Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-23518

ATDome is mixing in position

    Details

    • Type: Story
    • Status: Done
    • Resolution: Done
    • Fix Version/s: None
    • Component/s: None
    • Labels:
    • Story Points:
      0
    • Sprint:
      TSSW Sprint - Feb 17 - Mar 2, TSSW Sprint - Mar 2 - Mar 16
    • Team:
      Telescope and Site
    • Urgent?:
      No

      Description

      It seems like ATDome is mixing in position with not in position. 

       

       

        Attachments

          Activity

          Hide
          rowen Russell Owen added a comment -

          For the ATDome CSC to report azimuth as "in position": azimuth must not be moving and must be within 0.2 degrees of its commanded position. The tolerance is specified by attribute az_tolerance and can easily be changed if need be.

          To determine if an axis is moving the CSC uses line 5 of status. Here is sample output:
          RR 000
          The second item is a bitmask with the following bits. All of these except HOME and ESTOP indicate whether the axis is moving in the direction specified:
          AZPOSITIVE = 1
          AZNEGATIVE = 2
          MAINDOORCLOSING = 4
          MAINDOOROPENING = 8
          DROPOUTDOORCLOSING = 16
          DROPOUTDOOROPENING = 32
          HOMING = 64
          ESTOP = 128

          Azimuth is stopped if none of AZPOSITIVE, AZNEGATIVE or HOMING are set.

          Show
          rowen Russell Owen added a comment - For the ATDome CSC to report azimuth as "in position": azimuth must not be moving and must be within 0.2 degrees of its commanded position. The tolerance is specified by attribute az_tolerance and can easily be changed if need be. To determine if an axis is moving the CSC uses line 5 of status. Here is sample output: RR 000 The second item is a bitmask with the following bits. All of these except HOME and ESTOP indicate whether the axis is moving in the direction specified: AZPOSITIVE = 1 AZNEGATIVE = 2 MAINDOORCLOSING = 4 MAINDOOROPENING = 8 DROPOUTDOORCLOSING = 16 DROPOUTDOOROPENING = 32 HOMING = 64 ESTOP = 128 Azimuth is stopped if none of AZPOSITIVE, AZNEGATIVE or HOMING are set.
          Hide
          rowen Russell Owen added a comment -

          Tiago figured out that the low level controller has a larger azimuth slop than the CSC was allowing for. Fixed by setting the CSC az_tolerance slightly larger than the "tolerance" reported by the low-level controller.

          Pull request: https://github.com/lsst-ts/ts_ATDome/pull/21

          Show
          rowen Russell Owen added a comment - Tiago figured out that the low level controller has a larger azimuth slop than the CSC was allowing for. Fixed by setting the CSC az_tolerance slightly larger than the "tolerance" reported by the low-level controller. Pull request: https://github.com/lsst-ts/ts_ATDome/pull/21
          Hide
          rowen Russell Owen added a comment -

          Tiago Ribeiro can you please test that this pull request fixes the problem?

          Show
          rowen Russell Owen added a comment - Tiago Ribeiro can you please test that this pull request fixes the problem?
          Hide
          tribeiro Tiago Ribeiro added a comment -

          I'll have to wait until next week to verify this with the real hardware. There's work being done at the atdome right now that prevents us from moving it. 

           

          Show
          tribeiro Tiago Ribeiro added a comment - I'll have to wait until next week to verify this with the real hardware. There's work being done at the atdome right now that prevents us from moving it.   
          Hide
          tribeiro Tiago Ribeiro added a comment -

          Review done in GitHub. After testing the code at the telescope we figure it is a deeper problem than we thought. We will need to implement a fix in the lower level controller and possibly a workaround in the CSC while we work on how to do that. 

          Show
          tribeiro Tiago Ribeiro added a comment - Review done in GitHub. After testing the code at the telescope we figure it is a deeper problem than we thought. We will need to implement a fix in the lower level controller and possibly a workaround in the CSC while we work on how to do that. 
          Hide
          rowen Russell Owen added a comment - - edited

          To expand on what Tiago just said: the CSC was being too picky when testing for "in position", but that fix is not good enough for reliable "in position" reports because the vendor's code is mis-behaving. When it stops it often stops outside its own tolerance and just stays there. Another move is required to fix that.

          Clearly this should be fixed in the vendor's code, but we can probably cobble together a workaround in the CSC.

          Merged to develop. I will hold off releasing until Tiago has some time to work on conda files, which he would like to get into this release. We may want to consider renaming the release 1.2.0 instead of 1.1.1 since it will then be more than bug fixes.

          Show
          rowen Russell Owen added a comment - - edited To expand on what Tiago just said: the CSC was being too picky when testing for "in position", but that fix is not good enough for reliable "in position" reports because the vendor's code is mis-behaving. When it stops it often stops outside its own tolerance and just stays there. Another move is required to fix that. Clearly this should be fixed in the vendor's code, but we can probably cobble together a workaround in the CSC. Merged to develop. I will hold off releasing until Tiago has some time to work on conda files, which he would like to get into this release. We may want to consider renaming the release 1.2.0 instead of 1.1.1 since it will then be more than bug fixes.

            People

            • Assignee:
              rowen Russell Owen
              Reporter:
              tribeiro Tiago Ribeiro
              Reviewers:
              Tiago Ribeiro
              Watchers:
              Russell Owen, Tiago Ribeiro
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: