Swope GUI Software Documentation - Troubleshooting

The Observatories of the Carnegie Institution for Science (Carnegie Observatories), Pasadena, CA


Location of this document: http://instrumentation.obs.carnegiescience.edu/Software/Swope/trouble.html


  • Common Error Codes
  • How does the Init procedure for the stages (X,Y,Focus) work?
  • Aligning the Focus Limit trigger
  • What are the trigger points of the ion-pump Auto mode?
  • Mac OS-X Arp Unicast workaround


    Common Error Codes

    Error Code Message Problem Recommendation
    7 ccdserver: busy The ccdserver is running an exposure and cannot take the requestion action. Try again; if the problem persists a reboot of the ccdXX computer may be necessary.
    12 ccdserver: too late The exposure is almost done and cannot be paused, nor the exposure time changed.
    17 ccdserver: line errors from DSP The ccdserver is receiving incomplete data lines from the DSP. Reset DSP/CCD-Server using the "Camera - CCD Hardhat" menu.
    1008 TCPIP: timeout on socket The device did not respond in to a request.
    1009 TCPIP: opening socket failed The device is not accepting connections. Check the network cabling (ObsPC and device). Power cycle the device.
    1010 TCPIP: internal locking error The TCP/IP connection is blocked by another software component. Check the network settings of the observer workstation. Check the router and cabling.
    2001 Galil: invalid command The Galil controller did not accept a command. For this to happen the controller has to be in a "confused" state. Please power-cycle the Galil controller and re-start the GUI.
    2007 Galil: position not found The software failed 3 times to move the mechanism to the target position.
    2009 Galil: motion aborted A home-sequence or other motor move has been stopped by the Galil-Hardhat.
    3001 CCDIO: ccdserver busy Connection to ccdserver is busy by another task. This should be a temporary issue, please try again.
    3005 CCDIO: DSP error The DSP response was inconsistent with the command issued by the ccdserver/Swope-GUI Try again; if the error persists you may have to use the Reset DSp/CCD-Server button on the CCD-Hardhat window and/or power-cycle the DSP.
    3007 CCDIO: no data from ccdserver The GUI did not receive (enough) data from the ccdserver. Try again; if the error persists you may have to use the Reset DSp/CCD-Server button on the CCD-Hardhat window.
    Note: This may be caused by intermittend network problems and is currenly under investigation.
    4001 exposure timeout The ccdserver is in exposure mode even though the GUI expects the exposure to be done. Take a short test exposure; if the problem persists, reboot the ccdXX computer and re-start the Swope-GUI.


    How does the Init procedure for the stages (X,Y,Focus) work?

    1. Drives the mechanism into the reverse limit.
    2. Slowly backs out of the limit. When the limit sensor goes OFF, it stops the motor and zeroes the encoder.


    The focus limit trigger "wheel" becomes miss-aligned.
    This may be caused by a loose set screw.

    The "wheel" set screw must be tightend at the proper position of the "wheel" because it defines the home position and protects the guider camera from running into the focus stage or IMB enclosure.
    Please follow the instructions on the
    HowTo page.


    Trigger points of the ion-pump Auto mode

    The Auto mode turns the ion-pump ON if
  • pressure below 1.2e-4 (from the GP375 vaccum gauge) and
  • CCD temperature (2nd highest) below -100 [C]
  • The Auto mode turns the ion-pump OFF if
  • pressure above 1.8e-4 (from the GP375 vaccum gauge or ion-pump sensor) or
  • CCD temperature (2nd highest) above -95 [C]

  • Mac OS-X Arp Unicast workaround

    In ... OS X Mavericks it appears that Apple has implemented section 2.3.2.1 of RFC1122, "ARP Cache Validation." In this scenario it appears that Mavericks installations are performing unicast ARP requests and timing out ARP for the gateway on the host if it does not receive a corresponding response - thus assuming its ARP entry is no longer valid. Based on preliminary evidence after 5 unsuccessful attempts at unicast ARP OSX reverts to standard broadcast ARP and is successful. This is causing a symptom that resembles short period of unresponsiveness or packet loss.

    sudo sysctl -w net.link.ether.inet.arp_unicast_lim=0

    /etc/sysctl.conf:
    net.link.ether.inet.arp_unicast_lim=0


    2014-06-13, Christoph C. Birk, Carnegie Observatories