Plate Solving woes in APT

[smr]

I've been imaging for several nights on NGC 2023 and at the beginning of each session it's been fine. However, when I come to do a Meridian Flip, and have to plate solve again, it just endlessly searches and then timeout. I've uninstalled PS2 and reinstalled everything including all sky plate solver etc. and checked all parameters (pixel size, focal length, sensor dimensions etc.) Well, doing some testing, I just loaded an image from last night, to see if it would plate solve on NGC 2023 (which is the blue nebulosity near the Horsehead and Flame Nebula - between them), I'm choosing this to centre my fov for framing. Again it wouldn't plate solve on NGC 2023 - so, unable to think of what else could be causing it, I decided to choose the Flame Nebula from the list of objects in the objects browser, and doing that it solved within a few seconds. So why is it that it's very temperamental with plate solving on NGC 2023? Last night I wasted a couple of hours trying to get it to plate solve but it just wasn't having it!

Edited January 13, 2022 by smr

[Dinglem]

@smr have you tried Astap https://sourceforge.net/projects/astap-program/   it seems to work well for me and it's very quick.

[smr]

Thanks. Looks like good software.

I've fixed the plate solving problem, in APT under Tools I had the CCD width as 23.5x15.7    - I've changed it to 23x15 and it's working fine. I can't believe I've wasted a couple of hours of imaging, and a couple of hours this morning troubleshooting that!

[newbie alert]

Glad you got it rectified but do load up ASTAP it's very fast, even as a blindsolver it's faster than ps2

[Budgie1]

#3 for ASTAP. 👍

I had a few plate solving failures when I only had PS2 and ASPS installed. Added ASTAP, it's really fast and I haven't had a failure since.

[scotty38]

ASTAP here too

[smr]

Does it work with APT though?

 

[smr]

... or do I even need it to? Will it plate solve for me itself? I don't want to plate solve an image, I need to put in RA and DEC values for the mount to slew to.

[smr]

Ah I see it's integrated in APT, I've installed it. I think I can see what you mean in terms of how fast it is but I'm generally having problems with plate solving so even that is failing to detect stars.

I really don't understand what the problem is, seen quite a few errors with PS2 over the last couple of days...

Overflow errors 

Solves, but then when I click on Goto++ it doesn't solve, it just endlessly searches, upon clicking on PS2 and 'Show image' (imagetosolve.fit) the image is just black, like a dark frame, hence why it can't solve. I don't understand the reason behind that.

I've tried it with the image being stretched, without, made sure all the settings are correct - focal length, pixel size, sensor width. 

I've uninstalled and removed everything to do with plate solving - even all sky plate solver, deleted PS2 and APM directories and reinstalled everything, even uninstalled APT and then installed the latest beta ( but I don't think it's a complete uninstallation as all the settings were still there so they must be saved separately to the default installation folder)

The only thing I can think of is formatting my laptop and reinstalling, because if that doesn't fix it nothing will. 

The laptop itself is getting on a bit (i3 CPU, 4GB RAM but it has an SSD I installed in it so it's quick enough) and it has been absolutely fine plate solving for a couple of years now, I've only just started having this problem where I'll need to re-solve mid way through the session and it just sits there trying to solve for ages before timing out for whatever reason. At least I can see one reason why it isn't solving when I click on Goto++ (with the black imagetosolve image) but I don't know why that is.

[Budgie1]

First off, check the exposure setting in Point Craft. Open Point Craft > Settings > about half way down the window "Default PointCraft Exposure (s)". This setting is what APT will use during the GoTo++ part of plate solving. I think the gain is taken from the Camera tab gain setting. 

Also in the setting window, check that you don't have the box marked "Use PS2 for Near Solving" selected.

I also have the "Solve un-stretched DSLR images" selected in my settings but I don't know if this will have any effect.

Edited January 14, 2022 by Budgie1

[newbie alert]

10 hours ago, smr said:

Does it work with APT though?

 

Of course

Although reading further down your other replies suggests to me you have either a camera setting issue or a camera issue..

Personality I've never had to use more that a 4 secs for luminance  and 6 secs for ha exposure length to see plenty of stars... if it can't see stars it can't solve, whatever software is being used

Have a look in your camera settings under tools

Edited January 14, 2022 by newbie alert
Added info

[smr]

The strange thing is that it does solve with ASTAP when you click Solve, but when I click on Goto++ it doesn't - saying there aren't enough stars. Very strange, it can obviously see enough stars to do the normal 'Solve' but not the Goto++ !! 

Quite infuriating, the only thing left to do is format and reinstall everything I think.

[smr]

Just to recap as I've just posted this on APT...

Strange problem with plate solving - It's been working fine for over a year or two since I started using plate solving PS2 in APT. The past week it's been really problematic. It's been very temperamental in whether it solves an image or not. Sometimes it will solve when you click on Solve but when you then click on Goto++ to get a more accurate centred plate solve it just endlessly tries solving before timing out. What I noticed, interestingly, is that when I click 'Show Image' it just shows a dark / black frame with no stars, it looks basically like a dark frame. So I'm wondering why this would be, it's (sometimes) fine with the first Solve and image, so why not when I click on 'Goto++' ? I installed ASTAP last night and that is the same thing, it will Solve within seconds, and then clicking on Goto++ it says it could not find enough stars. Weird! I've attached an image, if anyone could have a look (specifically Object Calculator) to see if I've put the settings in right - my scope is 430mm and 2600MC Pro is 3.76 pixels, 23x15mm Sensor dimensions. I'd really appreciate any help, I'm an IT tech so I consider myself quite proficient at troubleshooting, but I've ran out of ideas and the only thing left would be to format and reinstall Windows and all software... unless anyone can help!?

[Budgie1]

Can you post a screenshot of your Camera Tab settings and the Point Craft Settings window? 

This sounds more like a setting issue than a hardware/software problem, so seeing what settings you're using may identify where to look.

[smr]

Camera tab

 

 

 

Plate solve settings..

 

Choosing an image from last night and solving works as seen below, but if I was imaging, and clicked on Goto++ ASTAP would say could not find enough stars and PS2 would time out

 

[smr]

Here's an excerpt from the log last night where it shows successfully solving an image, but then not able to solve when clicking on Goto++

 

2022/01/14 00:12:36 (UT 2022/01/14 00:12:36)    Op    Bulb Exp : Bin(1x1), Gain(100), Offset(N/A), Duration(40") CCD T(2°C)
2022/01/14 00:12:36 (UT 2022/01/14 00:12:36)    DBG    Image Timer
2022/01/14 00:12:36 (UT 2022/01/14 00:12:36)    DBG    CCD_IMAGE_TIMER, Image Ready - false
2022/01/14 00:12:37 (UT 2022/01/14 00:12:37)    DBG    CCD Image Ready
2022/01/14 00:12:37 (UT 2022/01/14 00:12:37)    DBG    CCD Image Ready 2
2022/01/14 00:12:37 (UT 2022/01/14 00:12:37)    DBG    CCD Image Ready3
2022/01/14 00:12:38 (UT 2022/01/14 00:12:38)    DBG    Get BlackLevel -  Exception occurred.
2022/01/14 00:12:38 (UT 2022/01/14 00:12:38)    DBG    Get BlackLevel -  Exception occurred.
2022/01/14 00:12:39 (UT 2022/01/14 00:12:39)    DBG    File saved: C:\APT_Images\CameraCCD_1\2022-01-13\Single__0073_Bin1x1_40s__2C.fit
2022/01/14 00:12:42 (UT 2022/01/14 00:12:42)    DBG    Image Saved
2022/01/14 00:12:42 (UT 2022/01/14 00:12:42)    DBG    FocusCraft - Min:414, Max:65535, bits:16
2022/01/14 00:12:47 (UT 2022/01/14 00:12:47)    Op    FocusCraft called - HFD: 9.76, FWHM: 1.20, Stars: 86, Focuser: 2572
2022/01/14 00:12:47 (UT 2022/01/14 00:12:47)    DBG    CCD Image Ready4
2022/01/14 00:12:47 (UT 2022/01/14 00:12:47)    DBG    ResumeAfterTransferCCD
2022/01/14 00:12:47 (UT 2022/01/14 00:12:47)    DBG    No pause
2022/01/14 00:12:47 (UT 2022/01/14 00:12:47)    Op    Finished Plan - Bulb Shoot
2022/01/14 00:12:55 (UT 2022/01/14 00:12:55)    DBG    PointCraft (ASTAP): Ra 05:41:38, Dec -02:15:33, FOVx 3.06577530, FOVy 1.99941864
2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    DBG    SC Status - checking for plan triggers
2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    Op    PointCraft: Image solved in 5s.
2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    Op    PointCraft: Detected FOV size (arc-minutes) - 182.78 x 122.16; focal length - 441.75mm
2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    DBG    PointCraft: Found Ra 5.6939783096, Dec -2.2593517303
2022/01/14 00:13:08 (UT 2022/01/14 00:13:08)    Op    Synchronized to the given coordinates.
2022/01/14 00:13:21 (UT 2022/01/14 00:13:21)    Op    Stopping PHD2 Guiding.
2022/01/14 00:13:22 (UT 2022/01/14 00:13:22)    DBG    Slew start detected (1).
2022/01/14 00:13:22 (UT 2022/01/14 00:13:22)    DBG    Slew command to RA/Dec started.
2022/01/14 00:13:22 (UT 2022/01/14 00:13:22)    DBG    Slew move started.
2022/01/14 00:13:22 (UT 2022/01/14 00:13:22)    DBG    Generated coordinates for APT Dithering.
2022/01/14 00:13:22 (UT 2022/01/14 00:13:22)    DBG    PHD2 Guiding Stopped.
2022/01/14 00:13:26 (UT 2022/01/14 00:13:26)    DBG    Slew move finished.
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Going to change Bin
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    No Binning change
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Checking Exp
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    No filter change
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Exposure is starting
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Gain set with Camera Tab value
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Get BlackLevel -  Exception occurred.
2022/01/14 00:13:31 (UT 2022/01/14 00:13:31)    DBG    Exposure Started
2022/01/14 00:14:11 (UT 2022/01/14 00:14:11)    DBG    Exposure is ending
2022/01/14 00:14:11 (UT 2022/01/14 00:14:11)    Op    PointCraft Exp : Bin(1x1), Gain(100), Offset(N/A), Duration(40") CCD T(2°C)
2022/01/14 00:14:11 (UT 2022/01/14 00:14:11)    DBG    Image Timer
2022/01/14 00:14:11 (UT 2022/01/14 00:14:11)    DBG    CCD_IMAGE_TIMER, Image Ready - false
2022/01/14 00:14:12 (UT 2022/01/14 00:14:12)    DBG    CCD Image Ready
2022/01/14 00:14:12 (UT 2022/01/14 00:14:12)    DBG    CCD Image Ready 2
2022/01/14 00:14:12 (UT 2022/01/14 00:14:12)    DBG    CCD Image Ready3
2022/01/14 00:14:13 (UT 2022/01/14 00:14:13)    DBG    Get BlackLevel -  Exception occurred.
2022/01/14 00:14:14 (UT 2022/01/14 00:14:14)    DBG    File saved: C:\APT_Images\CameraCCD_1\TemporaryStorage\Focusing_CCD.fit
2022/01/14 00:14:16 (UT 2022/01/14 00:14:16)    DBG    Image Saved
2022/01/14 00:14:16 (UT 2022/01/14 00:14:16)    DBG    CCD Image Ready4
2022/01/14 00:14:16 (UT 2022/01/14 00:14:16)    DBG    ResumeAfterTransferCCD
2022/01/14 00:14:16 (UT 2022/01/14 00:14:16)    DBG    Make a pause
2022/01/14 00:14:23 (UT 2022/01/14 00:14:23)    OpErr    PointCraft: Solving failed. ASTAP returned 'Not enough stars.'
2022/01/14 00:14:23 (UT 2022/01/14 00:14:23)    OpErr    PointCraft Error: Solving failed, will try to make new image after 30s and solve it. Attempt 1.

[Budgie1]

Okay, so on the top image you have the exposure set to 20s with gain 100. This will be the settings for the image when you hit the "Shoot" button. This image is being plate solved with no issues?

Normally I would increase the gain on my ASI294MC Pro to 300 for plate solving with an exposure of 10s. This give a nice bright image for plate solving and then I put it back to Unity Gain of 120 in the imaging plan.

In the Point Craft settings, you don't have a default exposure set, so it should default to the same settings you have on the Camera page (20s @ 100 Gain), but I wonder if it is. Unfortunately the log you posted doesn't show an image taken during the GoTo++ phase, so we can't see what exposure has been used. It may be worth adding something in there, like 10 or 20s, then that will always be used by Point Craft, no matter what you have set in the Bulb setting on the Camera tab.

I can't see anything else obvious at this stage, but I would try increasing the gain when plate solving.  

[scotty38]

I cannot tell if ASTAP is being used in both cases but did you install, say, the H17 database with ASTAP?

[smr]

2 hours ago, Budgie1 said:

Okay, so on the top image you have the exposure set to 20s with gain 100. This will be the settings for the image when you hit the "Shoot" button. This image is being plate solved with no issues?

Normally I would increase the gain on my ASI294MC Pro to 300 for plate solving with an exposure of 10s. This give a nice bright image for plate solving and then I put it back to Unity Gain of 120 in the imaging plan.

In the Point Craft settings, you don't have a default exposure set, so it should default to the same settings you have on the Camera page (20s @ 100 Gain), but I wonder if it is. Unfortunately the log you posted doesn't show an image taken during the GoTo++ phase, so we can't see what exposure has been used. It may be worth adding something in there, like 10 or 20s, then that will always be used by Point Craft, no matter what you have set in the Bulb setting on the Camera tab.

I can't see anything else obvious at this stage, but I would try increasing the gain when plate solving.  

Yep it uses the Camera tab bulb exposure for the plate solving time, usually I have that set to 10 and it's always been fine, I was trying 20s to see if that would work.

The weird thing is, like I say, it does plate solve properly when clicking on 'Solve' - most of the time, but just refused completely to solve when clicking on Goto++ after taking an image and trying to solve it just won't.

In the log you can see this...

When I clicked on 'Solve' for the current image it did so successfully:

2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    Op    PointCraft: Image solved in 5s.

When clicking on Goto ++ (after selecting the same object in the object list) it fails:

2022/01/14 00:14:23 (UT 2022/01/14 00:14:23)    OpErr    PointCraft: Solving failed. ASTAP returned 'Not enough stars.'

I'm not sure it's anything to do with gain as I've always used Gain 100 since I bought my Camera which was in February and plate solving has always been flawless with that gain, until this week.

 

[smr]

19 minutes ago, scotty38 said:

I cannot tell if ASTAP is being used in both cases but did you install, say, the H17 database with ASTAP?

H18

[Budgie1]

3 hours ago, smr said:

Yep it uses the Camera tab bulb exposure for the plate solving time, usually I have that set to 10 and it's always been fine, I was trying 20s to see if that would work.

The weird thing is, like I say, it does plate solve properly when clicking on 'Solve' - most of the time, but just refused completely to solve when clicking on Goto++ after taking an image and trying to solve it just won't.

In the log you can see this...

When I clicked on 'Solve' for the current image it did so successfully:

2022/01/14 00:13:00 (UT 2022/01/14 00:13:00)    Op    PointCraft: Image solved in 5s.

When clicking on Goto ++ (after selecting the same object in the object list) it fails:

2022/01/14 00:14:23 (UT 2022/01/14 00:14:23)    OpErr    PointCraft: Solving failed. ASTAP returned 'Not enough stars.'

I'm not sure it's anything to do with gain as I've always used Gain 100 since I bought my Camera which was in February and plate solving has always been flawless with that gain, until this week.

 

What I don't know is whether the ATP uses the un-stretched or histogram stretched version of the image for plate solving.

Your log says it couldn't find enough stars and you said above that the image taken when using GoTo++ was black, like a dark frame. Did you apply a histogram stretch to this image to see if there were stars in it or whether it was just a black image? If there were stars with the stretch then increasing the gain setting is faster & more effective than increasing the exposure time.

One other thing; when you were doing this plate solving, was there a LP or narrowband filter in the imaging train? If so, it would make the image darker then with just a UV/IR cut filter.

[newbie alert]

Jusr to add , in your log it says get blacklevel, exception occurred..

And to add to all this my goto++ have been playing up, not to the point it won't solve, but at the stage where it moves the mount and then solves again, except it doesn't solve... so I just redo it until the solve and the goto++ match

[smr]

Right, I've finally fixed my plate solving. For a few days I've had lots of problems and errors trying to plate solve. Sometimes, more often than not it would 'Solve' but when clicking on Goto++ it would time out and not solve on that... I downloaded ASTAP and that would 'Solve' but Goto++ said it didn't return enough Stars - not sure why as how did it solve in the first place? Anyway, after lots of frustration, I've installed 3.88 over the top of 3.92 (I tried 3.92 and it was the same) and after installing 3.88 plate solving worked first time, straight away, target within 1 pixel. Thought this might help if anyone else has plate solving problems after 3.88

Thanks for your help anyway guys, appreciated.

[simmo39]

On 17/01/2022 at 19:14, smr said:

Right, I've finally fixed my plate solving. For a few days I've had lots of problems and errors trying to plate solve. Sometimes, more often than not it would 'Solve' but when clicking on Goto++ it would time out and not solve on that... I downloaded ASTAP and that would 'Solve' but Goto++ said it didn't return enough Stars - not sure why as how did it solve in the first place? Anyway, after lots of frustration, I've installed 3.88 over the top of 3.92 (I tried 3.92 and it was the same) and after installing 3.88 plate solving worked first time, straight away, target within 1 pixel. Thought this might help if anyone else has plate solving problems after 3.88

Thanks for your help anyway guys, appreciated.

I have had similar problems and have now back dated to 3.88. Will see if it helps next clear night, thanks for the pointer.

[Same old newbie alert]

New versions solve the go-to++ issues... It did have a blip in a few versions so use a version 3.9 and under or the last 2 cutrent versions..