From slaughte@fnal.gov Wed Dec 31 12:44:09 2003 Date: Wed, 31 Dec 2003 12:43:18 -0600 (CST) From: Jean Slaughter Reply-To: slaughter@fnal.gov To: slaughter@fnal.gov Subject: database notes Notes from the "Magnets and Alignment Database" meeting of Dec. 19, 2003 It's been updated with some information from Dennis Box and John Greenwood. A. Action items - --------------------- 1. use case 1 - method for adding new magnet alignment data to BLAST (Beams Lattice Alignment Survey Tracker). Input will be from file with a format to be determined by Norm and John Greenwood. (See section B for draft 0, worked out in a post-meeting meeting). 2. Use case 2 - method to output a file suitable for input into Norm's program 3. Fill in the descriptions table in the database and disseminate.. a) Dennis Box will fill in what he already has from Norm b) Joel will get Norm's memo scanned and put in database c) Hire someone, possibly Kathy Lebrun to do the data entry? Joel to look into this 5. Make a first pass at what we are trying to achieve. Dave Harding, then Valerie. Action item response - post meeting form Dennis ------------------------------------ 1. My latest attempt at defining uses cases is at http://fndapl.fnal.gov/~dbox/php/documents/use_cases.html 2. The database browser I demo'ed is at http://fndapl.fnal.gov/~dbox/php/browser/blast.php . There were problems with the com (comments ) table that were due to my own mistakes, these are now corrected. 3. I have filled in as many of the column_description fields as I could manage, I will have to bug people for the rest. I created and populated an sgrade_codes table to define an important column that shows up in many tables. B. Post-meeting meeting notes (Dennis, John, Dave, Ray, Norm ( for first part), Jean -------------------------------------------------------------------------- 1. Draft file for input to BLAST 1. date 2. magnet location 3. magnet number 4. x,y,z in FSCS Z coordinates ( modernized DUSAF) 5. yaw, pitch, roll in FSCS Z coordinates 6. Twist defined as upstream roll - downstream roll upstream = direction the protons see first? 7. condition defined as either as found as set Note : FSCS Z is a flat world system, centered at A0 anbd tangent to A0. I know this is not quite enough...) 2. suggest a runii meeting talk by John Greenwood on coordinate systems and alignment 3. questions 1. Does Norm's program use the x,y,z? 2. Does anyone's tracking program use x,y,z, i.e. dispaced magnets relative to the central ray? 3. Do all beam physics codes use a central orbit coordinate system? 4. Interesting facts - which we hope we understood 1. The Main injector has an ideal orbit defined in effectively the FSCS system 2. The Tev does not. C. Comment by John Greenwood on beam sheets -------------------------------------------- Just to close the loop on last Friday's follow-on discussion about whether a formal beamsheet exists, or not, it is clear that a de facto beamsheet always exists. The issue is whether it represents a 'real' machine or an unacceptable approximation of one. For years we have been asked to make magnet moves relative to the present position of a component in the machine and/or wrt. the Murphy Line. That request constitutes a redefining of the machine's design orbit and, therefore, the issuance of a new 'beamsheet.' The problem the TeV has had over the years is that no one has been able to accurately catalog all the moves that have been made, so there is no machine design document. Once the TeVnet data is processed and the position of the machine and its components are documented, a statement of the current machine definition can be made - a beamsheet, if you will. As the requests to move various components are submitted, a new 'beamsheet' evolves. Whether the request says "move this dipole 50 mils right and set the roll at 0.1 milliradians" or it says "set the magnetic center at these coordinates with this roll, pitch, yaw set", doesn't change this at all. Our objectives for a database have always been to create an abstract machine definition that allows us to position all the components in 'local' machine space and know the relationship to various other machines, in both a spatial and temporal reference frame. This is most easily done by using a neutral, coherent, reproducible coordinate system (ECEF - Earth Centered, Earth Fixed), for which transformation parameters are developed to address 'local' needs.