How to interpret field curvature? [ED72 + OVL]

[billyharris72]

Hi all:

I've just had a first go with my new scope (Skywatcher ED72 with OVL flattener) and early signs are good - think it's going to be a keeper. However, I'm struggling with what I assume is field curvature (never imaged with a frac before so this is new for me). The stars in and around the centre are great but at the edges of the field they are both out of focus and elongated (pointing towards and away from the centre of the image). I just had a couple of questions that I was hoping someone more experienced could weigh in on:

1) Does that sound like field curvature? (90% sure it is but worth asking)?

2) The effect is not equal in all corners (some are actually not too bad). Now I think the setup is square - could this be a distance issue, or do I need to work more on getting the setup square (and if so, any ideas on how / what helps)?

3) On the off chance someone has already dealt with this, does anyone out there know the correct sensor distance for this scope and flattener combo? Initial messing about suggests I might need more than 55mm, but any pointers would be really useful.

4) If not, is it possible to infer anything about the spacing (whether I need more or less) from looking at the curvature?

Sorry ... that's not a couple of questions after all.

Cheers,

Billy

[choochoo_baloo]

Would also like to know the answer to this Billy - watching with interest.

[alexbb]

I did not spend too much time on this as I waited for the "dedicated" (cough) flattener to appear.

However, not having the stars shapes the same towards all corners means that something's not quite square. I somehow doubt that the optics are tilted, I would bet that the flattener doesn't stay perfect in the 2" holder. Try to improve on that.

On the real issue: you need to add more space. How much? I don't know. I put a few delrin spacers on the M42 thread and, since it wasn't enough and I did not have an available 7.5mm extender, I put also some more spacers between the flattener's last element (not really on the lens, but on it's holder) and the adapter to M42. I'm maybe 5mm further away now and the stars shape improved. Perhaps a few more mm would be even better.

All in all, I don't have an exact answer, but I believe an M42 7.5mm extender would do. Try also to tighten the flattener without tilt.

[Louis D]

Yes, you probably need more spacing.  The shorter the focal length, the greater the field curvature in a refractor.  That flattener is optimized for a certain focal ratio range, but curvature mostly varies by focal length.  I use a TS - 2'' Field Flattener - Universal Field Flattening Lens visually on my 72ED and have to space it out to about 127mm to get about the best correction for most of my eyepieces that focus at the shoulder.  Their values suggest the following:

• Achromatic refractor f/5: 100mm to 105mm
• Refractor f/6.6: 103mm to 109mm
• Refractor f/7.5: 105mm to 110mm

For the very short 72ED, these values are no where near long enough.  As a guess, I'd say to try adding 8mm additional spacing to start with.  I'm basing this on your corrector doing the same work as mine in half the distance (55mm vs 110mm) and I needed to add about 17mm additional spacing.

[Stub Mandrel]

I'm interested in the outcome here, as I'm getting stretched stars in one corner of my images despite a field flattener.

[vlaiv]

It's a spacing issue, and a tilt issue.

Spacing issue is going to be trial and error approach. It depends on size of chip, focal length of scope, focal ratio of scope, so it is not always "prescribed" distance. I've looked at some flattner (with reduction) specs, and they can vary greatly in "optimum" distance based on telescope type (F/ratio and focal length). Also some flatteners are designed to certain corrected field - so one can expect correction over given field, but if sensor is larger, outer parts will suffer and not be fully corrected.

Procedure is simple - start at some distance and increase/decrease as long as you see improvement. Combine different length extension tubes and use distancing rings for fine adjustments (0.5-1mm range).

First step in solving tilt issue is to go with threaded connection. One can probably fix major issues by using high quality / self centering standard connections (baader click lock and alike), but I think that best solution is threaded connection. Adjusting focuser for being square with lens is another thing that can be done (if there is some play in focuser, or if it can be collimated). As a last resort - there is tilt adapter that can be used (but it adds optical path, and probably best for permanent setups).

[billyharris72]

Interesting thoughts there and plenty to work with. Sounds like I definitely have some tilt - wondering if the compression ring adapted that FLO sell would help here, as might be more rigid than the set screw adapter it comes with. As yet I don't hink there is a screw fit on the market (though let me know if I#m wrong!). As for the distance, I'm thinking perhaps a Baader adjustable tube is in order.

Billy.

[Louis D]

I recall reading about a C11 user who got so frustrated with Crayford focuser sag and slippage that he switched to a Clement focuser instead.  Don't discount your focuser's ability to counteract sag under heavy imager loads.

[RayD]

I agree with mosie, it is probably spacing.  I use the same configuration and you need nearer 59mm between the FF and sensor.  I use all screwed, so long as you pop the FF in with the scope facing downwards and make sure it is nice and tight, then it should work fine.  I've not noticed any draw tube sag as it doesn't need much projection to reach focus.

Start with the spacing and once right (take short subs pointing straight at the Zenith) see if you then have sag.

[billyharris72]

Thanks - pointing at the zenith seems a good idea.

I think I've identified the set-screw eyepiece holder as a definite source of a little flex in the system, so that seems an easy one to fix. Focuser movement would be a lot worse, but I tend to agree with Ray - the tube is only racked out maybe 10mm or so and the camera and mini filter wheel is not that heavy (fingers crossed...)

Something that concerns me - is it common to have this elongation predominantly in one corner of the image? I would have thought if the camera is tilted it would affact the whole image plane, at the angle of the tilt, but the elongation I'm seeing tend to be concentrated in one corner and looks radial, with the elongation pointing towards the centre of the image. I've attached a test image (see how the elongation is much worse top tight corner)  and an output from CCD inspector in case anyone can make some sense of it.

Thanks,

Billy.

[billyharris72]

Good news - unscrewed the nosepiece on the OVl flattener to reveal an M48 female thread. And with this on FLO ...

https://www.firstlightoptics.com/adapters/lynx-astro-m48-adapter-for-sky-watcher-72ed.html

Goodbye tilt (hopefully)!

[RayD]

Yes that does point towards a very slight tilt, but the spacing looks to be out as there is a radial pattern across the whole image, just a little more noticeable at the top.

I did a short intro video here showing the option of the screw fit.  This is what I use and it seems to work very well, and definitely an advisably option in my view for imaging.

Edited September 13, 2018 by RayD

[billyharris72]

Hmmm.. Well, I've tried imaging again, having done the following:

1) Replaced the push fit eyepiece holder with a screw fit so the whole imaging train is threaded.

2) Tried varying the spacing (currently up to 59mm and increasing - or will when the weather clears).

Results so far are not great. CCD inspector shows the curvature lowest at 58mm and 61.5mm, and higher in the intermediate values (is this normal)?

The elongated stars are still there with all setups - with some spacings it seems to be in all corners, in others only the right hand side of the sensor. The whole setup is threaded and I can't detect any movement in the focuser (if I hold the back of the camera and try to move it up and down the whole thing feels rigid - could move it slighly by applying force I'm sure but there is absolutely no play or wobble that I can detect). Pointing the setup at the zenith for test shots makes no difference at all - elongation remains and is about the same.

Again, I've attached a couple of images for reference - I think there might be a slight improvement, but the edges (especially on the right) are still very poor.

Struggling to know what to try next, so thoughts are much appreciated. Is it worth continuing to adjust the distance, or do the elongated stars mean I'm wasting my time?

Billy.

 

[RayD]

To me that looks a little like the elongation is bottom left to to right across the whole field, so possibly some tracking issues?  What length subs are these?

I have used CCDI many times, and if you have a read up of the author's recommendations it is quite specific on star field, length and number of subs and orientation.  I think to test flatness of the field you need to totally eliminate any other possibility, so very short subs, and a number of them (20 or so) in a very even but dense star field.

[Stub Mandrel]

Sorry to gatecrash, can anyone help me interpret this - it looks bad!

[billyharris72]

1 hour ago, RayD said:

To me that looks a little like the elongation is bottom left to to right across the whole field, so possibly some tracking issues?  What length subs are these?

I have used CCDI many times, and if you have a read up of the author's recommendations it is quite specific on star field, length and number of subs and orientation.  I think to test flatness of the field you need to totally eliminate any other possibility, so very short subs, and a number of them (20 or so) in a very even but dense star field. 

Interesting and useful to know - I did some variation between individual subs despite nothing changing in the setup so makes sense to take a larger sample. Sub length for these was only 5 seconds, and I'd checked the alignement. With the target area being in Cas. I'd be surprised (though not astonished) it was tracking. Worth double checking I suppose, but I'd expect to see the same elongation across the field. I'm a bit stumped but think I need to check the performance with slightly greater backfocus. I'll also try a few shots without the flattener or any spacers - that should at least help me narrow down where the tilt is in the system.

Billy.

[billyharris72]

12 minutes ago, Stub Mandrel said:

Sorry to gatecrash, can anyone help me interpret this - it looks bad!

Ow! Was that shot on tghe 130 P-DS (I though Newts had a pretty flat field generally)? That far out would have me wondering about focus and collimation (FWHM looks huge). What did the actual sub come out like?

[Stub Mandrel]

5 minutes ago, billyharris72 said:

Ow! Was that shot on tghe 130 P-DS (I though Newts had a pretty flat field generally)? That far out would have me wondering about focus and collimation (FWHM looks huge). What did the actual sub come out like?

No, the ED66.

The field flattener seems to a have a wide full diameter section before the taper section, and the fixing screws seem to line up with the 'step'. I suspect this is tilting the camera and field flattener.

This is one from the 130P-DS. I assume this isn't perfect but is far better than the other one, and suggests my camera itself (same one for both images) has no major issues?:

[billyharris72]

Could well be that the flattener is causing the tilt. I'd try a few shots without it - field curvature might be horrible but at least it'll narrow down the source of the problem.

[Stub Mandrel]

1 hour ago, billyharris72 said:

Could well be that the flattener is causing the tilt. I'd try a few shots without it - field curvature might be horrible but at least it'll narrow down the source of the problem.

I had a close look in daylight, and tightening one screw forces the flattener out of alignment by about a millimetre. I will sort the new adaptor this evening and try with that first.

Another option is an adaptor screwed on both sides so the flattener screws direct to the focus tube, guaranteeing alignment.

Edited September 19, 2018 by Stub Mandrel

[ollypenrice]

Scroll down this thread to Sara's link (Swag72) for the 'too near/too close diagram: 

 

This should be helpful.

Olly

[Skipper Billy]

Is this any help ?

[billyharris72]

That's a result! Cheers Olly and Skipper Billy!

[Stub Mandrel]

On 19/09/2018 at 18:52, Skipper Billy said:

Is this any help ? 

Helps me.

For my second session I took care to make sure I eliminated tilt as far as possible.

The effect is now less marked, but is like your image 1 - too close. Better than too far!

I will 3D print a 1mm thick spacer for between the flattener and the T-mount adaptor and try again. Let's hope it doesn't take too many tries.

[Skipper Billy]

1 minute ago, Stub Mandrel said:

I will 3D print a 1mm thick spacer for between the flattener and the T-mount adaptor and try again. Let's hope it doesn't take too many tries.

Hi Neil,

What ID are the spacers you need - I have loads of Delrin spacers from 0.5mm, 0.75mm and 1mm - PM me your address if you would like them. (FOC).