Problem repeats - "something terrible has happened.."

Sequence Generator Plate Solving
1

The message that I get from SGP every time is the above.

I have sequences set to go to a target, solve the image, correct, and image. Every single time it gets there quite close, takes and solves and image, and then makes an irrelevant correction, and then solves and tells me that “something terrible has happened”; it then tells me that my evening is over and it will continue to try to resurrect. This happens when I am virtually spot on target!!

I posted this some time back with the logs. There was no response.

Since then (some weeks) I have been manually going to targets, one solve with focus, and then using focus and renaming files because I don’t know what the problem is.

Lawrence Harris :disappointed:

2

Can you post a screenshot of your control panel/plate solve tab settings?

Something I’m wondering about…you don’t use a rotator. Are you sure that the rotator setting has been disabled?

Rotator.png822×497 47.4 KB

3

I have just checked; everything is set correctly.

SGP and controlpanel etc.jpg1366×768 116 KB

As mentioned before, the ‘attempt …’ setting was previously at 3 but it caused the scope to move randomnly so I reduced it to 1.

Lawrence

4

If you only have the attempt set to 1 then unless you have a really
fantastic exceptionally good expensive mount it will fail every time.

When you say having the attempts at 3 caused the scope to move randomly,
what do you mean by randomly? Does it slew to a completely different part
of the sky, or near your target but just off?

I am not saying that SGP doesn’t have something to do with this, but to me
this sounds like a communication problem at the mount level, especially if
you are getting wild slews.

5

That’s fair comment. To be accurate, the target can be visually centred and I leave a good margin of pixel error, but when I had it set to eg 3, I watched the scope slew further off target, solve correctly, then slew further off target. The solves are fine so surely it should know where to slew? That is why I changed it to 1 and then started to merely image when on target. I would love to get SGP to solve, move a small distance and then image. I even tried very large pixel errors but the slews were not correct. I have previously posted logs (long time back…) but without response.

Lawrence :worried:

6

One minor query: I have long wondered why a control panel setting cannot be saved. I can save a sequence but not a control panel. I presume that it is part of something else but when I have saved other stuff, the control panel continued with wrong settings. What am I missing?

Lawrence

7

When you save your sequence, the control panel settings will be saved with
that sequence.

But I suspect you are not understanding the relationship between an
equipment profile and the control panel. Your equipment profile acts as a
template for new sequences. You set up your equipment profile with your
usual settings, then when you create a new sequence based on the equipment
profile the settings in the control panel will reflect all the same
settings that you set up in the equipment profile.

But here’s the key point: once you create a new sequence, that sequence
(including it’s control panel settings) are totally separate from the
equipment profile. So if you make a change on the equipment profile it
will NOT be reflected in the existing sequence. And if you make a change
in the sequence’s control panel, it will NOT affect the equipment profile.
Also, every separate sequence has it’s own control panel settings.

Now, back to the original problem. What is your mount, and are you using
any kind of a pointing model?

8

Thanks Joel - I will read and memorise :slight_smile:

I used to use a Meade RCX400 with SGP and experienced comparable problems with slewing. I am now using a Celestron 11 Edge with Hyperstar (560mm f.l.) and have taken care to generate accurate profile+control panel + sequences. Various sequences for different things, eg sun (no slewing required). For any session I use Carte du Ciel to slew to target and then do focusing with SGP. This is working very well and landing spot on target ( 560mm fl is easy!). I do a full model (two stars on W and then two calibration stars on E - all good. So the first target is close to centre (sorry, center?). Last night was a classic; I was ready to try the first sequence target but as always, got the error message and then cancelled when it started the long wait to ‘rescue’. From then, I did manual imaging.
FWIW I have last night’s logfile …

Lawrence

9

From the screenshot earlier in the thread it looks like you use a Celestron mount: is that correct? Maybe a Celestron user will post up.

Mount pointing models can interfere with plate solving. After completing PA, have you tried not creating the basic alignment on the West and then East? If you try this you will of course need to increase the number of iterations.

If it is a Celestron mount you use, I’m not familiar with its ascom driver. However, you may want to check that it isn’t storing a model built up over your sessions; you do not want a stored model. Check also synch settings to see whether it adds a synch point to its internal model (you do not want this option) or whether it aligns a model (you want this option).

If you make a change to one of your settings in the control panel and wish to preserve this in your equipment profile you can select File/Save Sequence as Profile.

Any rotation prompt can arise if you leave the rotation option ticked when setting a target in the Framing & Mosaic Wizard, this prompt is made irrespective of whether you have a rotator as you can ‘manually’ rotate. Just make sure you untick the option.

HTH

10

Thanks for all these points. Yes, I am now using a new C11 Edge with Hyperstar. The initial alignment is two stars on the W and then two on the E unless I opt to keep adding. There are no further alignments added. Control panel saving comment understood.
Framing module when used, I untick rotation.

Thanks again. (ASAIK I am so far doing everything correctly.

Lawrence

11

Hi Lawrence

C11 hyperstar is the telescope an reducer system, yes?

Which mount are you using? This is the only relevant point for the centering plate solving.

I suggest you do not do the alignment as this has the potential to be the cause of your problems. All you need to do is a PA.

12

I am using a CGX mount controlled by Carte du Ciel and it does accurate pointing, mostly perhaps because of the effective focal length - 560mm. Goto targets are normally very close to centre. I have polar aligned the mount quite well.

Lawrence

13

Excellent.

I would certainly try without doing the two star West/two star East alignment after clearing any existing model. This may make all the difference and is the general advice, eg for the Losmandy and Skywatcher mounts, to not have a stored model (even if it is only a few points) unless the mount has a specialised control system.

HTH.

14

Do you mean use the CGX from power up with ‘use last alignment’ option? Its easy enough for me to try, but I would have thought that SGP would solve a test image and accept it as on target?

BTW I don’t think that there is any option to ‘ignore all pointing models’. It knows enough basic info to go to the correct target.

Lawrence

15

Hi Lawrence

The internal model stored will conflict with plate solving: this is the reason you are experiencing worsening pointing with each iteration. So I am suggesting you start your evening’s session without any model stored in the mount at all as an experiment.

I’m not familiar with Celestron’s terminology (I’m hoping a Celestron user will help here) but you do not want any stored alignment active when trying to auto centre with plate solving. Choose whichever options allows this in this experiment.

There are exceptions, eg 10 Micron or Bisque mounts, however their control systems are quite specialised.

Barry

16

OK- I understand. I will use the mode that simply takes basic location/time data and estimates a reasonable start place. Next clear sky. [Still don’t understand why SGP doesn’t say ‘close enough for me’ and start imaging.

Lawrence (with thanks)

17

The “close enough for me” comes with the tolerance in pixels that you set in the plate solving tab, eg 30 pixels.

Good luck - not sure when you’ll get clear skies :smiley:

18

You are correct Barry; I have that set to 100 pixels. I still get the ‘something terrible’ message.

Lawrence

19

Hi Lawrence

Just to be clear, you had some cler skies and tried centring with no model stored in your mount and you received the ‘something terrible’ message and your mount slewed further away from your target co-ordinates?

What happens if you solve a Frame and Focus image of any region of the sky (right click for context message and Plate Solve) and then clicking somewhere in the frame (but not the centre) and right click and select Center Here? Does this solving and centring work and the mount moves to place the region of the cursor centre frame?

Some basic questions (apologies):

  • Are the date/time and location set accurately in your mount?

  • How do you connect to your mount?

  • Is there any chance of Epoch confusion?

  • What version of SGP are you using?

  • Which ‘synch’ option do you have enabled in SGP?

20

Hi Barry

Thank you for your continuing interest in helping me crack this basic
problem.

For every session I power up the mount. The immediate choice of mount
modelling is then made. I wanted to do a ‘new’ model (because the
’terrible’ problem hit the previous session) so I did the requested two
stars (W) then two (calibration) stars E. It then has pretty good
modelling (there are various options). As usual, I used Carte du Ciel to
go to the first target for focus - currently Altair. Good focus
achieved, so I went to first target on my list. Spot on, so I triggered
SGP after checking precise centring. It quickly flashes through expose,
solve, ‘correct’, expose, solve - complain … At that point I
just tell it to do the schedule without ‘precise centring’ (knowing that
it is right on target.) To answer your query about correcting further
off target, I cannot exactly answer that. I cannot follow the log file
to see where it went.

Subsequently I find that the images are fine (except that my chosen
filenames are not used). I will note for next time to solve and ‘centre
here’; that;s easy to test - and I should have thought of that!

See below:

Hi Lawrence

Just to be clear, you had some cler skies and tried centring with no
model stored in your mount and you received the 'something terrible’
message and your mount slewed further away from your target co-ordinates?

What happens if you solve a Frame and Focus image of any region of the
sky (right click for context message and Plate Solve) and then
clicking somewhere in the frame (but not the centre) and right click
and select Center Here? Does this solving and centring work and the
mount moves to place the region of the cursor centre frame?

Some basic questions (apologies):

Are the date/time and location set accurately in your mount?

Yes, its checked at the start of every session (within a few seconds)

How do you connect to your mount?

Via a USB hub. This provides camera connection as well.

Is there any chance of Epoch confusion?

Sorry, I don’t know. I have CdC

What version of SGP are you using?

Latest - v 2.6.0.23

Which 'synch' option do you have enabled in SGP?

In the control panel, telescope setting ‘auto meridian flip’ has 'sync
behaviour set to ‘sync’ (I’m not familiar with the meaning of these
settings. I haven’t found an explanation of preferred settings in the
manual.

regards

Lawrence