Doing experiments at CMC will require a complete Spec installation, with the interface to Epics set up properly talking to all of the motors, with pseudomotor and macro code completed, and motor motions and macros tested. This memo documents some of the things done in July 1999 and some still to be done.
Spec is installed on the computer by the FOE (scooter.cmc.aps.anl.gov), but has not been configured. The computer is currrently being used to run FOE components through Epics.
Fourc version 4.04.04 has been installed on the office computer (cmc2006.cmc.aps.anl.gov), and configured for all real motors currently anticipated for FOE, SOE, and B hutch components, including the six-circle diffractometer. Here is a backup copy of the config file. The config file contains everything necessary to access real motors through Epics on the first three VME crates, which are:
The interface to 9ida: was tested, then the IP address removed from the access list to avoid accidents with the FOE components. Communication with 9idcs1: was established successfully. The 9idb: crate still needs to be built.
No counters or other non-VME hardware were defined in Spec. Surf (needed for the liquid surface spectrometer) is not installed on either machine.
The SOE VME crate has been built and booted, and one motor record configured in an attempt to spin one of the table leg motors. We did succeed in getting this motor to twitch, with Spec, through Epics! Now, the entire Epics database for all SOE and B hutch motors has to be entered, and each motor has to be tested to check the parameters. I'm sorry to say that at least one of the 16 25-pin driver cables we made has a short in it. I checked for opens and shorts to the shield, but they need to be checked for shorts between pins.
The SOE and B hutch tables, and six-circle diffractometer, are now in the hutches, with motors hooked up. The six-circle is wearing its four-phase motors at the moment.
Wherever Epics defines a pseudomotor that Spec needs to access, an additional Epics record must be made to allow Spec to access the pseudomotor. This task has hardly been started. Until it is done, Spec will have no control over white beam masks or slits. There may also be problems with the B hutch table and the monochromator control. After this has been worked out, these pseudomotor records have to be added to Spec's config file.
Macro files have been written to control the beamline slits, the SOE table and mirror, and the B hutch flight path. All of these macros have been tested cosmetically in a dummy setup of Spec,
and seem to be working. They need next to be tested with real motors. Macros for the B table have not been written. The B table has three horizontal translators which I would like to define as pseudomotors in Epics. I plan to provide the geometry for this, then Arun can enter the records.
The monochromator macros are not really started, I would like more info before making certain decisions about these. I made an attempt to find out how the Kohzu is used at SRI-cat, but haven't connected with the right people yet. Once this happens, the Escan-type Spec macros need to be checked out. I should probably talk more with Arun about the best way to change the energy: for example. by using epics_put on the energy variable? On the lambda variable? And so on.
It would be possible to go over this list and find out which ones depend on the others, and work out the best order for proceeding. I can help with this if requested. Meanwhile, this is a core dump.
Additions to this memo are welcome, just send them to me and in they go.
Elaine (dimasi@bnl.gov)