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

My Journey with RASA BackFocus, Image Sensor Tilt, and CCD Inspector

  • Please log in to reply
18 replies to this topic

#1 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 21 July 2021 - 11:03 PM

It seems like just about everything in this hobby is a journey. For me, owning a RASA 8 is a part of that journey. It's both a joy and a challenge. I like my RASA 8. In fact, I would even say I LOVE it. But wow - I would have to imagine that it's about like having a loving spouse who is beautiful, playful, picky, stubborn, and WAY moody all on the same day. Here's what I'm learning so far about backfocus, image sensor tilt, and CCD Inspector, in no particular order, after 7 months in the journey:

 

1) Don't pass up the journey with an f/2 scope just because the critical focus zone (or CFZ) is measured in microns. (CFZ is the distance encompassing the range of focuser positions for which the telescope's image will still perceived as being in focus. "Depth of field" is actually a related term in normal photography, so to speak.) But don't take on an f/2 scope if you are faint of heart. If you're curious about calculating critical focus zone, remember, you can calculate your CFZ by using this formula:

 

CFZ (in microns) = focal ratio2 X 2.2

 

The bad news, as you can see by glancing at that equation, is that... as the focal ratio becomes a smaller number (representing a larger aperture for the scope and, thus, letting in more light more quickly and more efficiently), the critical focus zone gets smaller exponentially. It's an INVERSE square relationship. It's not just slightly harder to find focus. It's exponentially harder -- literally. The good news is, if there is any, is that, fortunately, in the real world, it is said that the CFZ is actually perceived to be 10-30% greater than the theoretical values. Awesome. I'll take all I can get. I need it. (I'm the one that puts the "A" for Amateur in the phrase, Amateur Astronomy.) This basically means that instead of 8.8 microns, the CFZ on a RASA 8 is actually more like 11 microns. That's 11 millionths of a millimeter or .011 millimeters. To put that in a different perspective, that's less than 1/2 of 1/thousandth of an inch. A human hair is around 17 to 70 microns thick. To focus a RASA, you have to be so good that you can rack it down to 1/2 of the thickness of a human hair.

 

2) If your image sensor is slightly tilted (say, by the thickness of a human hair), then one end of your image is going to be completely out of focus. Wait. Let me put the word "completely" in bold. One end of your image is going to be completely out of focus. hahahaha I'm not kidding. The thickness of the felt strip on the back of the stock camera adaptor/holder for a Celestron RASA 8 is WAY thicker than a human hair, sadly. If that felt wears down by 50% on one side of that adaptor plate, welcome to the world of wacky star shapes. They'll look like seashells, chevrons, arrows, shrimp, or -- worse -- they'll just look, in general, like malformed blobs.

 

3) If your image train is at all mixed up - say, by the width of a human hair? -- maybe because one of your spacers has threads and the guy who machined it didn't happen to run the threads as deeply on your part as he did on the last part, your backfocus is going to be off, meaning that your image will no longer be projected on a flat plain. Instead, corners, edges, and maybe the entire image will be puckered, warped, and mis-aligned.

 

4) The focusing mechanism itself (deep inside the portion of a RASA's "business end" -- where the primary mirror is held) can, itself, "shift" or slightly tilt the mirror. Remember, all it takes is 1/2 the thickness of a human hair.

 

5) CCD Inspector is a piece of software that can measure, expose, and illustrate how bad your image really is. So basically, you're paying $179.95 to purchase a really bad day. : ) The GOOD news is, CCD Inspector is available as a free trial for 30 days. The BAD news is... Wait... let me put the word "bad" in bold. The BAD news is - it will take you longer than a month just to underSTAND your RASA's tilt, let alone address it. (My thinking is -- conspiracy theorists will eventually determine that the Celestron RASA's development costs were totally sponsored by the makers of CCD Inspector.)

 

6) With CCD Inspector, it helps to have a wonderfully-rich star field of equally distributed and equally bright stars. Finding that is like looking for a needle in a haystack. The best star field I've found is to center on the random star, HD 192911 (which is near Sadr). No other star field in the sky will work. (hahaha I jest. But that star field DOES seem particularly good.)

 

hd192911-field_of_view-500.jpg

(Above: The star, HD192911, just happens to take us to a field of view that CCD Inspector really seems to like. Please make a note. : )  In this image using Sky Safari, the field of view rectangle is that of a ZWO ASI2600MC Pro imaging through a RASA 8.)

 

7) With CCD Inspector, it helps to take 5 images for each position or change that you'd like to test. (By that I mean - every shim, every spacer, every slightly-turned device, every adjustment - essentially, everything that's keeping you from doing the observing you wanted to do but can't yet because your stars still look like shrimps.) It helps if those 5 images are slightly different exposure lengths. For example, 10 secs at gain 300, 20 secs at gain 200, 30 secs at gain 200, 30 secs at gain 100, and 20 secs at gain 100. Then open those five images in CCD inspector's image list, highlight them, then click on the "Curvature" button to determine your imaging train's percentage of curvature at any given position or adjustment.

 

RASA-tilt-2021-07-21-500.jpg

 

8) With CCD Inspector, it helps to keep track of your adjustments. For instance, Octopi-Astro recently contacted me to ask if I'd beta test their new camera holder device for the RASA 8. (Yes, I knelt in prayer to give thanks to God.) We've gone through a couple of design changes with them and I had to ship one of my RASA 8 parts to them so they could spec out one of those design changes but, the great news is, they've developed a design that essentially allows a RASA 8 owner to (almost) infinitely adjust for all these values -- backfocus and image tilt included. I'm still learning to USE that device, mind you - but I truly believe the potential is now there to fix all these problems. Wait. Let me bold the word "all." The potential is now there to fix all these problems... permanently. Here's a chart from last night's callibration testing session:

 

RASA-curvature-and-tilt.jpg

 

It seems like the key columns to include would be

  • Position (I arbitrarily named each position according to the changes on the adjustment screws. But just today, Octopi-Astro suggested I start using a Vernier caliper tool to measure the "push" screw on their device to determine a more tangible "name" for the position. I'm really looking foward to that prospect.)
  • Action (the adjustment it took to get to the measurements in this row) - for example, 1/2 turn "in" or 1/4 turn "out"
  • Which adjustment screw I adjusted to get there.
  • How many screws (e.g., in the case of backfocus, it's usually all 3 adjustment screws)
  • Percentage of Curvature measured for this position
  • Tilt for this position
  • The direction of the tilt (as reported by CCD Inspector)
  • The readout of one's focus motor (which is more or less essential for this entire process).
  • The number of images used for these measurements. (For me, it was consistently always 5.)

At the bottom, as you can see, there's a graph illustrating the changes in curvature and tilt.

 

9) It's helpful to focus on backspacing first. (Pun intended.) We shouldn't worry about tilt until backfocus of "around 20%" has been achieved. My new rule of thumb is "around 20%" means anything near 20%. Wait, let me put the word "near" in bold. : ) hahahaha Anything near 20% is, from now on, fair game. For example, after working on this for three nights, with an average of 5 hours each night in testing, if I can just get back to position 13.5 with a focus of 40300 -- or position 15 .. or anything close to it, I'll be in backfocus heaven.

 

10) Astronomy is a journey. Enjoy the journey - or it might drive us all crazy. haha (By the way, if I can find position 13.5 or position 15 or ANYTHING close to it on the next clear night, I'm going to observe, not test. Note "observe" is in bold.)

 

Once Octopi-Astro has finalized this device, it'll be available at their website at https://www.octopi-astro.com/. Make no mistake: Octopi Astro's parts are machined at incredible levels of precision. (Wait... let me bold incredible. Truly.) Their customer service is amazing. I don't know what this camera holder is going to cost. But RASA lovers who love their RASA will want this for Christmas -- there is no doubt. They didn't pay me to say this. In fact, they didn't pay me at all. In fact, the deal they presented was - if I decided to keep the part, I'd have to pay at least for the parts. But the truth is - RASA lovers are just going to want this. This is the set of adjustments we have all wished for on a RASA 8. Artesky filter holders are nice. But they let us adjust exactly zero factors. (Wait - let me bold the word zero.) The Octopi-Astro device is going to let us adjust everything, including lateral placement of the camera. I'm guessing the device will be available soon. Octopi-Astro seems determined to serve this need.

 

For more on CCD Inspector, visit https://www.ccdware....s/ccdinspector/. My next step with their product is to learn to use the Aspect Ratio Mapping tool.

 

For a video discussion about all these factors, see this Youtube video, below. I suggest you watch it on some kind of device that will let you watch at 1.5 speed or something similar, because it took me about an hour to talk through all these issues and demo the work flow at the telescope in the field under the stars.

 

octopi_camera_holder-review-screenshot-500.jpg

 

https://www.youtube....h?v=or7jRVgcgGE

 

Thanks to CloudyNights for being the greatest bunch of guys under the stars - to inspire me when I'm frustrated about sensor tilt! : )


Edited by EmeraldHills, 21 July 2021 - 11:06 PM.

  • dswtan, Bob Denny, mikenoname and 5 others like this

#2 hornjs

hornjs

    Messenger

  • -----
  • Posts: 474
  • Joined: 04 Sep 2020
  • Loc: Billings, MT

Posted 21 July 2021 - 11:23 PM

Oh my gosh Doug what a journey indeed!


  • EmeraldHills likes this

#3 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • Posts: 9,253
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 21 July 2021 - 11:29 PM

You have it backwards.  Dave Rowe who designed the RASA for Celestron (and then the PlaneWave CDK line) did it to increase sales of CCD Inspector. This is well known. 

 

Rgrds-Ross


  • EmeraldHills, Jim Waters and Bob Denny like this

#4 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 125,225
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 22 July 2021 - 04:41 PM

Doug, Thanks for putting this all together for us. I think that I'm now sufficiently scared of ever touching a RASA, even though I occasionally dream of one. smile.gif


  • EmeraldHills and psandelle like this

#5 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 22 July 2021 - 10:59 PM

Just remember, Errol, there was probably a point in your past in which everything about your current scope seemed a bit overwhelming, right? : ) Like I say, seems like ALL of this hobby constitutes a journey. : )


  • alphatripleplus likes this

#6 alphatripleplus

alphatripleplus

    World Controller

  • *****
  • Moderators
  • Posts: 125,225
  • Joined: 09 Mar 2012
  • Loc: Georgia

Posted 23 July 2021 - 07:05 AM

That's a valid point, Doug. What I find interesting here though is the level of the challenge - new software and hardware adjustments  -  to tune this set-up. 


  • EmeraldHills likes this

#7 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 23 July 2021 - 08:41 AM

I agree. I think one of the lessons learned is that "with great aperture comes greater focal challenge." Just ordered a precision digital vernier caliper. Hoping that will help me measure the height of these grub screws so I can more precisely quantify the position of the sensor relative to the primary.



#8 rgsalinger

rgsalinger

    Cosmos

  • *****
  • Moderators
  • Posts: 9,253
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 23 July 2021 - 01:52 PM

I don't think that aperture is really the issue. I think that the issue is that pesky F number. At F2 any tiny tilt or bad spacing is going to be very apparent. By the time you get to F5, not so much. My F8 scope worked right out of the box because, I'm certain, it's F8 and the amount of out of focus is tiny even with a bit of tilt. 

 

This is my 12.5" CDK on a good but not great night with a full frame chipped camera. All that I've ever done is screw things together. 

 

aftercollimation.PNG

 

I totally doubt I could ever get this with an F2 scope. So, my hat is off to you for your diligence.

 

Rgrds-Ross


  • psandelle likes this

#9 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 23 July 2021 - 08:26 PM

When I said, "With great aperture," I meant exactly that -- low focal ratio -- f/2. I'm with you, Ross. It's super-challenging.



#10 Ron in Michigan

Ron in Michigan

    Ranger 4

  • *****
  • Posts: 360
  • Joined: 20 Feb 2008
  • Loc: W. Michigan

Posted 24 July 2021 - 09:52 AM

Keith is killing it - I had a similar journey over a RASA 11 V1 and V2 over about 3 years.   I then found the (beta)  Octi-pi tilt plate unit (back then) for my RASA V2 and problem solved.
He's updated me a couple times.  RECOMMEND HIGHLY.

IMG_7827.jpg
 

 


  • EmeraldHills likes this

#11 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 24 July 2021 - 04:43 PM

Ron, I can spot his handiwork in your unit. Similar craftsmanship. Glad to hear it's working for you. yay!



#12 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 2,158
  • Joined: 09 Aug 2016
  • Loc: Ellensburg, WA, USA

Posted 30 July 2021 - 03:04 PM

I use and make a micrometer head adjuster for the RASA 11. The micrometers let you jot down how much you have actually moved each adjuster.
  • EmeraldHills likes this

#13 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 30 July 2021 - 07:16 PM

That's awesome, 555aaa. I picked up a caliper. I hope I can learn to use it to measure adjustments I'm making --- starting Monday night.



#14 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 08 August 2021 - 07:12 AM

Updated: The Calipers helped a great deal for me. See this video:

 

https://youtu.be/2_0gvHzMKrA

and this

https://youtu.be/l0SVShuIAu8

 

At last. I can breathe again.



#15 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 24 August 2021 - 05:10 AM

Updated: A subscriber to my YouTube Channel ....  (Gasp: You didn't know I had a YouTube Channel????      hahahahaha)

http://www.YouTube.c...eraldHillsSkies

Anyway... A subscriber to my YouTube Channel asked for a better/different tutorial for using CCD Inspector. (By the way, thankfully, I'm not affiliated with the company. If I were, I'd suggest that they charge a lot less for their software!!! hahaha) For whatever it's worth, here's the video tutorial I did night before last:

 

CCD-Inspector-tutorial-500.png

 

https://youtu.be/fCQZKJ_oLMw

 

Just remember - at this point, I'm an explorer, not a guide. And I'm certainly not Spielberg. : ) But this video does at least walk through the process I ended up using to help me enjoy my RASA scope. The corners (star shapes) now at least resemble the center. haha


  • dswtan and psandelle like this

#16 SanjeevJoshi

SanjeevJoshi

    Viking 1

  • *****
  • Posts: 835
  • Joined: 07 Sep 2020
  • Loc: Longwood FL

Posted 25 August 2021 - 02:36 AM

Great info as usual Doug.   I have found that CCD inspector readings are finicky, but useful as a quick check.  Pix insight is a bit more work but more repeatable.

 

I found that by nailing back focus to within 0.2 mm (camera, filter, adapter), most of the “alleged” tilt goes away.   I have posted some pics in my past threads.   What is left is so small, the level of focus issue left for a corner is smaller than a focus change in a few minutes.     Not noticeable at all.  Some people start adjusting tilt or collimation without nailing back focus, and it’s a tough spiral after.   While a lot of people in the early RASA threads from a couple years ago were complaining of tilt, a wise and quiet one educated me on back focus and the impact it has.   Back focus tuning is one time work per combo (camera, filter, adapter).

Once you nail back focus, the next key thing is to do the best focus - even with auto focus type adjustment, you can optimize focus for a central object or a wide field depending on type of object.

 

Excellent point about CFZ.   Using an electronic focuser, the CFZ will lie within roughly 20 to 50 units of position (depending on gear ratio and index)   A simple analogy could be using GPS to find a city block, not a specific house or entrance.   Not that hard to pin I think, although GPS can sometimes get lost smile.gif.  With careful eyes, easy to identify and fix.

 

With these two tasks (nailing back focus, focus optimally), RASA 8 can exceed your expectations!  So far I am using the factory collimation and no tilt plate.  I do own various adapters and tilt adjusters, I just stick to the basics.


Edited by SanjeevJoshi, 25 August 2021 - 03:06 AM.

  • EmeraldHills likes this

#17 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 25 August 2021 - 05:34 AM

I hope you can hear that coming through if you get a chance to watch the video, Sanjeev. At one point, while I'm bragging about how happy I am to have reached a tolerable back-focus, tilt, and focus, I actually say, "I know this sounds like a commercial for Celestron RASAs, but I just LOVE this scope!" hahahaha I couldn't agree more. I'm at that point - of loving it.

 

For me, being such a rank amateur, it took 5 different nights of work -- throughout 5 or 6 weeks. But I think it could have been cut short a great deal more if I'd just known to purchase Vernier calipers from the beginning. Once Octopi-Astro recommended them and I figured out how to use them, it actually was just one hour before I reached a tolerable level.

 

I say tolerable level because it's not quite *PERFECT* yet. But at 27% field curvature (back-focus is close but not perfect) and 16% tilt, all of the "seashell" and "tadpole" shapes are gone, I'm happy to report. The star shapes in the corners of my images are acceptable again.

 

I never heard for sure. Did all of your gear arrive in good form in your new location in Florida? You're doing ok?


  • SanjeevJoshi likes this

#18 SanjeevJoshi

SanjeevJoshi

    Viking 1

  • *****
  • Posts: 835
  • Joined: 07 Sep 2020
  • Loc: Longwood FL

Posted 25 August 2021 - 07:03 AM

Yes arrived and settling in finally, thanks for asking!

 

Good news on your end wrt star shapes!

 

Using CCDI, pointed at a non descript part of the sky, I consider this typical with my factory set up — 10% - 20% curvature with 0% to 10% tilt.  I am giving ranges because CCDI does not have repeatable results even when averaged over a reasonable number of captures.

 

I used to have numbers like yours or worse until I worked very carefully on back focus.   If you use Baader spacer rings, you can get it done well.

 

Just as a rough starting point, I found that all three of the ZWO cameras (including the 2600) set to  nominal back focus (add up the lengths) needed one 0.3 mm Baader spacer.   In one case, I got similar results with both 0.3 mm and 0.5 mm, so I left it at 0.3 mm.

 

Good luck, and thanks for sharing!


  • EmeraldHills likes this

#19 EmeraldHills

EmeraldHills

    Apollo

  • *****
  • topic starter
  • Posts: 1,451
  • Joined: 26 Nov 2006
  • Loc: Louisville, Kentucky

Posted 25 August 2021 - 08:21 AM

Yes, I'll get out there eventually to fine-tune. Just glad to have things in acceptable tolerances.

Thankfully, I won't those spacers because I have the Octopi-Astro camera holder - which is infinitely adjustable along every axis.


  • psandelle likes this


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