Why diffraction spikes more wide than they are tall?

[BrendanC]

Hi all,

Just processed this - it's supposed to be Mel 25, the Hyades cluster, although I think I probably either need to revisit my coordinates or put a mosaic together to get more of the cluster. I find it slightly underwhelming. I am underwhelmed. It underwhelmes me.

Anyway, my concern is really regarding the big bright star top left. I don't understand why the diffraction spikes from my Skywatcher 130PDS are wider than they are tall? Is this something wrong with my optical train? Or is this ok? I don't actually see how it's possible unless something isn't right.

Kit list is as per my signature.

Thanks, Brendan

[vlaiv]

I don't see your signature - that makes it a bit hard to determine focal length used to record this.

I'm also having trouble identifying stars in the image. Let me see what astrometry.net gives ...

Ah, yes, you are a bit off from your target.

Back to original question - it is in principle possible if you have more smear in one direction. Spikes get fainter as you move away from the star and if their light is smeared over more pixels - it becomes fainter faster.

This would usually be in RA or DEC - depending on which axis is causing trailing issues. With your image - this does not seem to be the case. Elongation is in horizontal direction - This is neither oriented with RA/DEC coordinate system, nor does it align with optical center (not towards nor perpendicular to line that joins star and center of the image).

Given that it is not guiding/tracking related (not aligned to RA/DEC), nor is has optical origin (not aligned to line joining star and center of the image) - then it must be processing. Fact that it is oriented up/down - left/right emphasizes this assumption.

Can you post raw linear image to see if effect is present there?

 

[Wyvern]

Hi Brendan,

As vlaiv notes, a bit off for the Hyades cluster. With a Field of View roughly 1 x 2 degrees, you'll only capture parts of it, so yes, a mosaic will be needed.  For reference, the bright blue star in your shot is Rho Tau  ( p ) and your field of view is roughly that of the stellarium screenshot No 1 below.  Screenshot No 2 shows the whole of the "V" of the Hyades cluster.

Regards,

Les

[wimvb]

2 hours ago, BrendanC said:

Anyway, my concern is really regarding the big bright star top left. I don't understand why the diffraction spikes from my Skywatcher 130PDS are wider than they are tall?

How are the spikes in single exposures? Long or short? If short, this means the spider may be twisted, or poor collimation. If long, the spikes are rejected /averaged out during stacking. Probably field rotation.

[BrendanC]

Ha! I knew it. You know, I did the Pleiades too last night and got it hopelessly off with that as well. I have a nasty feeling I copied the RA from one and the Dec from the other and, well, very accurately pinpointed two completely irrelevant areas of sky. Oh well - another learning: double-check your plans!

I also checked out Astrometry and Stellarium and that caused me lots of pondering and stroking of chin. Yeah... basically I got it wrong. Not the first time, won't be the last.

Anyway, thanks for the replies, really appreciate the helpful suggestions. I've attached one sub, and the stacked FTS file from Deep Sky Stacker. I totally get that my subs are probably not of the best quality, but if you could take a look that would be fantastic.

I don't think it's field rotation because going through the subs, apart from dithering, they're rock solid. I really hope it's not a bent spider. Would coma cause this at all, given it's at the very edge? It shouldn't because I have a coma corrector.

Anyway, over to you!

Autosave.fts L_9310_ISO800_60s__12C.CR2

[vlaiv]

I think you have combination of the two things (it might be one thing that fixes the other as well):

1. collimation issue

2. tilt in focal plain.

top left stars:

Ellipsis with strong bottom edge

Top right stars:

These two indicate spacing issue for CC (but it could be collimation as well as it tilts primary mirror)

Bottom right stars:

These look like small coma marks that have been written vertically (not consistent coma tail orientation - it should be away from the center but it's pointing "up" rather than "down")

Similarly bottom left of the stack:

This looks more like astigmatism then coma - and if it is coma - it is pointing completely in the wrong way (should be like 90 degrees to this).

This is causing vertical spikes to be split in two (this happens when you have poor focus):

Which makes them shorter.

I would first - check collimation without coma corrector and then see if things improve or do you need to deal with tilt as well.

[BrendanC]

Wow. Apart from that, it's perfect! 

I was aware I had some issues around the edges of the image which I was going to address, but didn't realise they might affect the spikes like this. 

Thank you so much for your help @vlaiv, you're a superstar. 

[Paul M]

Here is what ASTAP's CCD inspector utility makes of it. Certainly some tilt in there by the look of it. It's not a tool I've used much nor am I an expert in interpreting the result. I wouldn't use it on my own images. I don't need any more bad news!

 

 

[BrendanC]

Thank you @Paul M! Really appreciate you taking the time to do that. 

[BrendanC]

I just had a terrible thought,

About a week ago, I accidentally dropped the camera - only a couple of feet, against a relatively soft vinyl floor, not concrete. I only just remembered.

I've been checking previous subs from before that, and they're fine. 

So, could I have misaligned the sensor?

In which case, would the only really viable fix be to get a tilt corrector, something like this: https://www.firstlightoptics.com/adapters/astro-essentials-m48-t-tilt-adjuster.html

The next question would then be, given that I have, what, about 5mm backfocus left, would this even work with my setup?

I'm starting to get worried now...

Edited November 24, 2020 by BrendanC

[BrendanC]

So I've just been through old subs and looked at them through the marvellously useful ASTAP CCD inspector.

Interestingly, they all show tilt to varying degrees. It seems I've always had a small amount of tilt. This is both looking at subs with my previous camera, and with my recent replacement when the old one died (both EOS1000Ds).

However, even though they have tilt, they don't have the problems with the stars in the corners as per my latest subs, and the spikes are nice - sharp and even.

So this seems to point to collimation and focus being the issue this time, I think? I hope...?

It would also be handy to know whether a small drop of the camera could possibly increase the tilt - but as I say, it seems all my subs have a small amount.

Edited November 25, 2020 by BrendanC

[BrendanC]

So, in case anyone benefits from this thread, latest is that I've re-collimated it. I think the secondary wasn't quite aligned up properly relative to the focuser tube (I used my patent-pending technique for this as outlined here). 

Also, I'm kind of wondering whether my Bahtinov mask might be out. I have a feeling it's warped slightly, which could be damaging my focus. So, I might try and straighten that out or break the bank by buying another one.

Saturday night is looking promising round my neck of the woods so fingers crossed things look a bit better then...

[tooth_dr]

6 minutes ago, BrendanC said:

I'm kind of wondering whether my Bahtinov mask might be out. I have a feeling it's warped slightly, which could be damaging my focus.

My first Bahtinov mask was made from the back of a cornflake box - genuinely - cut out using a scalpel.  This lasted me for a few years.  It was well wonky, but didnt affect focusing so I think you should be ok.

  

On 09/08/2020 at 18:36, BrendanC said:

Then log in remotely from a laptop or PC to that camera using TeamViewer

Brendan do you mind explaining this?  My newt is pretty hard to collimate, and I used my phone in a holder like you have done.  However, it's only just possible to see the screen at a stretch, and working via TV would be ideal.  (I use AnyDesk, but perhaps the principle will be the same?

[BrendanC]

Thank you - another possibility struck off the list! Now it's down to tilt, collimation, and focus.

I think the collimation was off. Looking back through old subs, I've always had a small amount of tilt, which hasn't caused these issues. And perhaps I just need to be more careful with my focus (I think I trust the APT Bahtinov tool a bit too much, maybe I should actually use my eyes more).

I'm fairly confident the collimation is better now, so I just need a good night to see how things go. Fingers crossed. Everything crossed.

[BrendanC]

23 minutes ago, tooth_dr said:

Brendan do you mind explaining this?  My newt is pretty hard to collimate, and I used my phone in a holder like you have done.  However, it's only just possible to see the screen at a stretch, and working via TV would be ideal.  (I use AnyDesk, but perhaps the principle will be the same?

Ah, sorry, missed that. 

Yes, mine is hard to collimate too. In fact, I detest collimating it.

So, to get your mobile device's screen onto your PC or laptop's screen, first install TeamViewer, both on your laptop/PC, and the equivalent app on your mobile device (just search in your device's app market, and you also might need to install a supplementary app for your specific device).

Fire up TeamViewer on your PC, and set up your account.

Fire it up on your mobile device, log in, and then you should see it appear as one of your devices on the PC/laptop.

Then, when you've got the mobile in the bracket positioned over the collimation cap, and the image of the focuser tube and secondary mirror is viewable (using either your mobile camera's photo preview or video), go back to the PC/laptop, and click the mobile device to start controlling it. Your mobile device will ping and ask if you're sure you want the PC/laptop to take control, so you confirm that, and then your mobile device's screen should appear on your PC/laptop's screen.

Now fire up Mire de Collimation on your PC/laptop, and place it over the mobile device's screen, being shown on your PC/laptop screen. So you now have Mire de Collimation superimposed over the TeamViewer app on your laptop, which is streaming the image from your camera, on your mobile device(!).

Then, it's a case of pinching the mobile device screen to make it smaller or bigger, moving the Mire de Collimation app around, making it bigger/smaller, using more/fewer rings on it, until you have something that you can work with, to get the secondary mirror round, and centered in the focuser tube.

This is all achievable without this complexity I know, but I think it's more accurate, and when you've got it up and running, it's far easier, especially with replacement collimation thumb screws (which I just mentioned in an addendum to that post detailing this technique).

I hope this helps? Shout if not. I'd used TeamViewer a lot before, including to control my mobile device outdoors, so it was a fairly natural progression for me to then think about how it could help with this. Once you've got it working, it's fairly easy to get going with it again.

Edited November 25, 2020 by BrendanC

[tooth_dr]

26 minutes ago, BrendanC said:

Ah, sorry, missed that. 

Yes, mine is hard to collimate too. In fact, I detest collimating it.

So, to get your mobile device's screen onto your PC or laptop's screen, first install TeamViewer, both on your laptop/PC, and the equivalent app on your mobile device (just search in your device's app market, and you also might need to install a supplementary app for your specific device).

Fire up TeamViewer on your PC, and set up your account.

Fire it up on your mobile device, log in, and then you should see it appear as one of your devices on the PC/laptop.

Then, when you've got the mobile in the bracket positioned over the collimation cap, and the image of the focuser tube and secondary mirror is viewable (using either your mobile camera's photo preview or video), go back to the PC/laptop, and click the mobile device to start controlling it. Your mobile device will ping and ask if you're sure you want the PC/laptop to take control, so you confirm that, and then your mobile device's screen should appear on your PC/laptop's screen.

Now fire up Mire de Collimation on your PC/laptop, and place it over the mobile device's screen, being shown on your PC/laptop screen. So you now have Mire de Collimation superimposed over the TeamViewer app on your laptop, which is streaming the image from your camera, on your mobile device(!).

Then, it's a case of pinching the mobile device screen to make it smaller or bigger, moving the Mire de Collimation app around, making it bigger/smaller, using more/fewer rings on it, until you have something that you can work with, to get the secondary mirror round, and centered in the focuser tube.

This is all achievable without this complexity I know, but I think it's more accurate, and when you've got it up and running, it's far easier, especially with replacement collimation thumb screws (which I just mentioned in an addendum to that post detailing this technique).

I hope this helps? Shout if not. I'd used TeamViewer a lot before, including to control my mobile device outdoors, so it was a fairly natural progression for me to then think about how it could help with this. Once you've got it working, it's fairly easy to get going with it again.

Thanks for that Brendan.  Being larger too with definitely help. 

[BrendanC]

Imaging the Pleiades tonight, checking images as they come through and all is well. It was, I think, collimation and focus. I know what was wrong with the collimation, and I do think my visual judgement of the Bahtinov spikes is better than APT's Bahtinov aid. Onward and upward! 

[tooth_dr]

1 hour ago, BrendanC said:

Imaging the Pleiades tonight, checking images as they come through and all is well. It was, I think, collimation and focus. I know what was wrong with the collimation, and I do think my visual judgement of the Bahtinov spikes is better than APT's Bahtinov aid. Onward and upward! 

I personally do it by eye too, I think you can get a better feel for what is in and out of focus.

[BrendanC]

Hey, I just realised - you don't have to use TeamViewer at all if you can cast your mobile device's screen to your TV, for example if you have Chromecast. Just thought I'd mention that.

[BrendanC]

Just one more to add to this thread - as you can see, things are much better now (just a quick 30-minute of the Pleiades, not enough to pick up the lovely crinkly nebulosity but that was all the clouds would allow!)

So, much sharper, better diffraction spikes, but there is still a bit of divergence in the horizontal spikes which I can't see in the vertical.

I've played around with focus and it's not to do with that.

I did the unfocused star collimation test and it looked great, nice concentric circles.

So, is this still collimation, or tilt?

 

[tooth_dr]

1 hour ago, BrendanC said:

Just one more to add to this thread - as you can see, things are much better now (just a quick 30-minute of the Pleiades, not enough to pick up the lovely crinkly nebulosity but that was all the clouds would allow!)

So, much sharper, better diffraction spikes, but there is still a bit of divergence in the horizontal spikes which I can't see in the vertical.

I've played around with focus and it's not to do with that.

I did the unfocused star collimation test and it looked great, nice concentric circles.

So, is this still collimation, or tilt?

 

That’s nice. I’ll hazard a guess and say that your horizontal spider vanes arent perfectly in line, so you get two spikes that don’t overlap horizontally. 

[BrendanC]

Very good. Thank you! Time to check that too then... 

[BrendanC]

Right - I just read this missive: 

It's a three-pager, a tale of woe. I feel very sorry for @Spaced Out. And there is absolutely no way I'm going down that route!

So, I've tightened the vanes a little (which, naturally, means I have to check collimation yet again). I might do some investigative work such as retrieving a very old shot of the Pleiades I did over a year ago, to see if I had the issue then, and trying the camera without the coma corrector (and trying my guide camera on it too), to see if, having moved the focuser out of the optical field, that makes a difference. But I am not going to strip the thing down or chop bits off or move mirrors up for the want of absolutely perfect horizontal diffraction spikes. That way madness lies.

I'll also get in touch with @Spaced Out and see if he (or she) ever resolved this.

Btw, if it does turn out to be the focuser, I will not be impressed with Sky-Watcher. I actually know that the focuser draw tube does in fact impinge on the view, because, well, I can see it! Also I have the classic 'bitten' stars when I examine the subs. I just figured that, as it didn't really seem to affect the final image too badly, it was ok. If I have to live with slightly wonky spikes, then so be it. But really, for a company to produce a scope supposedly especially for astrophotography, and then for the tube not to offer enough backfocus with its own coma corrector, is pretty poor...

 

Edited December 2, 2020 by BrendanC

[tooth_dr]

On 25/11/2020 at 15:48, BrendanC said:

Ah, sorry, missed that. 

Yes, mine is hard to collimate too. In fact, I detest collimating it.

So, to get your mobile device's screen onto your PC or laptop's screen, first install TeamViewer, both on your laptop/PC, and the equivalent app on your mobile device (just search in your device's app market, and you also might need to install a supplementary app for your specific device).

Fire up TeamViewer on your PC, and set up your account.

Fire it up on your mobile device, log in, and then you should see it appear as one of your devices on the PC/laptop.

Then, when you've got the mobile in the bracket positioned over the collimation cap, and the image of the focuser tube and secondary mirror is viewable (using either your mobile camera's photo preview or video), go back to the PC/laptop, and click the mobile device to start controlling it. Your mobile device will ping and ask if you're sure you want the PC/laptop to take control, so you confirm that, and then your mobile device's screen should appear on your PC/laptop's screen.

Now fire up Mire de Collimation on your PC/laptop, and place it over the mobile device's screen, being shown on your PC/laptop screen. So you now have Mire de Collimation superimposed over the TeamViewer app on your laptop, which is streaming the image from your camera, on your mobile device(!).

Then, it's a case of pinching the mobile device screen to make it smaller or bigger, moving the Mire de Collimation app around, making it bigger/smaller, using more/fewer rings on it, until you have something that you can work with, to get the secondary mirror round, and centered in the focuser tube.

This is all achievable without this complexity I know, but I think it's more accurate, and when you've got it up and running, it's far easier, especially with replacement collimation thumb screws (which I just mentioned in an addendum to that post detailing this technique).

I hope this helps? Shout if not. I'd used TeamViewer a lot before, including to control my mobile device outdoors, so it was a fairly natural progression for me to then think about how it could help with this. Once you've got it working, it's fairly easy to get going with it again.

Finally got around to doing this, worked a treat.  Cheers