• Main INDEX
  • Monthly INDEX
  • PREV
  • NEXT

    User name stevensm

    Log entry time 13:21:49 on June17,2004

    Entry number 126077

    keyword=left septum J.T. valve problem

    Yesterday we responded to the loss of liquid helium in left Septum. We determined that the helium and nitrogen valves were not responding to any epics or plc control input, but would respond to local control. The plc program had become corrupted. We attempted to reprogram the plc, but it would not accept a new program without errors. We then changed the Septum plc cpu, and the new cpu accepted the program correctly, but would not allow us to configure the serial port through which EPICS communicates with Septum. We changed the cpu with two more spares, with identical results. Finally, just before 4:00 yesterday, Scot Spiegel figured out that the way we exited the software step of configuring the serial port can prevent the plc from accepting the reconfiguration. With that knowledge, we had a fully functioning plc at 4:00 P.M. yesterday.

    In the mean time, while we were troubleshooting the controls problems, we took local (manual) control of the helium and nitrogen valves, and proceeded to cool down and refill the left Septum. We slowly opened the supply valve to 85%, and watched the coil and yoke thermometers get colder. But what we did not realize was that the cold return valve had erroneously been left in remote control, and at some point the plc had nearly closed it, probably during one of the reprogram/reset evolutions. This resulted in left Septum not recovering helium level overnight, in spite of the J.T. valve being nearly wide open. I discovered the error this Thursday morning. I put the cold return valve in local control, and opened it wide open. The magnet filled with liquid helium in less than an hour.

    To prevent these problems from recurring, we are in the process of
    changing Heidi Fansler's "cheat sheet" of instructions for resetting and reprogramming the Septum plc into a formal checklist. This will verify the correct valve line ups and correct adjustable parameters in the Septa controls whenever the system is reset or reprogrammed. Because the Septa plc cpu seems to be susceptible to radiation induced failure, we will order additional plc cpu's to ensure adequate spares (at least 4 total).