9 replies to this topic

[radu.marinescu]

Hello.

So I took my ASI294MM PRO for first light last night when I caught a few hours of clear skies. Decided to try the RED filter first. However the images look somewhat distorted. This night was not intended for actualy imaging but rather an experimental shoot.

This is a stack from 7 x 420sec exposures. As you can see, the image is a little blurred (although I remember focusing well with a bahtinov mask) and the details are not as sharp as I expected from this camera. Obviously the camera is well, there is something with the other instruments.

- First thing I notice are overall bloated stars. I don't know if this is the result of unproper functioning coma corrector, or oversampling from the 2x2 binning I shot these with (in relation to the seeing conditions) and/or lack of exact focus. Or maybe unproper collimation although I used a laser collimator 3 weeks ago and perfectly aligned the mirros. Can the RED filter also cause bloated stars (haven't tried the luminance one yet). Or it could be little from every reason I mentioned. Is there any-way to pin point the problem?

- Second thing I notice are elongated stars, especially in the top-right corner. This effect can be seen in the other corners as well but it's extremely proeminent in the top-right one. Weirdly enough even the stars above the galaxy seem just a little elongated. The camera is attached with its standard 1.25" adapter (17.5mm), T2 extender (21mm), M48-T2 (16.5mm) and then right in the corrector, so it creates a total of 55mm back focus. I just read that Baader Planetarium recommends a 57.5mm backfocus for M48 adapter, could such a small difference of 2.5mm be the cause for the elongated stars? By the pattern of the elongation (towards the somewhat center of the image) it does look that the sensor is too close but can someone confirm that such a small difference can create such bad results? Or is there a different cause?

- Third thing I notice is unusal vignette in the top-left corner. Could this be tilt for the coma corrector or focuser? I know the filter is improbable of creating vignette because it's attached directly to the 1.25" adapter of the camera so it's closest to the sensor it can get. Is there a way to correctly diagnose and adjust tilt?

P.S. The guiding was great, sitting around 0.4-0.5 total RMS outside the dithering intervals and the telescope is in my portable dome so wind or other such factors are out of the question. Also no calibration frames were shot.

Relevant equipment: SkyWatcher 250PDS/1200mm f/4.8; EQ6-R Pro, ZWO ASI294MM Pro, Baader 3 MPCC Coma Corrector, ZWO LRGB filter set.

Can anyone who had these issues guide me towards solving or diminishing them? Thanks a lot.

Edited by radu.marinescu, 20 March 2023 - 06:42 AM.

[Oort Cloud]

The only thing I see wrong here is poor tracking (elongation). Field has some bright stars and some dim ones. Global stretching will always cause the brighter ones to swell up, star reduction techniques in post-processing can help, but most of those techniques require the stars to be round for good effect.

So we're back to square one, the tracking needs to be fixed before anything else is really worth spending time or energy on.

Edit: I just noticed you said guiding was
.4-.5" all night. If that's the case, then either your pixel size/focal length in PHD2 is wrong, leading to incorrect RMS calculation, or your guidescope is moving relative to the main scope (differential flexure), and needs to be secured better.

True .4-.5" guiding will result in tight, round stars unless you're imaging at very long focal lengths (like 2m or more)

Edited by Oort Cloud, 20 March 2023 - 09:31 AM.

[radu.marinescu]

The only thing I see wrong here is poor tracking (elongation). Field has some bright stars and some dim ones. Global stretching will always cause the brighter ones to swell up, star reduction techniques in post-processing can help, but most of those techniques require the stars to be round for good effect.

So we're back to square one, the tracking needs to be fixed before anything else is really worth spending time or energy on.

Edit: I just noticed you said guiding was
.4-.5" all night. If that's the case, then either your pixel size/focal length in PHD2 is wrong, leading to incorrect RMS calculation, or your guidescope is moving relative to the main scope (differential flexure), and needs to be secured better.

True .4-.5" guiding will result in tight, round stars unless you're imaging at very long focal lengths (like 2m or more)

Oh boy, out of every possible cause I enumerated, differencial flexure in guiding is one I forgot about. I found various samples around the internet with before and after fixing the flexure and it seems you might be right about this one, it does look oddly familiar to my image. Still might be able to lower star size with some precise focusing and trying out the 57.5mm backfocus but the star shapes could be cause by flexure. Even if my giudescope is pretty well locked, I don't rule this out. I guess it's time to get an OAG and see how things improve.

[TelescopeGreg]

Oh boy, out of every possible cause I enumerated, differencial flexure in guiding is one I forgot about. I found various samples around the internet with before and after fixing the flexure and it seems you might be right about this one, it does look oddly familiar to my image. Still might be able to lower star size with some precise focusing and trying out the 57.5mm backfocus but the star shapes could be cause by flexure. Even if my giudescope is pretty well locked, I don't rule this out. I guess it's time to get an OAG and see how things improve.

You say the guide scope is "pretty well locked", but how is it actually mounted?  If it's in the "finder shoe", that's likely the source of the trouble.  Also, what guide scope and camera are you using?  Some have a flexible mounting ring that's more appropriate for an optical finder scope than a guide scope. 

 

Guide scopes need to be supported in two places, widely separated.  They do not need to be fiddled with every outing, so ease of adjustment is not an advantage.

[radu.marinescu]

You say the guide scope is "pretty well locked", but how is it actually mounted?  If it's in the "finder shoe", that's likely the source of the trouble.  Also, what guide scope and camera are you using?  Some have a flexible mounting ring that's more appropriate for an optical finder scope than a guide scope. 

 

Guide scopes need to be supported in two places, widely separated.  They do not need to be fiddled with every outing, so ease of adjustment is not an advantage.

It is not installed in the finder shoe but on the dovetail on the OTA's mounting rings. It is in the opposite side of the mount head, specifically beacause I didn't want it to cause balance issues in the finder shoe. It's a TS Optics Deluxe 60mm/240mm. The guide camera is an ASI120MM Mini. The scope is held in place by two rings, which are not only screwed in place on the dovetail, but they each have two counter-screws if I can call them as such that push agains the dovetail without getting inside any threads basically adding extra pressure and making sure they don't move no matter what. I am extremely confident there is absolutely 0 flexure in them. What I'm not confident about is the guide scope focuser. That's the only piece that might or could have flexure of some sort (as noted on other forums). Other than that the primary mirror of the Newt itslef.

BUT, I actually don't think flexure is the culprit after all. After further investigatin the issue, I noticed the extremely elongated stars are featured in one corner only, the opposite corner has almost round stars. This kind of indicates that guiding wasn't an issue and the problem is of optical nature. Could be collimation, could be insufficient backfocus, could be focuser tilt or even backfocus adapter tilt of some sort. I don't have a clear explanation but without someone that had my exact issue, I am at the mercy of experimenting and finding the issue by rulling out factor by factor.
 

[TelescopeGreg]

Ok, yeah.  That guide scope is mounted about as perfectly as one can do, so that's not it.

 

If the stars are elongated toward one corner, it might be that the camera is slightly tilted.  The ever-useful utility ASTAP has a tilt analysis function which can be used to display what's going on, and to help in any adjustments.  Just load up an image and go to Tools / Image Inspector, then Tilt F4.  Example from my scope, below.  Tilt isn't bad - the 4 corners are similar, but I've got some curvature as seen comparing the center and corner numbers.  Expected since the scope is a refractor and I don't have a flattener attached.

 

[radu.marinescu]

Ok, yeah.  That guide scope is mounted about as perfectly as one can do, so that's not it.

 

If the stars are elongated toward one corner, it might be that the camera is slightly tilted.  The ever-useful utility ASTAP has a tilt analysis function which can be used to display what's going on, and to help in any adjustments.  Just load up an image and go to Tools / Image Inspector, then Tilt F4.  Example from my scope, below.  Tilt isn't bad - the 4 corners are similar, but I've got some curvature as seen comparing the center and corner numbers.  Expected since the scope is a refractor and I don't have a flattener attached.

 

ASTAP Tilt analysis.jpg

Oh boy, that's the issue. Thanks a lot for the hint, I had no idea about this tool, although I had Astap working with NINA for platesolving. I attached the result. As one can see, there is a large tilt in one of the corners. However, I don't understand why I also have some curvature. Could it be because of the coma corrector? Now it remains for me to establish the tilted piece, wether it's the camera, some adapter in the image train not screwed properly or the focuser of the Newt itself.

At first I noticed the tilt is in the opposite side compared to the elongated stars from my picture at the beginning of this topic but for some reason DSS has flipped /mirrored my image. It's the same corner.

I think the next step is to check the image train, take some pictures on a rich star field and take several exposures at different camera rotations.

Edited by radu.marinescu, 21 March 2023 - 01:34 AM.

[TelescopeGreg]

Yes, ASTAP and DSS seem to have things flipped with respect to each other.  Not sure which is "right", if that term even applies. 

 

I've found ASTAP to be an amazing utility.  It's probably best known for its plate solving function, but it has a number of image analysis tools.  It also does a pretty good job at image stacking, sometimes better than DSS in my experience, though DSS does a better job at getting rid of satellite trails.  It even can stitch together image mosaics.

[Gert]

Hello radu,

When Astap refers to tilt it kind of unrealistically puts all effects into one basket that make elongated or enlarged stars. Yes, that includes tilt but also collimation and coma corrector issues. I would say in your case that you should investigate the collimation of the mirrors. I have found with my own 10inch f4 and TS corrector that only a small fraction of a mm in collimation will change the shape of stars in the corners.

Clear Skies,
Gert

[radu.marinescu]

Hello radu,

When Astap refers to tilt it kind of unrealistically puts all effects into one basket that make elongated or enlarged stars. Yes, that includes tilt but also collimation and coma corrector issues. I would say in your case that you should investigate the collimation of the mirrors. I have found with my own 10inch f4 and TS corrector that only a small fraction of a mm in collimation will change the shape of stars in the corners.

Clear Skies,
Gert

Yes, ASTAP and DSS seem to have things flipped with respect to each other.  Not sure which is "right", if that term even applies. 

 

I've found ASTAP to be an amazing utility.  It's probably best known for its plate solving function, but it has a number of image analysis tools.  It also does a pretty good job at image stacking, sometimes better than DSS in my experience, though DSS does a better job at getting rid of satellite trails.  It even can stitch together image mosaics.

First thing I will address is collimation and then coma. The collimation is very tricky thoug. I have a laser collimator as well as a chestshire cap that I use. The tricky part is that with the laser in, the collimation results differ. Like, if I unscrew the collimator + 1.25" adapter and take it out from the focuser, then put it back and screw it in, the laster will point to a different location on the primary mirror compared to last time. The error is not huge, like the pointer is about half a cm away from the mirror center, but I guess that value is enough to cause issues. And honestly I don't really know how to address this issue, it's clear to me that the Crayford focuser on my 10" isn't that good and allows for too much play in the 2" mount, which means the center point of any 2" or 1.25"-2" adapter put in it never actually coincides with the center point of the focal plane, which makes laser collimation very very inaccurate. I might keep the laser for secondary mirror alignement then use the chestshite cap only for visual alignment then a defocused centered star for confirmation.

If I manage to collimate it well, second thing I'll try is to see if I got tilt, starting with in my image train (without focuser). That will be relatively easy to be diagnosed indoors, I will shoot flat frames with a white image on my phone, placed on the image train (which contains ASI294, a T2 extender, a M48 extender and the coma corrector. If the flat frames will have decentered vignette, it will mean that I have sensor tilt. Second tilt test will be performed with the image train in the focuser, shooting flat frames but not with a flat panel on the telescope, with the phone inside the telescope, basically shooting them on the focuser inner tube. That will show if I have tilt because of the 2" focuser mount. Third test will be with a casual flat shot with the flat panel and some tshirts over the telescope, that will show if the focuser is decentered from the focal plane of the mirrors.

Last but not least, I will test different backfocus spacers to achieve the best star shape I can, but that will require getting the telescope out and do star tests and 14 days from now it seems there's going to be very cloudy . I'll keep this thread posted with what I find.
 

Edited by radu.marinescu, 22 March 2023 - 06:03 AM.