Goto Plate Solve issues

[Forum_2015]

I am currently demoing AstroArt 5 and tonight was my first chance to test the software out in the field.
My main requirement for AA is Plate Solving. Having tested Plate Solving on various images in demo mode I am familiar with the method of returning accurate resolves.
However after performing the same steps with my polar aligned mount I found that I did not get the expected results.

I wanted to test AA with a RAW image I had taken on of the Double Cluster a few months ago. I used two methods both of which failed to give me the results I was looking for.
I began by inputting my Focal length and CCD image size. Next I loaded an image of the Double Cluster and used Find Co-Ordinates, this resolved correctly. Then after connecting the mount I went to Telescope control and tab Goto Plate Solve. Left all settings on default. Clicked Center the mount moved to the co-ordinates. I took an image with Back Yard EOS and the outputted image was not aligned with the plate solved image. The double cluster was on the screen but not at the same position as the plate solved images. Why is this, what am I missing or doing wrong?

I tried various other options like Synch (resynch to target) but nothing came close.

Can anyone please help?

Thanks

[Forum_2015]

Hello, I don't understand well the procedure, but it seems not correct.
The command "Center - Goto Plate solve" is used to center an object in the field of view of the camera, using an image which was taken a few seconds before in Astroart (where the subject is not in the center), after a telescope GOTO launched with Astroart as well.

[Forum_2015]

Thanks .

Okay that makes sense from the Center goto.
So is it possible to get AA to plate solve a previous image taken weeks before and get AA to goto to this exact same position?
The idea is that I can return to a target over a sequence of nights.
Hope this makes sense.

Thanks

[Forum_2015]

If the target has a name, it's simpler to write it in the GOTO page of the Telescope Window, and/or save it into a List. This works also on generic coordinates (e.g. 12 37 34 +45 58).
For an unknown field, there is not a direct command, but it could be done these ways (after the image has been calibrated).
A) Right click on the image, click Star Atlas, then click the Telescope icon.
B) This one-row script: Telescope.Goto(Image.RA,Image.DEC)

[Forum_2015]

Thanks again I will have a look into this further when I get clear skies.

[Forum_2015]

Droogie. let me know how you get on, im in the same boat..

[Forum_2015]

Hi Astrocot2

Sorry I did not get a notification that you had posted to this thread (or I missed it)

I will certainly let you know how it goes. There is a small window tonight but it very much depends on the weather. It has been either raining or cloudy for well over a month in the UK, really depressing.
I realise tonight is a full moon but it will be an ideal opportunity to test out what Facavi suggested.

So I have tested the following with the simulators and it works as required just need to test in the Field!

Here is what I plan to do.

The assumption here is you have the mount connected accordingly and it is talking to AA correctly.

To begin with Plate solve the image I want my mount to slew to. So I have loaded an image I took a 2 months ago and Plate solved. Then open the 'Statistics' view copy the numbers on the 'Center' row e.g. 00 41 55.6 +41 21 24
Create a regular text file with the exact co-ordinates listed within it. e. "M31" 00 41 57.7 +41 21 18 (other syntax's are possible but this is my current preferred method)
*Note*. The above details must be acquired from an AA plate solve and not the standard International RA & DEC co-ordinates for M31. They have to be your individual plate solved co-ordinates or else you mount will take you to the default location which defeats the whole point of this.

Save the text file.

Open AA and then open the Telescope Control panel then the 'GOTO' tab. Click on the Yellow folder icon and open your custom text file. You will see the Name of the target, in this example M31, select and click the Goto Target icon (double arrow icon)
Your mount should now slew to those exact co-ordinates ready for imaging.
I will then take an image with the same exposure time and settings and then overlay the result with one of my original RAW images using Backyard EOS masking tool (Photoshop will also do this). This tool allows you to fade between two images to see if they are aligned.
Obviously I do not expected them to be absolutely identical but they should be reasonably aligned. Deep Sky Stacker (or other stacking software) should be able to take care of any discrepancies, in fact it should be the same as performing dithering on the images.

As I mentioned I have tested this works with the Simulators and the Mount Simulator does indeed slew exactly to the RA & DEC I have asked it to go to.

So all of this needs to tested in the field. I will post my results...

[Forum_2015]

Okay so I tried last night. Polar Aligned my mount 3 times and felt I had good alignment.
Started AA5 and slewed to the Goto co-ordinates I had in my text file for the Double Cluster.
Took an image which was only a 30 secs exposure but enough to plate solve from.
The results are below:
Plate Solve location that the mount was asked to slew to: 02 20 01.09 +57 13 21 (NGC 869)
Actual Location it slewed to after performing a Plate solve on it. 02 17 50.4 +57 09 38
So I believe I was out by 2.5 arc minutes?
After blending the two images with Backyard EOS this shows as being quite a lot of but I can see that the actual angle is fairly correct but the distance between them is where the main issue is. I may post a composited image to give an idea but no time at the moment. I think stacking on this image location would be very poor and in some circumstances I could potentially see swathes of Nebula would be missing (not in this case of course as it was the Double cluster)
I realise it probably all comes down to my mounts alignment however I am still a bit disappointed.
Stupidly I should have also taken a simple image with just slewing my mount via its own goto the Double cluster but I did not do this. This would have been useful to see what effect (if any) that the AA5 Plate solve co-ordinates have. I also tested on M31 and the same results were yielded.
I am not sure how much I should have expected from this test.
What and how do people get very precise results? Am I missing a second / last step. In that I need to plate solve a second time? Or do I have simply nudge the mount to the correct location? I want to avoid this as the directions buttons do not go in the direction you would expect due to the GEM mount and of course it defeats the purpose of plate solving.
Any suggestions would be great.

[Forum_2015]

Hello, I still don't understand well the problem. Plate solve and centering are two different tasks, it is possible that plate solve is now currently tuned (but you should write which paramaters you are using), but centering (GOTO) fails for some other reasons.
By the way 2.5 RA arcminutes are indeed 30 arcminuts, so 0.5 degrees, which can be a lot, depending your FOV.

[Forum_2015]

Hi Facavi. Thank you for your help and patience!

Ultimately what I trying to achieve is to be able to return to a target over several nights. As I do not have an Observatory or a Pier I have to setup from scratch each night. Therefore the chances of aligning my camera and scope exactly as before are very remote.

Since I posted earlier I decided to follow an idea in this link:

http://stargazerslounge.com/topic/19919 ... e-session/

This covers the same idea.

So I decided to follow the steps outlined near the end of the post with a stack of the small amount of images I acquired last night to see what happened.
I have to say the results look okay. As long as you make one of the light frames as a reference frame then any frames which are not as well aligned i.e. the Next nights frames then it will only cover the reference frame.
I am using Deep Sky Stacker and the output meant I get a line across the frame. This must be where the two alignments meet.
I will attach the image to this post so hopefully you will see.

So it has worked but I would like it to work much better.

Facavi. Can you tell me how I would improve the Centering after a Plate solve. I may simply be missing a step. I should have noted what RA & DEC AA said on the Goto parameters but I forgot. The cloud was crazy last night so only had a hour to do anything.

I have stretched the image to show the two diagonal line. So one RAW image was taken 2 months ago and another was from last night. Due to the conditions last night I could only take a 30 secs exposure at a different ISO. The original image was 300 secs @ ISO

[Forum_2015]

Hi Droogie ?, I just want to encourage you to spend the time to perfect this technique. I do it using scripts and it works great and is a huge time saver. I set up on my patio so I probably don't break my setup down as far as you need to. It looks like you need to find a way to reposition your scope and camera more accurately. Maybe try putting some blue painters tape on the scope and camera and draw a line across the two pieces to realign to. Good luck!
Iver

[Forum_2015]

Hi Iver.

Thanks for the tip I will give that a try.
Also please can you paste a copy of the script that you use in this thread?
I am only using a custom text file which I use for plate solving. Here is an extract from the text file that I load into the Telescope Control, GOTO tab. Obviously these RA & DEC co-ordinates have be gathered from previous images I have taken.

"NGC869-DoubleCluster" 02 20 01.9 +57 13 21
"M16-Eagle-Nebula" 18 18 53.0 -13 30 25
"M31" 00 41 57.7 +41 21 18
"IC1805-Heart-Neb" 02 31 11.0 +61 34 13

Thanks