41 replies to this topic

[AstroNikko]

Thanks, Björn. I'm going to try again with the working distance at the recommended 85mm from last lens to sensor, and plate solve to confirm. My target image scale will be 0.628 arcsec/px, which should put the focal ratio at f/6.3 with an effective focal length of 1280mm.

 

The Meade model 2080 SCT I'm using does not have ACF optics. It's an older scope, deforked from an LX3 mount. Should still be good to use with the Meade f/6.3 reducer/flattener. What I'm unsure about is whether I can expect to have round stars from edge to edge of the frame, or if I should expect to see crescent shaped stars with increasing intensity the further off axis they are. Even with a flat field. I'm not yet familiar enough with this kit to know either way.

 

I may need to use something like CCD Inspector to evaluate the field curvature.

 

Thanks again!

[barnold84]

Thanks, Björn. I'm going to try again with the working distance at the recommended 85mm from last lens to sensor, and plate solve to confirm. My target image scale will be 0.628 arcsec/px, which should put the focal ratio at f/6.3 with an effective focal length of 1280mm.

 

The Meade model 2080 SCT I'm using does not have ACF optics. It's an older scope, deforked from an LX3 mount. Should still be good to use with the Meade f/6.3 reducer/flattener. What I'm unsure about is whether I can expect to have round stars from edge to edge of the frame, or if I should expect to see crescent shaped stars with increasing intensity the further off axis they are. Even with a flat field. I'm not yet familiar enough with this kit to know either way.

 

I may need to use something like CCD Inspector to evaluate the field curvature.

 

Thanks again!

Depending on your goals, I would recommend not making a science out of it. For me, the most important things are: reduction of focal length to get a faster optics with reduced resolution (I have about 0.3arcsecs/pixel which is way too aggressive given normal seeing) and a (satisfyingly) sharp image to the edges. If I can achieve that, I don't care if its f/6.3, 6.4 or 6.1.

 

If you want to see the behavior of the reducer do extreme tunings, i.e.: take an image with working distance of say 30mm, 85 and 105mm. Compare the images, do plate solving and then you should see where the trend goes. If you do plate solving for the three setups, you could even compute the optimal distance. Check the image quality in the edges and you are probably get a feeling what a good setup will be for you.

 

Björn

[AstroNikko]

Pretty much the same for me. My main goal is a flat field with uniform star shapes throughout the field. Don't care as much about the reduction factor.

 

After seeing the results at 87.2mm working distance, I tried out a few different working distances up to 104.2mm. Wasn't particularly happy with any of them. Stars did appear tighter at 104.2mm than at 87.2mm, but that makes sense with a wider field of view. Stars would be smaller. It's difficult to evaluate the differences visually, though. Although I should be able to evaluate field curvature visually using a Bahtinov mask. Looking over some of the test images I took, the stars in the center of the field are in focus, while those at the edges of the frame are out of focus.

[barnold84]

Hi Nikko,

 

I think I might have read somewhere that software like PixInsight could determine the field curvature but don't nail me down on that.

 

Just an idea on what you could try to do, to visually inspect the image quality is:

a. Overlay the images at different distances that you have after the images are aligned (rotation, translation and scale). My image editing software (Affinity Photo) can align images. So you would see if the stars match. If the curvature is constant, they would match.

b. At a given working distance, capture an image. Capture one or more images, where you are centering the stars in the edges. Then align and overlay again. If there is no field curvature, the overlay should be a nice mosaic. If there is curvature, then you'll see the difference.

 

What's the size of the imaging chip you are using?

You wrote it in the initial post. It's APS-C size. One thing I am not sure is if the field correction is/can be perfect for very large sensors. Unfortunately, I don't have any reference for that.

Edited by barnold84, 23 February 2021 - 11:51 PM.

[AntMan1]

Just a quick note, you will never get a flat field with the meade focal reducer. It does more harm than good. It distorts the stars bad. It says its a reducer / flattener but its not. It only a reducer.  Try the candle test, the image will converge with just using the reducer. If there was a flattener in there you would need the lense of the telescope for it to converge. just my 2C throw it out. You will never get round stars with it.

 

Now this is a much better sct reducer

 

https://starizona.co...-coma-corrector

Edited by AntMan1, 24 February 2021 - 12:51 AM.

[barnold84]

Just a quick note, you will never get a flat field with the meade focal reducer. It does more harm than good. It distorts the stars bad. It says its a reducer / flattener but its not. It only a reducer.  Try the candle test, the image will converge with just using the reducer. If there was a flattener in there you would need the lense of the telescope for it to converge. just my 2C throw it out. You will never get round stars with it.

 

Now this is a much better sct reducer

 

https://starizona.co...-coma-corrector

I'd like to object on your statement that the Meade reducer does more harm than good. You're writing in your initial post that you use a de-forked LX90, given that this is the same that you have in your signature, i.e. a LX90 ACF, you shouldn't be surprised that the reducer does indeed more harm than good as the Meade reducer is not suitable for ACF optics, as Meade states itself.

 

The Meade reducer does flatten the field curvature of a regular Schmidt-Cassegrain optics, which has two effects: 1. a flat field leads to consistent distances on the image (i.e. if two objects are n pixels apart when in the center of FOV, hey will also be n pixels apart when seen at the edge. Caveat: if the field would be perfectly flat, which it won't be but much better than without flattener) and 2. objects in the edges will be sharper than without flattener as the focal plan becomes a focal plane and not a focal "sphere".

 

Coma is another type of optical aberration that is corrected with the ACF. For many people, a corrected coma is more important than a fully flat field. 

[AstroNikko]

Thanks for the feedback, guys. If the results at 85mm aren’t satisfactory, I’m going to try the Bahtinov test for field curvature at those working distances suggested (30mm, 85mm, 105mm) and compare the alignment of the diffraction spikes. If the pattern is the same regardless of working distance, then it’s probably a good indication that this particular reducer lens isn’t actually flattening the field.

I’m not sure what the corrected field size is supposed to be for the Meade f/6.3 FR/FF. The Starizona SCT corrector lists a nominal image circle of 27mm. The Fuji X-T100 has a 28.26mm sensor diagonal.

I’ll keep testing the Mead f/6.3 FR/FF. I may end up returning it if it looks like a flat field isn’t possible, but it seems coma correction may not be a feature of this lens. I don’t fully understand the relationship between field flatness and coma, though. If I understand correctly, I think it is possible to have a flat field and yet still have coma, albeit with tighter airy rings.

[barnold84]

Thanks for the feedback, guys. If the results at 85mm aren’t satisfactory, I’m going to try the Bahtinov test for field curvature at those working distances suggested (30mm, 85mm, 105mm) and compare the alignment of the diffraction spikes. If the pattern is the same regardless of working distance, then it’s probably a good indication that this particular reducer lens isn’t actually flattening the field.

I’m not sure what the corrected field size is supposed to be for the Meade f/6.3 FR/FF. The Starizona SCT corrector lists a nominal image circle of 27mm. The Fuji X-T100 has a 28.26mm sensor diagonal.

I’ll keep testing the Mead f/6.3 FR/FF. I may end up returning it if it looks like a flat field isn’t possible, but it seems coma correction may not be a feature of this lens. I don’t fully understand the relationship between field flatness and coma, though. If I understand correctly, I think it is possible to have a flat field and yet still have coma, albeit with tighter airy rings.

Hi Nikko,

 

You might try to contact Meade support. They should be able to help you. Was in touch with them some days ago and surprisingly, they responded within a few hours.

 

Somebody on CN put the manual: https://www.cloudyni...r/#entry8366691

You can see that one combination is T-Adapter, T-Ring, and Camera. This will lead to 105mm back focus/working distance. That's consistent what I've heard and read from other people about this FR/FF.

 

Let me try to roughly describe field curvature and coma:

Fiel curvature means that the focus of off-axis objects, like stars in the edge, are not focused on the plane of the chip but a bit before it, while the center objects are focused on the plane of the chip. So it's not a focal plane but rather a focal sphere. Hence, stars in the edges will look elongated/egg shaped and not perfectly focused. A lens flattens this sphere so that it becomes a plane.

 

Coma is different and has nothing to do with Airy rings. Airy rings a diffraction effect due to the aperture of the optical system. Coma makes stars at the edge of the FOV look like tiny comets. It's a bit more difficult to explain. Here's a link to an example image: https://images.app.g...5B8AKSDUEU4tkdA

The left image shows coma, the right is corrected.

 

Björn

[AstroNikko]

Thanks, Björn. I’ll reach out to Meade Support.

My understanding of coma is that it is the result of airy rings being offset away from the central axis, being further offset the further off axis the star is positioned. That overlapping of rings along one side of the airy disk with offset is what gives it that coma shape.

I’ve noticed that with my HRCC, if I don’t have the spacing dialed in just right, the field curvature will offset the airy rings away from the central access, but still be in focus across the field. Instead of the typical coma shape, they look like elongated stars radiating from the center of the field, or conversely spread out into crescent shapes depending on too much or too little spacing.

The optics behind this phenomena I still haven’t wrapped my head around yet.

[barnold84]

Thanks, Björn. I’ll reach out to Meade Support.

My understanding of coma is that it is the result of airy rings being offset away from the central axis, being further offset the further off axis the star is positioned. That overlapping of rings along one side of the airy disk with offset is what gives it that coma shape.

I’ve noticed that with my HRCC, if I don’t have the spacing dialed in just right, the field curvature will offset the airy rings away from the central access, but still be in focus across the field. Instead of the typical coma shape, they look like elongated stars radiating from the center of the field, or conversely spread out into crescent shapes depending on too much or too little spacing.

The optics behind this phenomena I still haven’t wrapped my head around yet.

You don‘t need the Airy pattern to have coma. These are two different things. Maybe this Wikipedia article might help you https://en.wikipedia...s)?wprov=sfti1 

 

One more thing: did you check that your SCT is nicely collimated? Before investing time into coma and curvature, the collimation should be ok.

[AstroNikko]

Thanks. I’ll check the collimation again. I think it may be slightly out of collimation. It’s pretty close, but that may not be good enough.

[AstroNikko]

Aaaaah, ok. After reading that Wikipedia article, I think I have a better understanding. The rings used to depict coma aren’t airy disk rings, but are representative of light reflecting off of the parabolic primary mirror at different distances from center to edge. The higher the angle of incidence with the surface of the primary mirror, the more coma it exhibits. Still not sure I understand it correctly. It’s a lot to wrap my head around.

Thanks for correcting me on this!

Edited by wcoastsands, 24 February 2021 - 02:44 PM.

[AstroNikko]

Was going through my test images again and found something interesting. One of the last images I took was at the shortest working distance possible with the T-ring threaded onto the WO SCT-M48 adapter. This provided a working distance of 45.7mm. Which roughly completes the data set I was planning to capture.

I was able to measure the off-axis aberration for each image using the CCD Inspector tool in ASTAP. The interesting part is that the off-axis aberration seems to decrease linearly with respect to working distance, but does not reach zero within a reasonable working distance.

Based on these results, it doesn't look like this lens is capable of producing a flat field. Will be reaching out to Meade for further explanation/clarification.





Astrometry.net Results:

• DSCF7403 (45.7mm working dist.)
• DSCF7396 (87.2mm working dist.)
• DSCF7404 (104.2mm working dist.)

Edited by wcoastsands, 28 February 2021 - 10:36 PM.

[han.k]

Any optical system no matter how good it is will still show an off-axis aberration. So star in the outer regions of the image will always show some kind of ovality.

 

I have an 20 year old SCT Celestron reducer, same make as Meade, but the optical quality is poor. I moved to a APO astrograph where all corrections are build-in. The images of this telescope show pinpoint sharp stars anywhere in the image.

 

 

Looking to you jpeg, your image contains a lot of hot pixels. They are abundant compared to the stars and spoil the measurement. The stars seems to have a HFD of 11 to 15. So to filter out hot pixels, I would increase the minimum HFD to something like 5. That will give you a more realistic measurement.

 

 

Han

 

 

 

 

[AstroNikko]

Looking to you jpeg, your image contains a lot of hot pixels. They are abundant compared to the stars and spoil the measurement. The stars seems to have a HFD of 11 to 15. So to filter out hot pixels, I would increase the minimum HFD to something like 5. That will give you a more realistic measurement.

Thank you, Han! I really appreciate the feedback.

 

Just updated to the latest version of ASTAP, changed the setting as you suggested, and ran the CCD Inspector tool again. Results are pretty close to what you suspected. Thank you for the guidance.

 

Updated values are as follows:

 

I don't think there's sufficient data here to draw any conclusions. Need to keep testing. Also need to keep in mind that field curvature is likely not the only factor contributing to the off-axis aberration value. The best way to isolate and test for field curvature might be with a Bahtinov mask.

 

Thanks again,

 

Nikko

Edited by wcoastsands, 01 March 2021 - 04:00 AM.

[han.k]

In the photometry tab of the stack menu, you can do something more. If you take for each working distance 7 or 10 images at different focus positions before, at focus and after the best focus position, feed the image to this photmetry tab. The program will then curve fit the HFD values for five areas. One area is in the center and the four corners at tell you at which focuser position that area is in focus. Knowing the best focus of each of the five areas you can express the curvature in delta focus positions.

 

This will only work if your acquisition program records the focuser position.

 

You could also do it manually. Try to focus at the center and try to focus at one of the corners. The curvature is the difference between the two focus positions. But you need an accurate focuser readout and move only in one direction to avoid backslash.

 

Not easy. It is easier to compare the images you already took and select then one with the best HFD value. The routine will take all stars and find the median value. The lower the better .

 

Han

[AstroNikko]

Thank you! Great suggestion. Unfortunately I don't have an automatic/motorized focuser. Not even able to use an acquisition program with this particular camera.

 

If I understand the concept correctly though, it may also be possible to evaluate curvature based on difference in image scale between focus points. I suspect image scale would be different between focus points for center of field and corners with field curvature, considering the focal length of the SCT would be different between the two focal points. If this does work, I'm not entirely sure where to go from there, though. Compare the deltas, and the working distance with the lowest delta would then be the optimal working distance?

 

I may be overcomplicating this. Best HFD value sounds like it may be the easiest, most viable solution.

 

Thanks,

 

Nikko