×

INDI Library v2.0.7 is Released (01 Apr 2024)

Bi-monthly release with minor bug fixes and improvements

Read More...

meridian flip fails:eq6, latest git

  • Posts: 969
  • Thank you received: 94

meridian flip fails:eq6, latest git was created by alacant

Hi everyone
Meridian flip at around 04:40 fails and the mount keeps trackling on the wrong side of the meridian.
The scheduler does not park the mount at sunrise despite the shutdown procedure 'Park Mount' being checked.
I was lucky: no damage.

Any ideas?
TIA and cleasr skies,
Steve

esl and log: drive.google.com/drive/folders/1vg9JM_c1...EmVpcLXQ?usp=sharing
3 years 9 months ago #56145

Please Log in or Create an account to join the conversation.

Replied by Jasem Mutlaq on topic meridian flip fails:eq6, latest git

So it's related to PHD2 again, but here is the log part that led to this event:
[2020-06-28T04:30:37.237 CEST DEBG ][ org.kde.kstars.ekos.scheduler] - Guide State "Dithering"
[2020-06-28T04:30:37.292 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: event: "{\"Event\":\"GuideStep\",\"Timestamp\":1593311435.618,\"Host\":\"E5430\",\"Inst\":1,\"Frame\":5527,\"Time\":11214.549,\"Mount\":\"INDI Mount [EQMod Mount]\",\"dx\":-0.057,\"dy\":-0.175,\"RADistanceRaw\":0.175,\"DECDistanceRaw\":-0.071,\"RADistanceGuide\":0.028,\"DECDistanceGuide\":0.000,\"RADuration\":13,\"RADirection\":\"West\",\"StarMass\":3129,\"SNR\":38.48,\"HFD\":4.19,\"AvgDist\":0.21}\r\n"
[2020-06-28T04:30:37.293 CEST DEBG ][   org.kde.kstars.ekos.capture] - meridianFlipStatusChanged:  "FLIP_PLANNED"
[2020-06-28T04:30:37.293 CEST DEBG ][   org.kde.kstars.ekos.capture] - setMeridianFlipStage:  "MF_REQUESTED"
[2020-06-28T04:30:37.293 CEST INFO ][     org.kde.kstars.ekos.mount] - "Meridian flip waiting."
[2020-06-28T04:30:37.301 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: event: "{\"Event\":\"LockPositionSet\",\"Timestamp\":1593311437.293,\"Host\":\"E5430\",\"Inst\":1,\"X\":89.528,\"Y\":204.160}\r\n"
[2020-06-28T04:30:37.302 CEST INFO ][     org.kde.kstars.ekos.guide] - "PHD2: Lock Position Set."
[2020-06-28T04:30:37.302 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: event: "{\"Event\":\"GuidingDithered\",\"Timestamp\":1593311437.295,\"Host\":\"E5430\",\"Inst\":1,\"dx\":-4.762,\"dy\":-4.487}\r\n"
[2020-06-28T04:30:37.302 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: event: "{\"Event\":\"SettleBegin\",\"Timestamp\":1593311437.295,\"Host\":\"E5430\",\"Inst\":1}\r\n"
[2020-06-28T04:30:37.303 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: response: "{\"jsonrpc\":\"2.0\",\"result\":0,\"id\":778}\r\n"
[2020-06-28T04:30:37.417 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[SCOPE] Compute local time: lst=20.92120993 (20:55:16.36) - julian date=2459028.60458347 "
[2020-06-28T04:30:37.417 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":j1\", 0 bytes written "
[2020-06-28T04:30:37.437 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=D6D6A2\", 8 bytes read "
[2020-06-28T04:30:37.438 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[SCOPE] GetRAEncoder() = 10671830 "
[2020-06-28T04:30:37.438 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":j2\", 0 bytes written "
[2020-06-28T04:30:37.455 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=7D8390\", 8 bytes read "
[2020-06-28T04:30:37.455 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[SCOPE] Current encoders RA=10671830 DE=9470845 "
[2020-06-28T04:30:37.455 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":f1\", 0 bytes written "
[2020-06-28T04:30:37.469 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=111\", 5 bytes read "
[2020-06-28T04:30:37.470 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":f2\", 0 bytes written "
[2020-06-28T04:30:37.486 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=301\", 5 bytes read "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[DEBUG] Timed guide West 2176 ms at rate 10.5287  "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[DEBUG] StartRATracking() : trackspeed = 25.5698 arcsecs/s, computed rate = 1.7 "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[DEBUG] SetRARate() : rate = 1.7 "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[MOUNT] SetMotion() : Axis = 1 -- dir=forward mode=slew speedmode=lowspeed "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[SCOPE] CheckMotorStatus() : Axis = 1 "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[MOUNT] SetSpeed() : Axis = 1 -- period=364 "
[2020-06-28T04:30:37.487 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":f1\", 0 bytes written "
[2020-06-28T04:30:37.501 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=111\", 5 bytes read "
[2020-06-28T04:30:37.501 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] dispatch_command: \":I16C0100\", 0 bytes written "
[2020-06-28T04:30:37.517 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=\", 2 bytes read "
[2020-06-28T04:30:37.556 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: event: "{\"Event\":\"Settling\",\"Timestamp\":1593311437.555,\"Host\":\"E5430\",\"Inst\":1,\"Distance\":6.70,\"Time\":0.0,\"SettleTime\":18.0,\"StarLocked\":true}\r\n"
[2020-06-28T04:30:39.065 CEST DEBG ][   org.kde.kstars.ekos.capture] - setMeridianFlipStage:  "MF_READY"
[2020-06-28T04:30:39.066 CEST INFO ][     org.kde.kstars.ekos.mount] - Meridian flip: slewing to RA= "20h 52m 27s" DEC= " 44° 06' 05\""  Hour Angle  "00h 00m 11s"
[2020-06-28T04:30:39.066 CEST INFO ][     org.kde.kstars.ekos.mount] - "Meridian flip started."
[2020-06-28T04:30:39.067 CEST DEBG ][   org.kde.kstars.ekos.capture] - meridianFlipStatusChanged:  "FLIP_RUNNING"
[2020-06-28T04:30:39.067 CEST DEBG ][   org.kde.kstars.ekos.capture] - setMeridianFlipStage:  "MF_INITIATED"
[2020-06-28T04:30:39.067 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: request: "{\"id\":779,\"jsonrpc\":\"2.0\",\"method\":\"stop_capture\"}"
[2020-06-28T04:30:39.072 CEST DEBG ][           org.kde.kstars.indi] - ISD:Telescope sending coords RA: "20h 52m 27s" ( 20.8744 ) DE: " 44° 06' 05\"" ( 44.1016 )
[2020-06-28T04:30:39.072 CEST INFO ][     org.kde.kstars.ekos.mount] - "Meridian flip slew started..."
[2020-06-28T04:30:39.075 CEST DEBG ][ org.kde.kstars.ekos.scheduler] - Capture State "Meridian Flip"
[2020-06-28T04:30:39.083 CEST DEBG ][     org.kde.kstars.ekos.guide] - PHD2: response: "{\"jsonrpc\":\"2.0\",\"result\":0,\"id\":779}\r\n"
[2020-06-28T04:30:39.084 CEST DEBG ][   org.kde.kstars.ekos.capture] - Guiding state changed from "Dithering" to "Aborted"
[2020-06-28T04:30:39.084 CEST INFO ][     org.kde.kstars.ekos.guide] - "Autoguiding aborted."
[2020-06-28T04:30:39.085 CEST DEBG ][ org.kde.kstars.ekos.scheduler] - Guide State "Aborted"

Basically, after image was captured, it was dithering. Then meridian flip occurred. After that dithering had to be stopped, but PHD2 event led to state changing to DITHERING_FAILED and then CAPTURE was abort.

After meridian flip was done, the scheduler decided that it cannot continue since the job is now close to twilight and decided to sleep and wait until the next day. It then initiated parking:
[2020-06-28T04:35:01.638 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[COMM] read_eqmod: \"=101\", 5 bytes read "
[2020-06-28T04:35:01.639 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[SCOPE] CheckMotorStatus() : Axis = 1 "
[2020-06-28T04:35:01.639 CEST DEBG ][           org.kde.kstars.indi] - EQMod Mount : "[DEBUG] IsRARunning() = true "
[2020-06-28T04:35:02.463 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'star' scheduled for execution at dom. jun. 28 23:27:00 2020. Parking the mount until the job is ready."
[2020-06-28T04:35:02.465 CEST DEBG ][           org.kde.kstars.indi] - ISD:Telescope: Parking... 
[2020-06-28T04:35:02.465 CEST INFO ][ org.kde.kstars.ekos.scheduler] - "Parking mount in progress..."
[2020-06-28T04:35:02.510 CEST INFO ][           org.kde.kstars.indi] - EQMod Mount :  "[INFO] Can not park while slewing... "
[2020-06-28T04:35:02.511 CEST DEBG ][           org.kde.kstars.indi] - Canon DSLR EOS 700D : "[DEBUG] Event timed out #2, retrying... "

The scheduler didn't handle the parking failure gracefully and went to sleep. So a couple of takeaways:

1. Handle dithering error while in meridian flip in Capture module. I believe this led to incorrect state machine. Should guide status affect anything during a meridian flip?
2. Handle parking failure in scheduler.
3 years 9 months ago #56169

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370

Replied by Wolfgang Reissenberger on topic meridian flip fails:eq6, latest git

OK, I'll check it.

Wolfgang
3 years 9 months ago #56171

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370

Replied by Wolfgang Reissenberger on topic meridian flip fails:eq6, latest git

We have indeed a significant weakness that a meridian flip might take place during the preparation tasks for a capture. The current implementation does not take care whether dithering or focusing are taking place when the mount signals that a meridian flip should take place. I think that is not that difficult to fix, since after the recent refactoring there is a single place to be corrected.
The following user(s) said Thank You: Jose Corazon
3 years 9 months ago #56176

Please Log in or Create an account to join the conversation.

  • Posts: 969
  • Thank you received: 94

Replied by alacant on topic meridian flip fails:eq6, latest git

Hi
Please let me know if there's anything needs testing. But maybe we can't test properly until the alignment is fixed:
indilib.org/forum/ekos/7256-align-fails-...ount-latest-git.html
Cheers and thanks for your time and interest.
Last edit: 3 years 9 months ago by alacant.
3 years 9 months ago #56189

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370

Replied by Wolfgang Reissenberger on topic meridian flip fails:eq6, latest git

I investigated the case further yesterday. As far as I can see meanwhile, Capture itself does not have a problem. It might happen that a meridian flip is started during dithering or focusing, but this will lead to failures of dithering or focusing - as we saw in your case with dithering. But the state of Capture remains correct.

The problem drills down meanwhile that the Scheduler issued a parking command while the mount was slewing for a meridian flip. This is indeed a problem. I am currently thinking about how to resolve this properly.

As a workaround: please be careful with termination conditions that are close to meridian flips when you expect proper parking afterwards. Better use a condition 30min beyond the meridian flip. That's far from elegant, I know...

Wolfgang
The following user(s) said Thank You: alacant
3 years 9 months ago #56200

Please Log in or Create an account to join the conversation.

  • Posts: 969
  • Thank you received: 94

Replied by alacant on topic meridian flip fails:eq6, latest git

Hi Wolfgang and thanks once again for your dedication.

The workaround: I'm currently imaging in Cygnus, catching astro darkness at around 23:30. and imaging until dawn when the scheduler shuts down at around 04:40. This coincides with the meridian flip and the telescope is almost vertical with the target being overhead. Even 5 minutes past the meridian the telescope is dangerously close to the mount. Is there a workaround for this set of conditions?
Cheers,
Steve
3 years 9 months ago #56205

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370

Replied by Wolfgang Reissenberger on topic meridian flip fails:eq6, latest git

Hi Steve,
the most critical point is that the mount executes the meridian flip, right? Why not ending the session significantly later than the meridian flip, e.g. at 5? You will get some additional frames that are worthless, but so what.

There are several ways to achieve this. You could change the dusk offset, you could set the job termination to a fixed value or you could set a fixed parking time on the mount tab. In the latter case do not forget to either push the "play" button or select the "Park at ... every day" box.
3 years 9 months ago #56228

Please Log in or Create an account to join the conversation.

  • Posts: 969
  • Thank you received: 94

Replied by alacant on topic meridian flip fails:eq6, latest git

Hi Wolfgang
OK. got it. I thought the scheduler was hard wired to shutdown at the end of astro darkness.
Cheers,
Steve
3 years 9 months ago #56229

Please Log in or Create an account to join the conversation.

  • Posts: 969
  • Thank you received: 94

Replied by alacant on topic meridian flip fails:eq6, latest git

Hi again. Unfortunately I'm not able to use the git ATM due to:
www.indilib.org/forum/ekos/7256-align-fa...ount-latest-git.html

Do you have a branch with your updates where the align module works?

TIA,
Steve
3 years 9 months ago #56230

Please Log in or Create an account to join the conversation.

  • Posts: 1187
  • Thank you received: 370

Replied by Wolfgang Reissenberger on topic meridian flip fails:eq6, latest git

This case is confusing...

What do we know:
  • At 04:30:39 the meridian flip limit is reached while dithering, the mount starts the meridian flip, dithering aborts immediately
  • a little bit later, dithering aborts
  • at 04:31:06 the following warning appears "Horizon Limits: Scope at AZ=0.458 ALT=-1.413 is outside limits. Nothing to abort."
  • at 04:35:00 the job reaches the twilight limit, aborts the job and issues a park command
  • at 04:35:02 EQMod issues a warning: "Can not park while slewing... "
From now on the mount remains in the state "slewing" until 07:07:07 another park command is sent. Again, the warning "Can not park while slewing... " appears.

It seems like the mount hit a mount limit while executing the meridian flip and got stuck there. Very strange. Does anybody have an idea?

Wolfgang
3 years 9 months ago #56288

Please Log in or Create an account to join the conversation.

Time to create page: 0.334 seconds