt shoeing up, which I think I can correct with the screws. If so, then I'll try again with the flattener. 

 

I also got rid of the cheap UV/IR filter just in case and simplified the adapters, again, just in case.

 

Fortunately we are having night after night of good weather at my house!

 

https://drive.google...4NW?usp=sharing is a link to 5 of the bad images.

 

Rgrds-Ross



#12 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 12,133
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 22 September 2021 - 03:54 PM

Ross,

 

I think SX collimation document is incorrect. It says to use CCD Cover Slip Spot which which I don't think it's correct. The correct method is to pay attention to CCD Surface Spot (part of sensor array) while rotating the camera. The spot is usually closest to CCD CoverSlip. You may notice the Cover Slip may wobble a little bit while rotating the camera and that's normal. Because the Cover Slip rotates a bit unevenly, you may notice the Cover Slip may approach and overlap the target CCD Surface Spot and the CCD Surface Spot will eventually reappear as you continue to rotate the camera.

 

Another way to pick the correct Surface Spot is to see the Spot that's most stationary or has the least amount of rotation. The collimation is done when this Surface Spot appears to be stationary. The bearing may wobble slightly and use your best judgment to determine the camera’s sensor is collimated.

 

The angle of the laser pointing to the sensor is not critical and also where on the sensor is not critical either. Terry Platt of SX said his preference where the laser points to the sensor is slightly off center of the sensor. 

 

Peter


Edited by Peter in Reno, 22 September 2021 - 03:55 PM.


#13 KNak

KNak

    Vostok 1

  • *****
  • Posts: 159
  • Joined: 30 Aug 2011

Posted 22 September 2021 - 04:23 PM

May I request that your report back whatever the outcome?  I've been perusing past threads about tilt in trying to diagnose my own issues, and I'd estimate that at least 75% die without resolution (or were resolved but never followed up on).



#14 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,856
  • Joined: 27 Oct 2014

Posted 22 September 2021 - 05:21 PM

Does anyone sell a camera tilt measurer?  Seems like that's a great business opportunity.

 

BUT.  I daresay many (most?) tilt issues are not internal to the camera.



#15 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 9,374
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 22 September 2021 - 05:22 PM

Does anyone sell a camera tilt measurer? Seems like that's a great business opportunity.

BUT. I daresay many (most?) tilt issues are not internal to the camera.


There's one built into ASTAP.

Edited by rockstarbill, 22 September 2021 - 05:23 PM.


#16 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 05:27 PM

Here's the curvature file from CCDI. It shows much less tilt and I think I can just do some adjustments and make it better than that. This is looking more and more like an optics issue. 

 

 

TV127is_Curvature.PNG


  • bobzeq25 likes this

#17 bugbit

bugbit

    Vendor - Octopi-Astro

  • *****
  • Vendors
  • Posts: 341
  • Joined: 28 Apr 2015
  • Loc: west of seattle

Posted 22 September 2021 - 05:31 PM

Just a fyi. I have a copy of CCDI that I've been using for a couple of years and recently tried a copy of ASTAP with it's version of CCDI and I prefer the ASTAP as it shows the actual HFD's of the star field and as a bonus it is free. Thank you for it Han.

 

EDIT to add: That graph you just uploaded looks more of a back focus issue than a tilt issue.


Edited by bugbit, 22 September 2021 - 05:33 PM.

  • ezwheels likes this

#18 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 05:40 PM

I looked at ASTAP this morning and I could see that it showed some tilt. I could not figure out how to have it average things out over 5 exposures though. The tilt was from lower left to upper right which was also confusing. I'll look at it somce more now that I've downloaded it. Maybe more reading of the documentation is in order. 

Rgrds-Ross



#19 bugbit

bugbit

    Vendor - Octopi-Astro

  • *****
  • Vendors
  • Posts: 341
  • Joined: 28 Apr 2015
  • Loc: west of seattle

Posted 22 September 2021 - 05:46 PM

With ASTAP, you want to get the outer numbers as close to the center number for a flatter field and pay attention to the hfd's on the edges as well. I was playing with it last night and found that with my refractor the numbers went down as I moved the camera closer to the scope. Also you want a fairly rich star field to get a better example.



#20 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 05:53 PM

The TV127is is a Petzval design and the reducer is effectively built in. They say that, therefore, there is no sweet spot when it comes to focus. However, as part of trying to optimize things I went back to the recommended 2.4 inches of spacers. That's supposed to  keep the focuser almost completely racked in when using the scope. It would be a real hoot if it turns out that spacing between the end of the OTA and the camera chip is important when not using the flattener (large field corrector). 

 

Rgrds-Ross


  • psandelle likes this

#21 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 10,846
  • Joined: 12 Oct 2007
  • Loc: Melbourne, Australia

Posted 22 September 2021 - 05:54 PM

That CCDInspector plot is not tilt.  Tilt increases relatively smoothly in one direction (not necessarily horizontal or vertical).  That goes up and back.  It's some kind of weird curvature.

 

I can't see how adjusting the usual 3 point tilt plate would help.  If you can get one side better, the other will be worse.

I view this as completely consistent with tilt.  There should be a line of best-focus - and away from that line the fwhm should increase quadratically - i.e. it is like a parabolic trough.

 

But I would much rather see actual star images than these derived heatmap plots.

 

A simple way to tell if it is tilt or not is to change focus a bit in and out from perfect focus.  The line of best focus should move laterally with the focus shift - in a linear way.  At the same time, one of the sides should get much worse.

 

Once it is confirmed to be tilt you can try to fix it.

 

Frank

 

[Addendum]  Is that image showing fwhm or some estimate of tilt?  If it is fwhm then it looks like tilt - but if it is something else then bobzeq may be right.  This is why I like actual star images - or the 9-tile versions showing close ups around the edge and center.


Edited by freestar8n, 22 September 2021 - 06:00 PM.

  • jdupton and ks__observer like this

#22 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 06:01 PM

Frank - see post number 11. That has a link to my google drive so that anyone who wants a laugh can see what's going on.In the worst of them the stars in the lower left hand corner look like crescent moons. In that past when I've seen something like that there was an obstacle sitting in the tube. That was another reason that I changed out to a simpler imaging stack. I may even remove the off axis guider tonight but those have never given me trouble in the past. 

 

One night the brass compression ring in my visual adapter came loose somehow. I was using the scope for visual one night using the compression fitting. When I removed it and put up my camera somehow it detached itself and ended up stuck in the tube. All stars looked like gibbous moons. I could not figure it out. So I brought the scope home and the next morning in the lightI saw the ring. Problem solved.

 

Rgrds-Ross



#23 bugbit

bugbit

    Vendor - Octopi-Astro

  • *****
  • Vendors
  • Posts: 341
  • Joined: 28 Apr 2015
  • Loc: west of seattle

Posted 22 September 2021 - 06:41 PM

Here is the ASTAP for your a fit. It shows a bit of tilt in the lower left but the other 3 corners look really good.

The difference between the perimeter numbers and the central number agree with CCDI that there is some form of back focus issue. You stated with that design it won't matter but something is off.

 

 

a fit.JPG

 

It should be more like this one from last night only with lower hfd's like you already have.

Capture.JPG


  • ezwheels likes this

#24 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • topic starter
  • Posts: 9,485
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 22 September 2021 - 06:53 PM

Well, it's not supposed to be sensitive to back spacing. It just comes to focus at a particular point. All that happens is that you end up with the focuser moved further in or further out. You can't adjust the spacing. However, I'm going to use the "approved" spacing tonight (or very close to it) and if we see a big improvement, I'll have to eat my San Diego Chargers vintage hat.

 

If not then I think my next step to use the corrector and if that fails I'll insist that Televue put the scope on their optical bench. As I said, I think that they star tested it a while back but now that I think about it, I don't see how that would test that the corners are correct - maybe there's some other issue that doesn't show up in the star test. 

 

The stars in the lower right hand corner are terribly distorted. The scope is almost 11 years old and perhaps something has "slipped". 

 

Rgrds-Ross



#25 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 10,846
  • Joined: 12 Oct 2007
  • Loc: Melbourne, Australia

Posted 22 September 2021 - 06:54 PM

Frank - see post number 11. That has a link to my google drive so that anyone who wants a laugh can see what's going on.In the worst of them the stars in the lower left hand corner look like crescent moons. In that past when I've seen something like that there was an obstacle sitting in the tube. That was another reason that I changed out to a simpler imaging stack. I may even remove the off axis guider tonight but those have never given me trouble in the past. 

 

One night the brass compression ring in my visual adapter came loose somehow. I was using the scope for visual one night using the compression fitting. When I removed it and put up my camera somehow it detached itself and ended up stuck in the tube. All stars looked like gibbous moons. I could not figure it out. So I brought the scope home and the next morning in the lightI saw the ring. Problem solved.

 

Rgrds-Ross

Thanks Ross-

 

I see a lot going on in the a.fit image and it sure isn't just tilt.  It sounds like you did fix it?  Sounds good - 

 

Frank




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