Plate solving with PHD2...

[rsf...@rsfotografia.com]

Hi,

As a spectrographer I would like to ask if it would be possible to Plate Solve using the image of PHD2.

Why? 

When doing spectroscopy we have the main camera behind a slit and what we see there is an image with a line = the spectrum of a star or whatever we want to do spectrogrphy on it. So we can not plate solve with the main camera.

Now for getting a star into the slit and for this we need to first plate solve the image of the guide camera in another software like maybe SharpCap and then switch the camera to PHD2.

Is it feasible to integrate plate solving into PHD2 or is it already there and I have not found it?

Thanks for your comments

Rainer

[Bruce Waddington]

Plate solving isn't part of PHD2 for a number of reasons.  First, the guide camera frames often have a pretty small field of view, and more importantly, they reference a different part of the sky from your main camera.  If you're using an off-axis-guider, the pick-off prism will be outside the main field of view so the guide camera image will normally be offset by at least the radius of the main camera image.  If you're using a separate guide scope, it's usually not practical or even necessary to get the two telescopes precisely co-axial.  For these reasons, I don't think a plate-solved image through the guider is very useful and isn't appropriate for synching the mount.  If you feel the need to pursue this, you can acquire guide camera image frames through the PHD2 event server and then do the plate-solving yourself.  A different question though is whether you need to do any of this.  Plate-solvers aren't necessarily thrown off by central obstructions or non-stellar bright areas - have you tried plate-solving an image through the main camera?  I'm attaching an example of a field with a large black rectangle in the center of the frame and it can be plate-solved very quickly.

Beyond all that, adding plate-solving to PHD2 would involve a lot of configuration, installation, and support headaches that we don't really want to deal with given the limited value of the end-result.  We have a substantial number of PHD2 users who do spectroscopy so I assume they've found workable solutions without the need for any new features in PHD2.

Regards,

Bruce

[NHSA]

Thanks Bruce.  I understand the amount of work.

Most spectrogrpahs do have the guider integrated imaging the image of the mirror with the slit and that is why I asked. I ahve no problem solving on the image delivered by the guide cam so far.

I at least have to first Plate Solve using let me say SharpCap and Plate Solve on the image delivered by my ASI 174MM mounted on the guider of the Spectrograph. After solving I go back to PHD2, center the star in the slit and start guiding be it on the star in the slit or any other visible star around the main star.

I have not yet found a video or paper explaining how they remotely center the star in the slit?

Thanks again and was just a curiosity question.

Rainer

[NHSA]

quote: "... you can acquire guide camera image frames through the PHD2 event server and then do the plate-solving yourself. "

How does this work and so far I understan I can share the PHD2 image with another program at the same time without disconnecting the camera from PHD2?

Thanks Rainer

[Bruce Waddington]

PHD2 has a server interface that is used by session manager apps or any other script/app that needs to interact with it.  The server interface is documented here:

https://github.com/OpenPHDGuiding/phd2/wiki/EventMonitoring

So you would probably use the capture_single_frame method followed by the save_image method to get an image and save it for subsequent plate-solving by whatever software you want to use.  There is no need for shared access to the camera.

Bruce

[NHSA]

Thanks, will take a look at that.

Rainer

[PartlyC]

I've been looking at this for a while as well.  The server does allow you to get an image and it will give you the name and path.  I have had less luck so far getting a plate solve to work. When it has, it has been very close - I started with a plate solve from an attached scope and got it very close, then did the guiding scope.  If I recall, I had to submit to Astrometery.net to get it to solve.   My seeing situation is pretty bad right now, however.   Hoping to get a lot better soon.

I suspect you have seen the various recent presentations on remote spectroscopy hosted by Shelyak:

From October of 2025

https://www.youtube.com/watch?v=IbIkk7GBlK4

and the one from December https://www.youtube.com/watch?v=30_KmGGytu4

They have some interesting options, and have Patrick has been using this for quite a bit of remote-automated data collection.  For PHD2, they tend to use the server and lock on a star and move it (I call it  a nudge) to the location of the center of the slit.  Some folks then use a different star to guide with, and keep it locked.  That's what I am trying to implement now.

With all the flex in the system that 23um slit can be hard to hit!

-Mike

[Bryan]

Mike

If speed of solution is desirable for your case, you might look into a local install of astrometry.net.  While a bit cumbersome to do this, once completed, the solves are MUCH faster...seconds, rather than minutes

Bryan

[PartlyC]

Bryan,

Wow... I did not realize you could do a local install! I will look at it right now.  I have been using NINA most recently, and have ASTAP primary with failover to astrometery.net web submission.  The latter can take some time.  Given the tight tolerance, I always feel like the clock is ticking, so this would be a great help!  ASTAP is quite fast but is less forgiving if the aim is far off..

Thank You!

-Mike

[Bryan]

Mike

I should have provided a better link.  This is the download link from what I originally posted.  https://astrometry.net/use.html

I have been using astrometry (AT) under Cygwin for years and had no reason to try WSL, but that might be less cumbersome.  What was easiest for Cygwin install was to use Andy Galasso's  AT install which includes Cygwin. It is not an quick process but his instructions are detailed. It is described on the NINA Platesolving page.  

NINA can directly use a local install of AT, but is not recommended for reasons listed there.  I use AT stand-alone and can't speak to that.  Full disclosure, I had more trouble getting ASTAP to work stand alone than AT and stayed with the latter.  That's likely due to the nut behind the wheel.

Bryan

[PartlyC]

Bryan,

Thanks again!  I dug into it quite a bit yesterday and saw the various implementation and build methods, and found the google group.  The searches popped up with that link at or near the top.  I was probably going to try ANSVR first, just to get a feel, but will also dig into the other options.  Your feedback is great, and I'll keep it in mind as I look at it.

Understand about ASTAP... my track record improved greatly when I realized I had entered the aperture and not the focal length <sigh>  Still, its so fast I'll probably leave it as the primary.

Thank you!!

-Mike

[NHSA]

Hi,

Thanks for the information but instead of getting easier it is getting more complicated  :-)
BTW, ¿do you image the spectrum with the full camera field? As far as I know (when I last read the manual)  in NINA one can not define a ROI somewhere ion the FOV of the camera. One can only define centered ROI's by using percentage numbers... for Autofocus or Guider but not for imaging... 

For example my camera has 6252 x 4176 pixels for the full imagen and I define a ROI of 5540 x 400 pixels centered at X-axis pixel 3725 and Y-axis pixel 1800. At the moment due to this limitation of NINA I am still using my old MaxImDL acquisition software.

Yes I have seen the french videos and how they do it.

Rainer

[PartlyC]

Hmm... I had not noticed that before but I am only somewhat surprised.  That said, I always get the full frames, though cropped images would be nice-to-have for the tracking images and others, in many cases. That may be left to the image processing software. It is really more a question for the NINA forum, however.... Are you on the thread "Platesolve Plus" in the NINA plugin-forum?

I am not 100% clear on what specifically you want to do with those images.  PHD2 supports a number of ROI capabilities for finding a star and getting its image in their server implementation.  I have not yet used them.  I do plan to in the near future.

-Mike

[NHSA]

quote: "I am not 100% clear on what specifically you want to do with those images."

Very simple. Why should I download 3008 x 3008 pixel big image if my spectroscope devilvers me a line a field ox 3008 x 400 pixels and that is enough for further processing in PixInsight and a Spectro analyze software?

Saves a lot of space on the hard disk :-)

Rainer

Thanks for the tip "Platesolve Plus" Unfortunately Spectroscopy in amateur field is a niche market... and for placing the star automatically into the PHD2 slit one needs to have a PHD in 10 programming languages :-) :-) and at my age 73 I am not able to learn that anymore Sorry and as far as I know nobody has made a software available be it paid or free...

[PartlyC]

Rainer,

Ya, I certainly can't argue about the space advantage of the cropped images!  

The availability of these tools is definitely an issue.  I am hopeful that there will be a couple of tools available soon to automate the "slit centering" issue, as well as others, at least as plugins like David's.  I see a number of people working it now.  Some do it with plate solving, and others by "nudging" the image in PHD2 (or another program<g>).  I know folks have implemented others, but are generally, well, embarrassed to share the code (not to mention getting into the "support" business).  Also, the best technique will vary by equipment, I think.. Spectrometer, telescope and mount.  I know the tracking images I get from my LHires III are VERY different from what I get on  my UVEX. 

And I am 68 myself, and yes, I want to figure this out SOON! <g>

-Mike

[NHSA]

Hi Mike,

I just looked into NINA discord and found PlatesolvePlus.

Is this already real or is it at the moment at a beta stage.

Can one download something?

Thanks Rainer

PD. yes I have seen the 2 Youtube presentations but they have solved it but not available for general public and as I wrote before tons of DIY programming...

[NHSA]

I found it in the NINA Plugins = PlateSolvePlus

Will test it these days.

Now we need a Plugin in order to define a random ROI :-) :-)

Rainer