APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO


mjb87@...
 

I tried my first ever attempt at building an automated pointing model using APPM.  (I've used manual T-Point runs in the past.) The plate-solves were all failing. I must be doing something very wrong here and would like to at least be pointed in the right direction. Is this an imaging issue, a plate-solving data problem, or what?

Equipment: Mach2, 130mm Starfire GTX, ASI2600MC-Pro
Software: All latest versions of APCC Pro, ASCOM driver, SG Pro
Platesolving: Pinpoint 6 using GSC11 downloaded file
Settings pretty much at default for a small sample (40-50 points) with horizon limits in use. I tried a range of exposure settings from 2s to 5s
Sky clear with half moon. (The failed images were not being taken in the vicinity of the moon.)
Focus: checked in SG Pro.

First issue: if I connect directly to the camera I can't adjust camera settings (e.g., cooling temperature). Plus, the process got hung up on imaging. So I abandoned that and connected the camera via SGPro and used that instead. It seemed to help. Is this the right approach?

Second (bigger) issue: All but one plate-solve failed. Here are the results of my fourth run attempt, before I abandoned the effort.



Lots of the following errors. What do these suggest is wrong?
- Operation stopped by client in Checkpoint event
- Failed to solve plate possibly due to input plate scale being too far off
- Bad Y scale
- Bad X scale

I want to try tomorrow if conditions permit. What am I doing wrong? Amy assistance would be greatly appreciated.

Marty


Ray Gralak
 

Hi Marty,

 

The errors are coming from the plate-solver application. Here are some suggestions:

 

1. Instead of using GSC11, use USNO A2.0 or UCAC4. Note that you will likely have to download those catalogs.

 

2. On APPM's Plate Solve Settings tab:

 

            a) Increase "Max Solve Time (sec)" to 60 seconds. The "Operation stopped by client checkpoint operation" means that the plate solve didn't solve before the time in this field. If your computer is slow, or you have large images this can happen. You can reduce image size by binning or doing a subframe on APPM's Camera Settings tab. Also, uncheck the Auto Dark Subtract option on that tab if you have it enabled.

 

            b) Enable "Refine image scale from next solved image".

 

            c) Enable "Use last plate solve's offset as hint for next plate solve".

 

            d) Enable "Try All-Sky Plate Solving if available". And set the timeout to 300 seconds.

 

            e) Make sure the X and Y plate solve scale (unbinned) is set approximately to your image scale.

 

3. On APPM's "Run" tab, enable "Recall near Zenith at start". You should always leave this enabled so that the all models you create will have the same starting reference point.

 

4. Before you start a run for the very first time try a "Plate Solve and Recal" on APPM's run tab. If this succeeds then everything should run smoothly.

 

Here is a screenshot of the Plate Solve Settings I mentioned above:

 

-Ray

 

> -----Original Message-----

> From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io

> Sent: Friday, March 19, 2021 6:16 PM

> To: main@ap-gto.groups.io

> Subject: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

>

> I tried my first ever attempt at building an automated pointing model using APPM.  (I've used manual T-Point runs in

> the past.) The plate-solves were all failing. I must be doing something very wrong here and would like to at least be

> pointed in the right direction. Is this an imaging issue, a plate-solving data problem, or what?

>

> Equipment: Mach2, 130mm Starfire GTX, ASI2600MC-Pro

> Software: All latest versions of APCC Pro, ASCOM driver, SG Pro

> Platesolving: Pinpoint 6 using GSC11 downloaded file

> Settings pretty much at default for a small sample (40-50 points) with horizon limits in use. I tried a range of

> exposure settings from 2s to 5s

> Sky clear with half moon. (The failed images were not being taken in the vicinity of the moon.)

> Focus: checked in SG Pro.

>

> First issue: if I connect directly to the camera I can't adjust camera settings (e.g., cooling temperature). Plus, the

> process got hung up on imaging. So I abandoned that and connected the camera via SGPro and used that instead.

> It seemed to help. Is this the right approach?

>

> Second (bigger) issue: All but one plate-solve failed. Here are the results of my fourth run attempt, before I

> abandoned the effort.

>

>

>

> Lots of the following errors. What do these suggest is wrong?

> - Operation stopped by client in Checkpoint event

> - Failed to solve plate possibly due to input plate scale being too far off

> - Bad Y scale

> - Bad X scale

>

> I want to try tomorrow if conditions permit. What am I doing wrong? Amy assistance would be greatly appreciated.

>

> Marty


mjb87@...
 
Edited

Thanks, Ray. This is very helpful.

I am getting big images (17MB) but I did check 1/2 subframe and Auto Dark Subtract was unchecked. The computer itself is very fast (6-core Xeon processor at 2.7GHz) but I'll add time for the solution set.

I'm surprised at the need for the USNO A2.0 (I do have it downloaded) given the focal length and thought the GSC11 would be enough. But I'll try USNO A2.0 tonight.

Pixel scale is off for sure: closer to 0.9 than 1.6. (3.76 pixel and 819mm) so I'll make that correction.

As for #3 (Recal Near Zenith at Start) my dome shutter configuration made that difficult so I gave up on that. When it moves to that position the telescope is on the same side of the mount as the unopened part of the shutter. So even though the shutter is open past zenith it still blocks about 1/2 of image. If I rotate the dome 180 degrees it would work. (Ironically, I got a good solution at one point at zenith even though the telescope was partially blocked.)

I'll check everything else and give it a try. Many thanks.

Marty


annaski
 

Hi Marty, I'm not sure if you've platesolved with this camera before.

I have the 6200 and find that I have to bin at least 2x2 to get a successful platesolve.


mjb87@...
 
Edited

Thanks. This is my first attempt at platesolving with any camera. Still learning.


mjb87@...
 

Ray -- one more question...

As mentioned above, the standard automated approach makes it difficult for me to do a Plate Solve and Recal at Zenith.

If I set up the dome/shutter for a clear shot in that configuration, can I do a Plate Solve and Recal (using that button) at zenith , bring back the dome to the original position (180 degree swing), and then run the full modeling WITHOUT checking the Recal Near Zenith at Start option? I assume that would be equivalent since I've just done a recal at that point.

Marty


Ray Gralak
 

Hi Marty,

As for #3 (Recal Near Zenith at Start) my dome shutter configuration made that difficult so I gave up on that. The
telescope is on the same side of the mount as the unopened part of the shutter. So even though the shutter is open
past zenith it still blocks about 1/2 of image. If I rotate the dome 180 degrees it would work.
Oh, okay, then you are right to disable "Recal Near Zenith at Start".

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io
Sent: Saturday, March 20, 2021 4:35 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

Thanks, Ray. This is very helpful.

I am getting big images (17MB) but I did check 1/2 subframe and Auto Dark Subtract was unchecked. The
computer itself is very fast (6-core Xeon processor at 2.7GHz) but I'll add time for the solution set.

I'm surprised at the need for the USNO A2.0 (I do have it downloaded) given the optical configuration and thought
the GSC11 would be enough. But I'll try USNO A2.0 tonight.

As for #3 (Recal Near Zenith at Start) my dome shutter configuration made that difficult so I gave up on that. The
telescope is on the same side of the mount as the unopened part of the shutter. So even though the shutter is open
past zenith it still blocks about 1/2 of image. If I rotate the dome 180 degrees it would work.

I'll check everything else and give it a try. Many thanks.

Marty


Ray Gralak
 

I don't see how I can change camera settings (cooler, binning) directly through APPM.
Marty, you can set binning on APPM's "Camera Settings" tab. However, you should setup other properties, like cooling temperature, in the manufacturer's camera control application or a third-party camera control. That being said, most of the new CMOS sensors are so clean that cooling the camera is not actually necessary for the purpose of plate-solving.

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io
Sent: Saturday, March 20, 2021 6:58 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

Thanks. This is my first attempt at platesolving with any camera. Still learning.

I don't see how I can change camera settings (cooler, binning) directly through APPM. So I assume the right way to
do this is to run (in my case) SG Pro, edit the camera settings there, and connect in APPM to SG Pro.


Ray Gralak
 

If I set up the dome/shutter for a clear shot in that configuration, can I do a Plate Solve and Recal (using that button)
at zenith , bring back the dome to the original position (180 degree swing), and then run the full modeling WITHOUT
checking the Recal Near Zenith at Start option? I assume that would be equivalent since I've just done a recal at
that point.
Yes, that would work.

-Ray


-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io
Sent: Saturday, March 20, 2021 7:08 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

Ray -- one more question...

As mentioned above, the standard automated approach makes it difficult for me to do a Plate Solve and Recal at
Zenith.

If I set up the dome/shutter for a clear shot in that configuration, can I do a Plate Solve and Recal (using that button)
at zenith , bring back the dome to the original position (180 degree swing), and then run the full modeling WITHOUT
checking the Recal Near Zenith at Start option? I assume that would be equivalent since I've just done a recal at
that point.

Marty


mjb87@...
 

Thanks, again, Ray. Yes, I now see the binning option in APPM. I'll try 2x2.


Ray Gralak
 

Thanks, again, Ray. Yes, I now see the binning option in APPM. I'll try 2x2.
Okay great! If you have any other issues, please post here.

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io
Sent: Saturday, March 20, 2021 8:35 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

Thanks, again, Ray. Yes, I now see the binning option in APPM. I'll try 2x2.


mjb87@...
 

Thanks, Ray and everyone else,

It worked. Only built a 35-point model, just to test the process. One failed solve -- I think it was related to focus. Need to figure out how to embed autofocus routine into APPM. I guess I can do that within SG Pro. No need for all-sky solve. My solutions took about 5-10 seconds.

Did a five-minute unguided image and got nice round stars. Good pointing. Good tracking.

However, I need to redo my horizon limits. I started getting star trails -- then realized I was outside the horizon limits (though I was getting good images) and I had set "turn tracking off." So of course I got trails. Once I realized that all was good.

Thanks for the help.

Marty


Ray Gralak
 

Hi Marty,

It worked. Only built a 35-point model, just to test the process. One failed solve -- I think it was related to focus.
Need to figure out how to embed autofocus routine into APPM. I guess I can do that within SG Pro. No need for all-
sky solve. My solutions took about 5-10 seconds.
I'm glad you got it working!

About needing to do autofocus midrun, is each point taking too long because it is waiting for dome movement to complete?

-Ray



-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of mjb87 via groups.io
Sent: Saturday, March 20, 2021 6:43 PM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO

Thanks, Ray and everyone else,

It worked. Only built a 35-point model, just to test the process. One failed solve -- I think it was related to focus.
Need to figure out how to embed autofocus routine into APPM. I guess I can do that within SG Pro. No need for all-
sky solve. My solutions took about 5-10 seconds.

Did a five-minute unguided image and got nice round stars. Good pointing. Good tracking.

However, I need to redo my horizon limits. I started getting star trails -- then realized I was outside the horizon limits
(though I was getting good images) and I had set "turn tracking off." So of course I got trails. Once I realized that all
was good.

Thanks for the help.

Marty


David Johnson
 

Last night, it was obvious my focus was changing while I was doing point mapping.  This is probably not an unusual problem since we tend to do point mapping early in the evening when the optics are cooling down relatively rapidly.  Fortunately, it was able to solve at all points.  However, I was thinking that maybe I could put APPM on Pause and then go over to SGPro and do an quick autofocus.  Would this be possible and/or advisable?


Bill Long
 

I run focus right before a model run. That should be more than sufficient for the task. I don't think ultra critical focus is needed, although Ray would know best.


From: main@ap-gto.groups.io <main@ap-gto.groups.io> on behalf of David Johnson <dajohns37@...>
Sent: Sunday, March 21, 2021 9:04 AM
To: main@ap-gto.groups.io <main@ap-gto.groups.io>
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO
 
Last night, it was obvious my focus was changing while I was doing point mapping.  This is probably not an unusual problem since we tend to do point mapping early in the evening when the optics are cooling down relatively rapidly.  Fortunately, it was able to solve at all points.  However, I was thinking that maybe I could put APPM on Pause and then go over to SGPro and do an quick autofocus.  Would this be possible and/or advisable?


mjb87@...
 

I share David's observation. My run only took 30-35 minutes. Normally I wouldn't need to refocus. However, the rapid temperature change must have led to a slight shift in focus.  I was actually impressed without well APPM did the plate-solving on images that I would have characterized as "not in great focus." I'm thinking ahead to a much longer run (200 points) some day.

I typically have SG Pro run autofocus every 10 frames or so (2 or 3 minute subs). Would APPM automatically pause while SG Pro went through its thing?


 

>>>I typically have SG Pro run autofocus every 10 frames or so (2 or 3 minute subs). Would APPM automatically pause while SG Pro went through its thing?

I'm pretty sure you would have to have an SGP sequence running (and be imaging) to have SGP do a timed autofocus run



On Sun, Mar 21, 2021 at 2:22 PM mjb87 via groups.io <mjb87=verizon.net@groups.io> wrote:
I share David's observation. My run only took 30-35 minutes. Normally I wouldn't need to refocus. However, the rapid temperature change must have led to a slight shift in focus.  I was actually impressed without well APPM did the plate-solving on images that I would have characterized as "not in great focus." I'm thinking ahead to a much longer run (200 points) some day.

I typically have SG Pro run autofocus every 10 frames or so (2 or 3 minute subs). Would APPM automatically pause while SG Pro went through its thing?



--
Brian 



Brian Valente


Dean Jacobsen
 

On Sun, Mar 21, 2021 at 02:57 PM, Brian Valente wrote:
I'm pretty sure you would have to have an SGP sequence running (and be imaging) to have SGP do a timed autofocus run
 
Yes, that would be my opinion as well.  However, the APPM "Run" page does have a pause button so if you saw that focus was shifting, then you could potentially pause the mapping run, refocus and and then resume the mapping run.  I haven't tried this however.
 
--
Dean Jacobsen
http://astrophoto.net/wp/
Image Gallery - http://astrophoto.net/wp/image-gallery/
Astrobin Image Gallery - https://www.astrobin.com/users/deanjacobsen/ 
Amateur Radio Call Sign - W6DBJ


Bill Long
 

The real question though -- is how critical focus needs to be for the point mapper to work accurately. I would suspect that it is not that critical and any drift in terms of focus during a run would have negligible impact on the results. 


From: main@ap-gto.groups.io <main@ap-gto.groups.io> on behalf of Dean Jacobsen <deanjacobsen@...>
Sent: Sunday, March 21, 2021 3:21 PM
To: main@ap-gto.groups.io <main@ap-gto.groups.io>
Subject: Re: [ap-gto] APPM Attempt Failed - Clearly I am doing something wrong #APCC #Mach2GTO
 
On Sun, Mar 21, 2021 at 02:57 PM, Brian Valente wrote:
I'm pretty sure you would have to have an SGP sequence running (and be imaging) to have SGP do a timed autofocus run
 
Yes, that would be my opinion as well.  However, the APPM "Run" page does have a pause button so if you saw that focus was shifting, then you could potentially pause the mapping run, refocus and and then resume the mapping run.  I haven't tried this however.
 
--
Dean Jacobsen
http://astrophoto.net/wp/
Image Gallery - http://astrophoto.net/wp/image-gallery/
Astrobin Image Gallery - https://www.astrobin.com/users/deanjacobsen/ 
Amateur Radio Call Sign - W6DBJ