Hello.
I tried ekos live for the first time last night to see if I could get everything working.
My mount, DSLR and focuser connected OK!

When trying to solve an image the online solver timed out though. I didn't know then how to tweak the settings but now afterwards I've read up on some things.
I tried submitting an image of mine to astrometry.net and it could resolve it without problem. I changed the settings in the ekos solver to reflect the same settings as the solver on astrometry web page. That meant setting -L 0.1 -H 180.0 -u dw. That is low scale is set to 0.1 degrees and high scale is set to 180 degrees.
And with that setting the solver in ekos managed to solve the image (using load and slew function to upload existing image).
However, after solving the image ekos updates the settings to -L 30 -H 33 -u dw. I was running ccd simulator when testing this and you cannot set resolution to larger than 2048x2048, which is smaller than my dslr. Is that why the solver updates the settings to these numbers?
Also the -u flag was not updated to aw when ekos changed the low and high scale to arcmin instead of degrees.

I hopefully will be able to test live again during the weekend and see if I can get it working now that I understand the settings better.

/Daniel

Ekos regenerates the image scale every time based on focal length, pixel size, and resolution. What are you values for variables? IMO, 0.1 to 180 doesn't sound right.

Hello.
Yes I understand that now. The focal length were set to my settings but when using the ccd simulator the resolution could not be set to higher than 2048x2048 which led to the solver setting wrong FOV and wrong scale variables.
The 0.1 - 180 degrees was used to set to the same default settings that nova.astrometry.net uses when you upload an image there. I wanted to see if ekos could solve an image that I know the online solver (at astrometry.net) had solved manually.

Last night I was able to set up and use the solver with my real equipment. Now the resolution was set correctly and the FOV was also correct. However the online solver didn't work. When turning on verbose logging the online solver hanged at status [status:processing]. I don't know if this is something Ekos has any control over though.


When I tried to upload an image to nova.astrometry.net it failed to solve the image as well. Maybe it wasn't good enough to solve, I don't know. You can see an example here. nova.astrometry.net/user_images/395531#original
Maybe the focus wasn't good enough, I couldn't get autofocus to work either, but that's for another post
Does Ekos get a correct response when solving fails? It seems that it did fail but ekos was just hanging on processing and not reporting that it had failed.

Another question, what settings in GPhoto do you need to have to make the images compatible with the online solver? When using the compress option and selecting a smaller image scale than raw (to reduce download time, but still using fits as output and not native) astrometry.net was not able to read the fits file. When I tried to open it in PixInsight I also got an error that it was not a valid fits file. Here is one of the files that couldn't be read. dl.dropboxusercontent.com/u/8802413/Ekos/temp8.fits
Later during the evening I got INTERNAL SERVER ERROR reply from the online solver but that has to be on their part right?

I'm going to try to install the offline solver and see if I can get that to work as well.

/Daniel

A small update.
I installed and tried the offline solver and with some changes to the low and high scale parameters I managed to get the image solved correctly.
So maybe next time I will finally get all things working

We have found there are several conditions which seem to upset the solver. Avoiding them may help your experience.

1: Trying to solve immediately after slewing. The gears of the mount may not have engaged firmly and tracking will be too poor.

2: Trying to solve on reasonably bright nebula or other DSO. We like to just bump the scope off a bit, then allow the solver to center it.

3: Forgetting to "push" the telescope coordinates into the solver. For some reason, we frequently end up with the "rough" coordinates not in the solver. Use the button to transfer them.

Also, you may find the first 2 helpful on autofocusing.

Dan

Just wanted to mention that the astrometry.net site doesn't seem to have a nice message implemented when you reach your limit of solves for the timeperiod. You'll get "INTERNAL SERVER ERROR" instead.

That might change at some point, but it is wholly under the control of the people who run astrometry.net.

Hello,

Might I suggest downloading the data files for the Astrometry. Once installed locally onto your machine, you may use the "Offline Solver" to avoid delays. While some take considerable time to download, its better to have a few too many, rather than too few of the data sets. Typical Offline solver times may be as short as <5 seconds.
And, without the lost time of upload to Astrometry.net, things are much more reliable and rewarding.

Most times when we have experienced failures, the cause has been related to a mistake on our end of Alignment. It is important to visually examine (Fits Viewer) the actual frame since settings for Alignment and settings for CCD may be different binnings and times. Often when visually examined, one can see the cause of the problem. (i.e. drift, bad focus, planes, satellites, clouds, too many or too few stars)

Hello.
Yes, I did download relevant data files from Astrometry and used the offline solver and it worked. So I'm just waiting for the next opportunity to try again.

/Daniel