Pixinsight solving error, I'm hitting a wall here - Forums

Topic starter

May 13, 2024

So I want to stitch my first mosaic of Rho Ophiuchi. I have two panels which I have stacked and want to combine in pixinsight.
Now I need to plate solve the image first, to get a wcs file, to then use for further Stitching.

Now every time I want to plate solve the images, I get this error:

Every single time. I feel like I change every parameter that can be changed in the solver it always comes out to this.

What is this error? how can I make it work? or are there other viable solutions on how to stitch panoramas together outside of Pixinsight?
I don't even know what it is trying to tell me

dkamen

May 13, 2024

Shouldn't you check the box that says [ ] S(outh) ?

Reggie Jones

May 13, 2024

You may want to double check the coordinates you're using. I show Rho Ophiuchi having RA = 16h 27' 3.43" Dec = -23 deg 29' 21.2"
Per the previous comment, you may need to check the "S" box since the declination is negative.

I'm assuming your equipment focal length and camera pixel size is correct, but you should check those as well. If you're using a focal reducer, that will affect the number you use for focal length.

Well Written Helpful Concise

Michael Stauning

May 13, 2024

Could it have something to do with your filename? Ill notice that they are called:

xxx..fts - dobble dot.

Can you platesolve any of these:
https://stenhuskost-my.sharepoint.com/:f:/g/personal/mich580j_stenhus-kostskole_dk/EsWh2WtLXI1EmMWJ4zEADaABdVIiLJQYSgDWqIVXlJ-SSQ?e=C9tZ8B

Michael Stauning

May 13, 2024

If you can then there is something wrong with the info in your files.
If you cant, then there is something wrong with the settings in PixInsight.

Giovanni Fiume

May 13, 2024

Check fits header, there is some value not allowed ("Invalid argument type …").
You can always delete the fits header and regenerate it with plate solving (clone the file before).

Jens

Topic starter

May 13, 2024

Shouldn't you check the box that says [ ] S(outh) ?

I'm in the north though. that's why I unticked it. but honestly, i tried with and without, nothing changes.

Jens

Topic starter

May 13, 2024

So what I noticed now: I stacked With deep sky stacker. those files, out of some reason don't work.
I just tried astap, the file cam out horrible but it did work.
I am now stacking the files with wbpp to see if that works, it's just really hard to get into because it's not super straight forwared with stacking multiple days.
I got so used to DSS, it is such a shame that I need to change now because of this.

I'll try your file too and maybe remove the second dot, you're right I didn't notice that it had two dots at all. maybe that creates some issues as well

Marcelof

May 13, 2024 · Edited: May 13, 2024

A few questions:

- Is the GAIA DR3 catalog configured properly? And also at least in my case in ImageSolver the catalog under Model Parameters is set to automatic.

- How have you calibrated your files? PI doesn't like anything that comes preprocessed with other programs, it likes everything processed in house. It seems you are using .fit files, PI only accepts that format initially because it is the standard for any astrocamera, but for any use inside PI you must use .XISF.

Helpful

Clayton Ostler

May 13, 2024

I sometimes get solve errors based on the following
1. My file is still too noisy to solve (try running some noise reduction or abe on it)
2. I cropped the file and it lost all it coordinates (make sure coordinates, date, time and pixel size is correct in the solve)
3. Related to number 2, the pixel size thing has killed me a few times
4. Try renaming the file to something that is simple and the solve might not assume is coordinates, dates etc. (maybe something like file1.fits)
5. More advanced… I am assuming you have already stacked and calibrated, maybe try resolving a single sub, If that works you know your stacking or calibration is messing things up
6. Does the solve anywhere else? Will sharpcap or ASI Studio solve it? Not saying this fixes things but it lets you know if its your file or something in PI
7. Quadruple check your coordinates and date and time. Make sure you havent mixed RA and Dec, or that your arent showing positive when it should be negative, or that your have date in correct order.

Good luck hope this helps.

Helpful Supportive

Michael Rushalko

May 13, 2024

Hello, I tend to plate solve one panel at a time. I noticed that the date may be off as it shows 11/5/2024.

Clayton Ostler

May 13, 2024

Michael Rushalko:
Hello, I tend to plate solve one panel at a time. I noticed that the date may be off as it shows 11/5/2024.

I'd bet this was supposed to be 5/11 instead of 11/5

Well Written

Kevin Boucher

May 13, 2024

I use the PlateSolver script all the time to prep an image for SpectroPhotometricColorCalibration but I only use it for a single view at a time. You might want to try that but I did replicate your issue when you try to plate solve multiple images. See no reason why that shouldn't work as I could plate solve two images individually but when I tried them as a batch using list of files I get the same message you do. Think this is a bug in script.

Just the same I would make sure you have your parameters similar to my setup except of course for the image parameters. Make sure your GAIA Process parameters a correct and the DR3 files listed when you click on wrench icon. Yu can also increase the search radius as if your picking coordinates where your mosaic panels meet...especially at 300mm FL if that is what your using then you may want to increase this value to 2-3 degrees. Also, I found sometimes using "Automatic Catalog" works better than using local server. Always check distortion correction checkbox highlighted at bottom. My two cents...Good luck!

Helpful Engaging Supportive

Michael Stauning

May 13, 2024

11/5/2024 can be correct, my files from that date are like that (we use a good date version here ;) ) So day, mth, yr !

Brian Puhl

IOTD Staff

May 13, 2024

Just a tip. When I have weird issues like this and I know my coordinates are right, I set my focal length to a little bit longer, or change my pixel scale to something a little bit tighter. It usually works.

If I were you I'd solve one image at a time, however, particularly because it's a mosaic. Also, Rho is definitely south of celestial equator, you need 'S' to be checked. Validate the center position of each frame from your sequencer settings, or FITS header. Solve accordingly.

Helpful Concise

Jens

Topic starter

May 13, 2024

Ok so I finished stacking the data with Pixinsight and it worked without error and so did your files @Michael Stauning.
I removed the two dots and nothing changed. The DSS files still show the error.

@Marcelof I use fit files because the only option dss gives me are Tiff and Fit files. I normaly work with tif files but the image solver can't read them out of some reason, so I have to work with fit files.

So with what I have figured out with all of this is: I can't use DSS if I want to platesolve which is so strange but that seems to be the origin of the problem.
It sucks because it's my favorite but for mosaic work I have to get used to other stacking software as it seems like.

thank you guys for all of you time, I hope the rest of the Mosaic stitching will work flawlessly (as it always does in astrophotography ;)

Michael Stauning

May 13, 2024

DSS do funny things with the info in the fits header, when ever Ill get a DSS file Ill allways open the fits header and change/input all the correct values by hand to make it work with any programs.

Jens

Topic starter

May 13, 2024

Kevin Boucher:
I use the PlateSolver script all the time to prep an image for SpectroPhotometricColorCalibration but I only use it for a single view at a time. You might want to try that but I did replicate your issue when you try to plate solve multiple images. See no reason why that shouldn't work as I could plate solve two images individually but when I tried them as a batch using list of files I get the same message you do. Think this is a bug in script.

Just the same I would make sure you have your parameters similar to my setup except of course for the image parameters. Make sure your GAIA Process parameters a correct and the DR3 files listed when you click on wrench icon. Yu can also increase the search radius as if your picking coordinates where your mosaic panels meet...especially at 300mm FL if that is what your using then you may want to increase this value to 2-3 degrees. Also, I found sometimes using "Automatic Catalog" works better than using local server. Always check distortion correction checkbox highlighted at bottom. My two cents...Good luck!

So the "Active Window" feature does work better but I don't get the WCS file that I seemingly need for the next step. That is the whole reason I'm going through this, to get that platesolved wcs file I need for the Mosaic By Coodinates script. Thank you so much for showing this, and I'm glad I'm not the only one getting this error. On the Internet I can barely find anything about it.

Jens

Topic starter

May 13, 2024

Michael Stauning:
DSS do funny things with the info in the fits header, when ever Ill get a DSS file Ill allways open the fits header and change/input all the correct values by hand to make it work with any programs.

this might be a oversight or a very stupid question but what do you mean by fits header? where can I find it? Or is it just the filename that you mean?

Kevin Boucher

May 13, 2024

You can also plate solve online using astrometry.net which does an all Sky plate solve and I've never seen it fail. You don't really need to supply accurate coordinates but it helps.

Michael Stauning

May 13, 2024

Its under:
Process - Image - Fitsheader

If you open one of mine you will see thats its a light frame of 120sec and alot more

Brian Puhl

IOTD Staff

May 13, 2024

Jens:
Michael Stauning:
DSS do funny things with the info in the fits header, when ever Ill get a DSS file Ill allways open the fits header and change/input all the correct values by hand to make it work with any programs.

this might be a oversight or a very stupid question but what do you mean by fits header? where can I find it? Or is it just the filename that you mean?

in Pix, go to process, fits header. It contains all the data of the image, similar to an EXIF header.

Concise

Dale Penkala

May 13, 2024 · Edited: May 13, 2024

Michael Stauning:
Its under:
Process - Image - Fitsheader

If you open one of mine you will see thats its a light frame of 120sec and alot more

You can also get to it from File> View Fits Header in PI

Brent Curtis

May 13, 2024

Sounds like your not using the new WCS files or created but the files that maybe you have open… if your working e files, you can save and close those , then drop the new files the image solver created for you with that suffix

Kevin Boucher

May 13, 2024

I think you misunderstand…the WCS data is put into the Header of the Image file your plate solving and that is the data that the Mosaic by Coordinates script uses. There is not a separate file.

Helpful Concise