i did that and still no success, also i tried with my ed80/atik383 which haw wider field and still no luck

For some reason the sensor size is locked at 1280x1024 and pixel size at 5.2. i found it by updating the image bounds from the solver options and checking it with 12dstring fov calculator. The sensor specs don't seem to change even when i save the configuration and although at the ccd simulator tab it has the correct values stored. And even whit these locked values the solver still fails.

i bump this thread again as i am looking at the log files and i notice these lines

[2018-11-30T08:20:42.842 EET DEBG ][ org.kde.kstars.indi] - CCD Simulator : "Error: Invalid range for Width (WIDTH). Valid range is from 1 to 1280. Requested value is 3362 "
[2018-11-30T08:20:42.962 EET DEBG ][ org.kde.kstars.indi] - < CCD Simulator >: < CCD_ROTATION >
[2018-11-30T08:22:36.460 EET DEBG ][ org.kde.kstars.indi] - CCD Simulator : "Error: Invalid range for Width (WIDTH). Valid range is from 1 to 500. Requested value is 3362 "
[2018-11-30T08:22:37.650 EET DEBG ][ org.kde.kstars.indi] - CCD Simulator : "Error: Invalid range for Width (WIDTH). Valid range is from 1 to 500. Requested value is 3362 "

Might seam a silly question but have you used a real CCD device and does the error occur then - might be a problem with the CCD Simulator. Especially as you implied you had reinstalled Ekos/Indi which would have cleared any SQL D/Base errors or XML file problems - Think the CCD details are in the SQL D/Base and for the simulator they come preloaded - might be wrong
It is also supported by he fact hat both Online/Offline platesolving fails !!!!

the problem started when i was on the field with my camera. i started my imaging procedure and when i got to polar align my mount the solver could not work.

Part of your issues might be related to your reinstallation of KStars.
In any case, please review carefully which ccd is used in the alignment tab each time you do a test. You reported a message about the CCD simulator complaining about a width that pertains to your Atik. Then, in the captures, the CCD in use is the guide simulator. This might be clear for you, but it complicates the task for us. Now it's possible your .local/share/kstars folder has issues, but I can't find evidence in the logs. You may rename that folder to test.
First thing off, as Wouter said, don't try to solve closer than 15 degrees to the pole.
Next, keep in mind that each time you change your profile, some Ekos settings tightly related to your equipment become incorrect and need review. Specifically, check that the solver location is properly set to "Remote" on each of your tests.
From there, once you settled for a profile, check that the proper telescope settings are associated to the proper CCD, so that the FOV is computed properly in the alignment tab. FOV is a good indicator that something is incorrect.
If you use the CCD simulator, I assume that GSC is installed properly and provides star points when you run a capture. Note that it is possible that GSC frames could not be solved because of their artificial nature, so don't spend too much time on that.
Preliminary operation to align is capturing a frame, so if no frame appears in the alignment tab, that should be investigated first.
You may also test-drive a star field from one of your previous sessions, with its approximate RA/DEC coordinates. Slew to the approximate location, and use the Load&Slew button in the alignment tab. You can use whatever image you want, as long as the FOV matches with your configuration.

-Eric

The plot thickens, i tested with an image of m31 i had and did a blind solve, both the online and offline solvers solved the image but the odd thing is that the resolution the solver gaveme is of 1.37 which is the one i have when i bin 2x2 my camera (it was a jpg image as i had deleted the fit file). I run another test at polaris but this time with bin 2x2 and the solver works but when i bin 4x4, which is the bin i use when in the field for plate solving, the solver does not work.

Note that i am still running a simulation profile for an atik 383l+ and an rc 8, i also tried a simulation profile with the ed80 and at bin 2x2 the solver works.

At the images below i have the results of a succesful plate solve at bin 2x2 and the solver runs a bin 4x4 which failed.

You should try to test only one thing at a time. First, you should drastically increase the exposure, to keep the quality of the capture out of the equation. Then, your screenshots are showing an "online" solver for the first and an "offline" solver for the second. Both are showing 4x4 binning for the running procedure, but the second out of three solves is failed in the first image and succeeded in the second. I have difficulties following...

-Eric

At the first image i had a successful plate solve of m31 at bin 2x2 with the online solver and i was doing an online plate solving with bin 4x4, the bin 4x4 plate solving failed, the simulated scope is the rc 8.

At the second image i had a successful plate solving of polaris at bin 2x2 with the offline solver and i was doing an offline plate solving at bin 4x4 which failed, the simulated scope is the ed80.

new tests for today and i tried this. I changed the resolution of the ccd camera from 3362x2504 to 2504x3362. When i do a plate solving at bin 4x4 INDI crahses and i have to disconect the equipment since i get a message that the camera is still capturing, but plate solving works fine at 2504x3362 at bin 2x2.

Also i noticed that INDI has a Simulators profile created that i did not create, is this profile created by default from the software?

I'll assume you are testing the solver on a frame captured by the CCD simulator.
Let's try to get rid of the width issue, because the debug session is about the plate solver. If the CCD simulator complains about a width greater than 1280, don't set a width larger than 1280. Consider your CCD is 1280x1024 and that's all.
Then about the CCD simulator crashing. This seems to happen at the moment the debug log is enabled. From the log, that log level is changed in the middle of the session. So instead, configure logging in the Logs options before starting the local indiserver.
About binning, reset this to 1x1 for now, again because your objective is to test the solver, not the particular binning abilities of the simulator.
Then we can go back to solving. Your picture still displays the solver as being the "online" one. If you want to test the local solver, change this to "offline", or "remote" if your profile includes the astrometry driver.
Run a simple session where you start kstars, start your local indiserver, point to a target that is 15 degrees away from the pole, select the "Sync" option of the alignment tab, click the "Capture&Solve", wait for the result and stop the local indiserver and kstars. Then you and we can study the log.

-Eric

ok, i used a 1280x1024 ccd resolution for a 203mm 1624mm (RC 8") scope which has a resolution of 0.66"/pixel and an FOV 14.09' x 11.27' (which the alignment module rounded up to 14.1' x 11.3'). i slew the scope at M 31 and took a 4sec bin 1x1 photo using the online solver and the solve failed. i include the log file