Go Back   IceInSpace > Equipment > Astrophotography and Imaging Equipment and Discussions

Reply
 
Thread Tools Rate Thread
  #1  
Old 30-07-2016, 08:38 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Help with spacing... or something

I tinkered around for ages trying to get the spacing between my sensor and flattener correct. I finally got it to a point that I was happy with (see the "good" image). Measured curvature over the entire field was about 11.7% as measured over a number of subs, with CCD Inspector (trial has since run out).

Now, without changing the spacing between the flattener and sensor, my image quality is terrible (see the "bad" image), and I'm at a loss as to why.

The observant may notice the following differences:

Good example: 5s single sub, L filter.
Bad example: 300s sub, OIII filter.

Note that I've also eyeballed 5s subs with the L filter and the bad example is representative of current state: there's very obvious elongation across almost the entire field. I just didn't save any of those test images so I can't give exact numbers.

The only things that I can think of that have changed are: I've increased the spacing between the OTA and the flattener by placing my OAG in between the flattener and the focuser tube. Flattener to sensor distance has remained the same.

I've rotated the flattener + camera etc chain relative to the OTA. The flattener to sensor is still in the same orientation.

Anyone have any thoughts/suggestions on this? I'm stumped.
Attached Thumbnails
Click for full-size image (good.png)
111.0 KB23 views
Click for full-size image (bad.png)
88.0 KB18 views
Reply With Quote
  #2  
Old 30-07-2016, 08:58 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,787
field rotation maybe? How tight is your PA? do you get the same effect using a long exposure with another filter?

or if you use a guidescope, it could possibly be due to differential flexure and rotation,
Reply With Quote
  #3  
Old 30-07-2016, 09:05 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Cheers Ray. PA is ok, within a few minutes. Not perfect, but good enough. I am getting RA drift of about 7"/min. The direction of the elongation changes over the frame. Top left is "down", centre is diagonal (/) roughly in the direction of RA.

I get the same issue with subs between 5s and 300s (tried 5s, 150s and 300s) and across all filters (oiii, sii, ha, l, r, g, b).

OAG was used in all of the 150s and 300s subs, no guiding used for either the good or bad 5s subs.

Uploaded one of the uncalibrated 300s subs so the elongation / direction etc can be seen rather than just explained.

Guiding total RMS was about 0.63" over the 300s frames. Image sampling was 0.91"/px. In retrospect I needed to drop the min move and aggression, but the guiding wasn't overall terrible, and in fact, except for the elongation, I got some pretty good subs.

Last edited by codemonkey; 30-07-2016 at 09:31 AM.
Reply With Quote
  #4  
Old 30-07-2016, 11:30 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Interesting. Looks like the spacing is wrong afterall. Hopefully I can confirm tonight and all my problems will go away. I now have it within about 0.5mm of what I now believe to be the correct spacing.

Odd thing is, I don't recall changing anything between the camera and the flattener yet I had some very different results as posted above. Maybe I did and I've just forgotten, but I can't imagine why I would have.

Anyway, fingers crossed tonight gets me some good results.
Reply With Quote
  #5  
Old 31-07-2016, 08:17 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Looks like there's multiple issues at play here. I think I have the flattener spacing sorted now, but there's still (sometimes) large shifts in FWHM over the image. The magnitude of those shifts are not the same from image to image so I don't think it's tilt. Off the top of my head, one of the best subs from last night was 2 - 2.4, whilst a randomly selected rejected sub was 2.4 - 3.6

I assume this is what field rotation could look like (if it was consistent), as it kinda looks like some fixed point off the side of the image is sharp, but the further you get away from it the worse it gets.

Everything appears tight and I'm using an OAG. Still having some issues with the mount in RA... had been hoping it was just seeing, but I think I need to drop Avalon a line.

I'll go through the PHD log and check what the guiding looks like on the better subs vs the bad ones, might shed some light...
Reply With Quote
  #6  
Old 31-07-2016, 08:43 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,787
maybe also try an out-of-focus star test and make sure the collimation has not shifted. I imagine that there is a fine line between clamping the lens elements tightly enough and overdoing it to the point of pinching - maybe something is drifting around sightly in it's holder?.

However, I think that field rotation would still be be the most likely explanation

Last edited by Shiraz; 31-07-2016 at 09:08 AM.
Reply With Quote
  #7  
Old 31-07-2016, 09:39 AM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 5,345
Lee it is frustrating I know, I am still trying to sort out my Hotech Field Flattener with the ASi1600 on my refractor. Have the spacing exactly as suggested by the instruction sheet but it's out of focus, even allowed for the filter, and sensor coverglass in the path. I suspect the rear corrector element in my scope is buggering up the spacing norms.
Did you adjust the spacing to take into account the filter and coverglass in the path to the sensor? Applying the 1/3 rule? I reckoned that my 2mm filter and the 1mm coverglass (=3mm) required an extra 1mm of spacing. Most flatteners are ok within .5mm aren't they?
Reply With Quote
  #8  
Old 31-07-2016, 10:36 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Quote:
Originally Posted by Shiraz View Post
maybe also try an out-of-focus star test and make sure the collimation has not shifted. I imagine that there is a fine line between clamping the lens elements tightly enough and overdoing it to the point of pinching - maybe something is drifting around sightly in it's holder?.

However, I think that field rotation would still be be the most likely explanation
Thanks Ray. I've never needed to check collimation before, being a refractor guy. I assume I need high magnification for this?

I think rotation due to my RA drift seems likely at this point. PHD warns me pretty much every time I try to calibrate about unexpectedly varying RA & DEC rates and I can see it correcting RA a lot in the same direction.

Quote:
Originally Posted by glend View Post
Lee it is frustrating I know, I am still trying to sort out my Hotech Field Flattener with the ASi1600 on my refractor. Have the spacing exactly as suggested by the instruction sheet but it's out of focus, even allowed for the filter, and sensor coverglass in the path. I suspect the rear corrector element in my scope is buggering up the spacing norms.
Did you adjust the spacing to take into account the filter and coverglass in the path to the sensor? Applying the 1/3 rule? I reckoned that my 2mm filter and the 1mm coverglass (=3mm) required an extra 1mm of spacing. Most flatteners are ok within .5mm aren't they?
Cheers Glen. It's a pain in the arse, isn't it? I'm actually over 0.5mm with spacing, but I think the 1/3 rule makes that 0.5mm under (that's the way it works, no?) with my LRGB filters. Not entirely sure about the narrowband stuff but I care less about that...
Reply With Quote
  #9  
Old 31-07-2016, 04:48 PM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Ok, I'm less convinced that this is any kind of drift / rotation. The stars appear more bloated in the top left than the bottom right, and it's the fwhm that changes the most, not the eccentricity. I'd expect rotation to cause eccentricity (elongation) issues rather than fwhm (bloat) issues.

Attached is the aberration inspector and fwhm/eccentricity plots of the best and worst (accepted) L subs I got last night on the trifid.

This has me thinking it's more likely to be something drifting around in its holder.
Attached Thumbnails
Click for full-size image (bad.jpg)
192.6 KB21 views
Click for full-size image (good.jpg)
197.5 KB19 views
Reply With Quote
  #10  
Old 31-07-2016, 04:57 PM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Oooh. Hmm. Now that I'm toggling the images on IIS, you can easily see that the field has rotated, and the rotation is in RA. The stars in the top left in particular have bloated, but you can see most of the extra bloat is in the direction of RA, and the rest is probably accounted for by the fact that the best part of that image is still 25% more bloated than on the good image.

yay, looks like I have more mount issues. Man, I think I have the world's worst luck when it comes to mounts. Hopefully they can send parts, as going back to Italy (manufacturer) via Germany (retailer) will be expensive and take a lot of time.

Last edited by codemonkey; 31-07-2016 at 05:13 PM.
Reply With Quote
  #11  
Old 31-07-2016, 05:32 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,787
the other possibility is that it is a focus effect and that the top left corner is always a little out of focus if the centre is good. Then, if you have a little misfocus in the centre, the top left corner ends up waay out. That might mean that there is a little bit of tilt somewhere in the system.

I think that field rotation is due to PA - not sure that RA rate error will cause the field to rotate - ie mount may be OK.

Last edited by Shiraz; 31-07-2016 at 06:41 PM.
Reply With Quote
  #12  
Old 31-07-2016, 06:46 PM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Yeah, it wouldn't surprise me if there was some tilt in there as well.

I plate solved the image and the rotation is in RA... if it was polar alignment, wouldn't it be in DEC?

I may have solved my mount issue (there is one, whether it's causative of this issue or not). I had the camera plugged into my rig runner before the mount, so the camera was getting first dibs at the power. I swapped it tonight and for the first time in a long time, I didn't get the PHD warning about RA and DEC rates being different, and from what I saw, the RA corrections didn't seem like they were constantly in the same direction.
Reply With Quote
  #13  
Old 31-07-2016, 07:15 PM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is offline
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,321
To me it looks like tilt, having a gradient where the stars become more bloated. Not sure why but I've had tilt both show up as this while still having round stars AND had tilt causing what appears like spherical aberration (coma in a reflector).
Reply With Quote
  #14  
Old 01-08-2016, 05:03 AM
Rex's Avatar
Rex
Registered User

Rex is offline
 
Join Date: Jun 2010
Location: Townsville, Australia
Posts: 991
Hi Lee,
Mate just a thought, the rotational difference in the two subs you have shown above could be caused by dithering, especially if they were several subs apart. This hobby is soooo frustrating at times.
Reply With Quote
  #15  
Old 01-08-2016, 06:41 AM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Quote:
Originally Posted by Atmos View Post
To me it looks like tilt, having a gradient where the stars become more bloated. Not sure why but I've had tilt both show up as this while still having round stars AND had tilt causing what appears like spherical aberration (coma in a reflector).
Cheers Colin. Just seems unusual to me that in the magnitude is inconsistent, and the bloat seems to be mostly in the direction of RA, where I've been having tracking/guiding issues. Maybe Ray's onto something though in terms of the focus being a bit off on the worst one, worsening the effect. I'll try rotating the camera and see if it still appears in RA or if it moves.

Quote:
Originally Posted by Rex View Post
Hi Lee,
Mate just a thought, the rotational difference in the two subs you have shown above could be caused by dithering, especially if they were several subs apart. This hobby is soooo frustrating at times.
Cheers Rex, good thinking. Unfortunately I wasn't dithering during these captures.
Reply With Quote
  #16  
Old 01-08-2016, 08:05 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,787
Quote:
Originally Posted by codemonkey View Post
Yeah, it wouldn't surprise me if there was some tilt in there as well.

I plate solved the image and the rotation is in RA... if it was polar alignment, wouldn't it be in DEC?
I think that if you have a PA error, the image will rotate about the guidestar - both axes are involved. In OAG systems it will look like the image is rotating about a point outside of the field of view, but with ONAG or a guidescope, the center of rotation can be in the fofv.
Reply With Quote
  #17  
Old 01-08-2016, 04:07 PM
codemonkey's Avatar
codemonkey (Lee)
Registered User

codemonkey is offline
 
Join Date: Jul 2013
Location: Kilcoy, QLD
Posts: 1,754
Quote:
Originally Posted by Shiraz View Post
I think that if you have a PA error, the image will rotate about the guidestar - both axes are involved. In OAG systems it will look like the image is rotating about a point outside of the field of view, but with ONAG or a guidescope, the center of rotation can be in the fofv.
Ah ok fair enough, thanks mate. Admittedly I've always been a bit less concerned with PA than some people, partly because the adjustments on my old EQ6 were so difficult to make that once I got it within a couple of arc minutes I didn't think I'd be able to adjust it any better than by pure luck.

Total PA error is probably 3-4' at the moment. Alt is a bit awkward to adjust on the Linear, but Az is easier so maybe I can tweak it a bit more in the near future.
Reply With Quote
  #18  
Old 02-08-2016, 11:49 AM
PRejto's Avatar
PRejto (Peter)
Registered User

PRejto is offline
 
Join Date: Jan 2011
Location: Sydney, Australia
Posts: 1,290
I was just about to post the same comment that Ray made. The guide star determines how the rotating field looks. The rotation is around the guide star. You might want to keep that in mind if you are comparing images from night to night and have not used the same star.

That thought leads naturally to a question I've not seen considered. Just how important is it to select the same star from night to night (or year to year!) in terms of image registration? I imagine the effect would be smaller when polar alignment is quite good, but it's always present to some degree. Two different guide stars will give 2 different fields of rotation. This might preclude perfect registration. Obviously the effect must be quite small to vanishing or we would all be concerned with this I suppose.

Peter
Reply With Quote
  #19  
Old 02-08-2016, 02:01 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 15,468
What you have here is a bad case of tilt. I have been doing a lot of work on tilt in my systems over the last year. That's exactly what it looks like.
In your case the tilt is severe.

Stars bottom right are perfect. Stars bottom left are quite good. Stars top right and top left are terrible. In particular the top right.

Tilt can also be flexure to some degree but in your case because it is so severe it must be the adapters, connections to the filter wheel perhaps the focuser itself tilts.

I am assuming you are using your ASI1600 camera and a Skywatcher 120mm refractor.

How solid is the focuser? Does it move when you grab it and shove it a bit one way and then the other?

How is your camera connected to the filter wheel? The filter wheel to the focuser or any other item in between?

With a small sensor you are getting a zoom effect so it multiplies your focal length. Like a crop digital camera. So the effect is magnified somewhat.

So you are chasing done the wrong thing. Its not spacing and its not guiding. Guiding errors gives elongation over the whole image not just parts of it. Spacing errors does the same. Tilt affects one corner or side compared to the other.

Here is my procedure for correcting tilt. Its similar really to a collimation error and you will see it in the image where it suddenly just looks sharper and snappier when you correct it.

1. You need to orient the image to the physical camera. So the top right corner of the computer image is which corner of the camera. I orient by referring to the camera from looking at it from behind it.

The way I ascertain this is to get a dim torch or put a tissue or rag over the end of the torch, set the camera to take a 15 second exposure which allows enough time to put the torch in front of the scope.

Now put the torch so its shining on the top half of the lens. Now look at what the image displays. Is it showing the top half of the image all white or the bottom half? Chances are it will be the bottom half. So now you know that the bottom of the computer image is actually the top half of the scope.

Do the same for left and right. Chances are they are also reversed.

So at the end of this step you now know for CERTAIN which corner of the image is which corner of the actual camera. In both of my mirrored scopes with correctors both top and bottom and left and right of the actual camera is the reverse of what is displayed on the monitor.

If you don't do this step well you will fail at correcting the tilt and it becomes hit and miss and its like doing a drift alignment when you forget which way you are supposed to adjust the mount for different drifts.

2. Now you know the orientation then get some brass shim material from a steel shop or a spark plug feeler gauge from an auto shop (that has little metal spacers in a set from very fine to a bit thicker.
Start with something like .10mm. Put it in the space between say your camera and the filter wheel and tighten the camera up so its locked in.

Now take another short exposure at 1x1 binning (2x2 tends to disguise the fine errors of tilt so always end with a 1x1 even if you start with 2x2).
Check the corner that you were attempting to correct. Is it better or worse or the same? It should be better (this assumes the worst corner needs to be packed out not pulled in).

You could add another step here which is to focus the camera so that worst corner improves and see if you had to rack focus in or out to improve it. Hopefully it was to rack it out. If you had to rack it in then you need to pack out the opposite corner right?

So simply repeat these steps until you have all 4 corners showing round stars with no elongation/distortions.

I could probably do it in 10 - 15 minutes as I have done it a lot lately.

You can also use CCDInspector to help but its not that intuitive as it doesn't really tell you which corner to pack out. But if you've done step 1 well and you know which corner of your camera is what corner of the monitor image then you can work it out from CCDinspectors graphic.

I bet you I could grab anyones setup here apart from a few and find it needs some minor tilt correction. I think its very common and easily mistaken for tracking errors. I made that error earlier this year and spent a lot of time chasing down "tracking errors" only to find that despite improving tracking I was not handling it.

Another clue is odd shaped stars like triangular but that probably only occurs on mirrored scopes.

What can make matters more complex though is if the problem is flexure and it only occurs or gets worse at certain angles then its going to be a difficult problem.

So really the first step is too make sure everything is strong and tight and nothing shifts easily when you try to physically rock or move your camera.
Watch the focuser - does it have some play.

The longer the focal length the more this comes into play. Also the faster the F ratio of the scope. Also the larger the sensor but it also affects small sensors as well due to the magnification factor.

Another possible source of tilt is the sensor is not set orthogonal in the camera. FLI features this as an item in their cameras that they guarantee orthogonality.

Greg.
Reply With Quote
  #20  
Old 02-08-2016, 02:23 PM
glend (Glen)
Registered User

glend is offline
 
Join Date: Jun 2013
Location: Lake Macquarie
Posts: 5,345
Lee, there are several type of T-thread tilt adjusters on the market, they work like the GSO focuser collination ring but are much smaller obviously. Teleskop-Express have a couple that would suit. With the 1600, the face plate is, well should be, flat and the sensor perpendicular. You might check to make sure the sensor is not raised on one corner, i have not heard of any mounting errors by ZWO but you never know. Good luck
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 09:49 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Advertisement
Lunatico Astronomical
Advertisement
Bintel
Advertisement
NexDome Observatories
Advertisement
Meade Australia
Advertisement
OzScopes Authorised Dealer
Advertisement
SkyWatcher Australia
Advertisement
Celestron Australia
Advertisement
Astronomy and Electronics Centre
Advertisement