[SOLVED]Can't seem to plate solve

Hello everyone,
I recently changed my imaging rig - I now have a Celestron C11 EdgeHD that I will be using natively at F10 (2800mm FL). I also changed my imaging camera to QHY600 at 3.76px binned 1x1. I wast trying to plate solve an image that I already have from the set up, but I could not get it to plate solve and at my wits end trying to figure out what is wrong.

I tried all possible methods available in Prism to plate solve, but failed in all of them, including using Astrometry.net online (not local). The funny thing is, I was able to plate solve on nova.astrometry.net by directly uploading on their site (not via Prism) - so I know, that should have worked.

I believe I have changed all the parameters in all places - but do let me know where all I need to make all the changes, in case I would have missed making the update somewhere.

Lastly, is there a way to “reset” all the personalization that I would have from my previous set up? If so, that would be awesome.

Any help will be greatly appreciated.

Thanks,
Dhaval

Hi Dhaval, I had some difficulties also but finally now I plate solve now in few seconds.
I recommend to use UCAC4 catalog (download and install it) and GSC catalog … thiose are fast and worthy for a 20’x10’ field without any problem. Also be sure you have the CCD angle set properly so if put CCD in the position that is N-up and E-left (since you have a SC) you should plate solve with 180° angle. about.
When plat solve fails it means the telescope mount AR/DEC is far away from the correct position… so first of all goto a known star and manually ensure you are centered. take a picture to ensure you are centered. try to right-click and move telescope using vector on the picture to center it perfectly. Finally sinchronize telescope to image cemter position.
If you are still incapacitated take a picture of a good starry frame and to plate solve try an ANALISYS->BLIND PLATE SOLVE… if it plate solve be sure focal and angle is as you expect. I found the focal reducer to be very inaccurate so you need to have Astrometry.net to tell you the real focal lenght.

Hello Paolo - thanks for reaching out, however, it seems that I have tried almost everything that you’ve listed and I am not getting anywhere.

Specifically -

I tried using UCAC4 and GSC catalogs.

I did blind solve my image on astrometry.net, took the RA/DEC from there and put them in the solving properties to solve a single image, that did not work.

My focal length is accurate (within 1%, when the recommendation is to be within 10%). I don’t use a reducer.

The only thing that I haven’t tried is setting the CCD angle, which I can do that right now.

Having said that, I would have expected blind solve to solve without any information - like I said, I was able to solve that when I uploaded my image directly on astrometry.net, so I know it is not the image that is the issue, but obviously something that I am missing in Prism.

Thanks,
Dhaval

Seems you have some NORTH-SOUTH inverted or EST-WEST reverted… if angle and focal is right. That was also my problem… pls be sure prism is managing automatically the meridian flip if you don’t have a fork mount.

I’m really happy about how prism is smart in platesolving in less than a sec… for any decent picture even without calibration…you will be there soon and have same fun I do just try again proper settings.
I have uploaded a camera config (my is a sbig) to let you see below the setting for a german mount to let prism to manage automirror the pier side compared to the pictrure orientation.

So, I have tried putting in an angle based on astrometry.net readings, but that has not worked as well.

At this point, I would really like either Hamza or the developers reach out to me so that someone can look at it.

Thanks,
Dhaval

Astrometry.net is telling you if you have NORTH-up and EST-left?
If you have a german mount w/ SC optics check the Xmirror

Where do I do that? Is that while collecting data or while plate solving? Just so you know, this image wasn’t captured using Prism, but it should plate solve, right?

Thanks,
Dhaval

If all the image information are properly written in the FIT you should plate solve. In your case try to upload the image in Astrometry.net web site and let it to tell you all picture info. In particular the orientation (X and Y mirror)… after that open a photographic catalog like ALADIN, search for your AR/DEC and compare the field of view of your picture with the one shown in aladin. You easily see mirror and angle differences.

If you want to send me your pic I can try a plate solve in prism for you
paolo.fini67@gmail.com

I tried to plate solve a .FIT from an unknown source.
If you ask Prism to platesolve it it fails because it uses your settings and you may be far in focal and position and angle. What I did:

  1. right click on image and plate solve with astronometry.net. takes few but finally it will provide you exact data.
  2. at this point you can run Analisys->PlateSolve->SIngle image and prism pop-up with the astronomy.net info pre-loaded (focal is correct, angle is correct, FOV is correct…).
  3. run the prism platesolve and in few millisecond is giving you the fit.
  4. finally save those setting as plate solve default in prism so you don’t need to fill all the time.
  5. if you open ALADIN at the position you will se the image X-mirrored for example.

Attached a final result

So after much testing - there were a few folks involved - Paolo from here, there was one other person (John) from Facebook and myself - I am finally coming to the conclusion that Prism cannot handle large images for plate solving. This is a 120MB image - at 0.28"/px, this is also very deep. I am sure this is a shortcoming for Prism. Over the past few months, CMOS cameras have come a long way in terms of resolution that they offer and it seems that image acquisition software needs to update with that. Unfortunately, I don’t believe Prism is there just yet.

I will say this, if the image is resampled down, then there is a chance that it will plate solve - Paolo did test and prove that theory. At this point, if someone has to use Prism for image acquisition, then for plate solving, you almost have to bin 3x3 and then it may plate solve. I am guessing even AF will be the same.

Apart from that, my true concern is support - there is simply not a large user base to provide support for these types of challenges. I don’t believe Hamza alone can provide support to all North America users and the developers are no where to be seen.

CS!

well, a few things:

  1. Can you please send me the raw FIT? I would like to take a shot at this
  2. Support: A very subjective topic. I have the most amount of videos out of most vendors out there. Have I been slow to respond, well yes. Reacting to a new way of doing business has been an issue.

Now, as a solution, I will be building a wiki for fast searching, I believe this is the best way to centralize information. Of course, your input is welcome

Hi Hamza,
Based on my understanding, it seems that plate solving is still an issue. Have the developers had a chance to take a look at why it won’t plate solve? I may employ Prism for backyard imaging over the next month or so before taking it remote.

With regards to support, I do feel you’re in a chicken and an egg situation - a wiki may or may not solve a ton of challenges. I truly feel that the best way to do this is to create champions for Prism that will be happy to provide support to others. It truly has to take community proportions. I don’t believe a wiki will solve for user specific issues (and as you know, given the multitude of cameras/mounts/telescopes/etc., it is just not possible for a wiki to cover all aspects). It is like train the trainer first and then the trainer can help provide support.

Hope that helps.

DB

Hi Hamza,
Is there an update on the ability to plate solve with a C11 EdgeHD/QHY600 camera?

I am still not able to plate solve and use Prism.

Thanks,
Dhaval

Why do you need platesolve? For pointing accuracy or photometric results afterwards?
Is it not possible to do a cropped window platesolve after an slew it it is for pointing accuracy?

120MB is a very large file to platesolve i think…

as we have discussed in private, binning the image 2x2 or 3x3 will plate solve the image with zero issues.