Anyone Using ASTAP for Plate Solving in SGP?

I am using the latest stable build, and see that ASTAP is a choice for Plate Solving in the equipment profile. However, I still see the instructions to replace PlateSolve2 by renaming the ASTAP executable.

I have tried the PlateSolve2 replacement approach, and it does not see to work. I also switched my equipment profile/sequence to select ASTAP directly, and those solves failed as well.

Any suggestions would be very much appreciated … thanks,

Richard

ASTAP works great for me. Not a single failure.

You don’t need to rename anything or replace PS2. Make sure that ASTAP is installed in “C:\Program Files\astap”.

From the Help file:
Setting up ASTAP

ASTAP is a free Plate Solver by Han Kleijn. More information can be found at http://www.hnsky.org/astap.htm

If you’d like to use this solver during an SGPro sequence, you’ll need to download it and set it up.

Now, you just need to tell SGPro where you installed ASTAP. Click the “Tools” menu item, the “Options”. Then, at the top of this dialog, click the “External Apps” tab. Find where you installed ASTAP:

Thanks Joel,

I checked all that and it looks like I had it all right …

image

But when I load an image into SGP, right-click and choose “Plate Solve”

image

I can see the ASTP solver running (upper left corner),

image

but then is fails …

image

Any ideas?

The image above you have no ra or dec coordinates.PS2 and ASTAP needs those numbers. Just load an image that has those or search for those and you should be fine. Also the scale is listed at 1.000 . You need that number as well.

ASTAP is not a blind solver. You need the RA, DEC and angle hints. If you took the image with SGP the image should have those hints in the fits header. If not, you need to supply these hints in the camera setup tab on the control panel.

Bos & Joel,

Thank you! The image I was testing was shot from another app with my DSLR (.CR2 RAW format) and did not have the data. Tonight I will shoot a few images with SGP and retest the solver.

Richard

Dunno if you are still around, however, you maybe able to Blind Solve in SGP using either PS2 or ASTAP that image you have. To load an image in SGP it needs to be a 16bit fits file.

I downloaded a fits file like yours above that had not RA / DEC info scale, etc. With ASTAP selected as the solver, I used “Blind Solve.” This uploads the image to astrometry.net and while you are getting a cup of coffee (lol) a couple minutes later it will solve the image and you can select to “use these for reference image.”

Here are the screenshots. I downloaded NGC7000 for reference image.

ASTAP

ASTAP%20RA%20DEC

This is where you want to select "Use these results as reference…

Sucessful

Try and see. The majority of the time I can get the coordinates I need this way.

Thanks Bos … very cool … it worked!

With Bos’s help, I am now able to blind solve my initial image and automatically populate the coordinates of my target. But when I attempt to “slew and center” the target, I get errors:

SGP%20Centering%20Error

And even after several recovery runs, SGP cannot center the target:

SGP%20Centering%20Failures

I am sure I still have a lot to learn, but after several days of adjusting, I still have not gotten SGP to complete my first imaging run.

Any more ideas on where I should be looking?

Thanks again for all the help,

Richard

PS: I sure miss the function to look up a target in a database to begin the plate solve process. I am currently using APT and it handles this beautifully. But I also want to use SGP for all of its automation features. I have about 38 days left in my trial, and I want to get this working before I purchase.

What mount are you using?

What is your mount, camera and scope?
It looks to me like you plate solve scope center options might be too tight for the solve and sync to be successful. Your mount might have too much backlash or play in the gears and so it swings too far one way, then too far the other way. What are your setting here:
platesolve

Try to increase the attempts to 10 and the error in pixels to 150.

TargetOffset
I suggest you try “Target Offset” for the Sync Behavior setting.
That is the only setting that works for my Paramount mount. May be necessary for your mount.

I think what you’re looking for is the “Framing and Mosaic Wizard” under the tools menu.

Celestron AVX with Pole Master … polar alignment is pretty good. Once I get an object centered, it tracks pretty well.

Thanks Joel,

Yes, I was asking too much by using the defaults (never changed these) …

image

Tonight I will give it a go as you suggested … 10 and 150.

Thanks for the tip … here is what I had for my Celestron AVX mount:

image

I will try that setting tonight …

If you already had it set to “Target Offset”, then maybe you need to try one of the others.

Hey Joel,

I had seen that function, but thought it was only for creating mosaic sequences. Now I see that it can be used to create a sequence with a single frame. This should work great for the final step in my alignment workflow.

I use CPWI with my Celestron gear (C8 SCT on an AVX mount) to load and update the alignment model for my backyard observatory. The final step after the model is loaded is to sync to a calibration star.

Now I can use SGP and the Frame and Mosaic Wizard to slew to and center that star. Then I tell CPWI we are all centered and voila, we are completely aligned and ready for an imaging session.

Thanks for the tip.

Your problem seems to come from sync adding points to your pointing model, this will just adjust the model a little in the part of the sky your are pointing so centering doesn’t really get closer after several tries.
After a full sky model is built what should be done is turning of the ability to add points to the model.
After the model is finished a sync should skew the whole model, this will make it work perfectly with regular sync. Offset pretty much does the same thou, but no sync is done in the mount.

To be honest a pointing model on cheaper mounts isn’t of much use when using SGP with centering, SGP will find it’s target anyway and usually it will be just as fast.
The real gain for pointing models are for mounts that can do unguided imaging because it will adjust for flex, nonperfect polar alignment, cone error to reach the accuracy needed for unguided imaging.

1 Like