Excellent guiding but eggy stars

[Bigfoot 9907]

Greetings all, 

I have a RASA8, EQ6 pro and ASI1600MC pro camera, using PHD2 as my guiding software. I have recently switched from taking 1 minute subs to 3 minute subs and have noticed egg shaped stars. They are not noticeable on the full image but when zoomed they are very noticeable. I have redone the PA using sharpcap and it gave me an 'Excellent' PA. the PHD2 graph looks pretty good too. So not sure where the issue is. 

Also, i have noticed that the bottom left of the image is slightly out of focus compared to the rest of the image (N again, only noticeable when zoomed in).

I attach my PHD2 graph and also a zoomed in imaged showing the elongated stars. This is the centre of the image where i would expect round stars.

Any advice would be greatly appreciated

 

[Fegato]

Were the stars looking fine with 1 minute subs? That would seem very strange, as your guiding stats are superb - way better than I ever get! So with your small focal length you shouldn't see any problems with star shapes caused by guiding however long your subs. 

Anyway, I think if you have funny shaped stars in the centre of the field, my first thought would be collimation. Have you checked that? 

As for worse focus in the corner - this is probably due to some slight tilt. With the RASA having such a fast focal ratio, it is very prone to showing up tiny issues with the backfocus and tilt. With your small pixels the critical focus zone is really small - only 10 microns!  I have a RASA 11, and have spent a long time trying to dial all this in to get a flat field (albeit with a full frame camera too, which makes matters worse).  Not sure what your image train is, but I had to invest in an Octopi device to give me really accurate control over the backfocus and tilt. Focusing itself is also very critical - do you use autofocus?  If so, I would recommend the Hocus Focus plug-in with NINA - this has an aberration inspector which allows you to very precisely measure what is going on with the centre and each corner of your frame. This has been a real boon for me in providing a system that provides feedback on every change I make (it's a bit time consuming - you really need to re-run after every small change, so you have to set aside some time for it - good for when the moon is up!).

 

[michael8554]

Image scale is 1.96arcsec/pixel.

Guide errors are RA = 0.16arcsec,  Dec = 0.16arcsec.

Both well under the image scale, so stars should be round.

As they're not I suspect you have Differential Flexture between the imaging and guiding setups.

What is your guiding setup, you didn't say ?

Robin suspects Collimation, but you say stars are round on shorter exposures ?

Michael

[Fegato]

24 minutes ago, michael8554 said:

Image scale is 1.96arcsec/pixel.

Guide errors are RA = 0.16arcsec,  Dec = 0.16arcsec.

Both well under the image scale, so stars should be round.

As they're not I suspect you have Differential Flexture between the imaging and guiding setups.

What is your guiding setup, you didn't say ?

Robin suspects Collimation, but you say stars are round on shorter exposures ?

Michael

 

Ah yes, good shout Michael - that makes more sense, especially if it's changing with different exposures. 

(The RASAs are normally delivered very well collimated and most people never touch it. I did make a tweak to mine though. It's easy to to do a quick check with a star test anyway - if it's pretty close, it would not explain stars like that certainly).

 

[ONIKKINEN]

Your guiding looks a bit too good to be true, how are your PHD2 settings, specifically the focal length of the guider?

[ollypenrice]

8 hours ago, Fegato said:

 

 

(The RASAs are normally delivered very well collimated and most people never touch it. I did make a tweak to mine though. It's easy to to do a quick check with a star test anyway - if it's pretty close, it would not explain stars like that certainly).

 

Who told you this? You shouldn't have believed them!  The RASA needs not only excellent collimation but excellent chip orthogonality (freedom from tilt) and this tilt can come from the camera or the alignment of the optics or the attachments or all three. The whole system is insanely sensitive and, after a year of effort, we are still not fully sorted with ours.

The key test of guiding is easy with a RASA because it can catch so much signal so quickly. On the same target at the same time, just take a set of subs, getting longer as you go. 10 secs, 30 secs, 1 min, 2 mins, 3 mins. Does the quality of star shapes alter much? If it doesn't, it isn't your guiding (and I don't think it will be.)

Another RASA thing: with regular optics, being slightly out of focus means having slightly larger stars. With the RASA it's more subtle: a tiny fraction out of critical focus either way will hardly affect your FWHM (star size) but it will affect your star shapes in the corners.

The world of F2 is a strange one. You need to try it to know this.

Olly

[Bigfoot 9907]

Thank you gents for the feedback. I am using an Orion 50mm mini guide scope with the Orion Starshoot camera. I have just checked PHD2 and found an incorrect setting. the focal length is set to 300 when it should be 162mm. However, i don't think this would cause star trails, but instead the wrong RMS error readings.

I have taken 60 second and two minute subs with perfectly round stars, but this was a couple of months ago. My first attempt at 3 minute subs with the RASA 8 and I'm getting trailing stars, although they are not really visible when seeing the image as a whole as seen in the screenshot. I will try again tonight with the 60 second and 120 second exposures and see if I'm still getting round stars. 

Many thanks

 

[newbie alert]

Sorry if this sounds harsh as it's not intended but I don't think your eq6 r pro is capable of .23 arc secs per pixel ...otherwise no-one would buy top end mounts 

I'd check that you guidescope is fully tightened down to eliminate flex, and usually say to check that you're guiding on actual stars and not noise but your guide graph suggests that the mount is reacting to corrections

With a Rasa focus and spacing are much more critical 

[Adam J]

23 hours ago, Bigfoot 9907 said:

Greetings all, 

I have a RASA8, EQ6 pro and ASI1600MC pro camera, using PHD2 as my guiding software. I have recently switched from taking 1 minute subs to 3 minute subs and have noticed egg shaped stars. They are not noticeable on the full image but when zoomed they are very noticeable. I have redone the PA using sharpcap and it gave me an 'Excellent' PA. the PHD2 graph looks pretty good too. So not sure where the issue is. 

Also, i have noticed that the bottom left of the image is slightly out of focus compared to the rest of the image (N again, only noticeable when zoomed in).

I attach my PHD2 graph and also a zoomed in imaged showing the elongated stars. This is the centre of the image where i would expect round stars.

Any advice would be greatly appreciated

 

Really need to see an uncropped sub to make any conclusions. 

[Bigfoot 9907]

53 minutes ago, newbie alert said:

Sorry if this sounds harsh as it's not intended but I don't think your eq6 r pro is capable of .23 arc secs per pixel ...otherwise no-one would buy top end mounts 

I'd check that you guidescope is fully tightened down to eliminate flex, and usually say to check that you're guiding on actual stars and not noise but your guide graph suggests that the mount is reacting to corrections

With a Rasa focus and spacing are much more critical 

0.23 arc seconds was a one off, normally I'm getting between 0.30 and 0.5. Also,  I'll make sure the guide scope is secured down.

 

Edited June 27 by Bigfoot 9907

[Bigfoot 9907]

@Adam J here is an uncropped image. 

Target 3_180sec_1x1__0023.fit

[Fegato]

2 hours ago, ollypenrice said:

Who told you this? You shouldn't have believed them!  The RASA needs not only excellent collimation but excellent chip orthogonality (freedom from tilt) and this tilt can come from the camera or the alignment of the optics or the attachments or all three. The whole system is insanely sensitive and, after a year of effort, we are still not fully sorted with ours.

The key test of guiding is easy with a RASA because it can catch so much signal so quickly. On the same target at the same time, just take a set of subs, getting longer as you go. 10 secs, 30 secs, 1 min, 2 mins, 3 mins. Does the quality of star shapes alter much? If it doesn't, it isn't your guiding (and I don't think it will be.)

Another RASA thing: with regular optics, being slightly out of focus means having slightly larger stars. With the RASA it's more subtle: a tiny fraction out of critical focus either way will hardly affect your FWHM (star size) but it will affect your star shapes in the corners.

The world of F2 is a strange one. You need to try it to know this.

Olly

 

Hi Olly - I've used Hyperstar and now a RASA, and I agree with all you say!  I just meant that if the collimation is pretty good and you're well focused, it wouldn't explain stars like that in the centre of the frame. Agree entirely about the corners, I've spent many an hour trying to dial mine in.

 

 

[michael8554]

To my eyes the left and centre of the 180sfit has round stars, but the right side has elongated stars.

This is not the same as elongated stars across the whole frame, which I thought was how you had described the problem - my bad.

IMO we are now seeing Tilt.

The Orion Mini Guidescope has the insecure finder shoe attachment and clamp-screw securing of the guidescope, but seems to be good enough for this short FL imaging setup.

Michael

 

[Fegato]

Yes I agree - looks like tilt.  But the image you posted seems different to your original up top - looks quite good in the middle and in the region you showed near the tip of the trunk is much better?

Anyway - to me - bottom left looks nearer the sensor (stars pointing in) than top right (stars pointing around sensor).  A quick image inspection in ASTAP shows this ( flipped in ASTAP to show the right orientation)