21/12/2012 Committed comi code to Launchpad Repository via Bazaar. Revision 16. (Model 42)
20/12/2012 There are now switches in the meshing resolution constants that permit the material for the panels on the compressors/racks to be turned to air. This allows the simulation of the removal of a panel. The code that generates the racks has had to be rewritten to permit this functionality. It should be possilbe to extend this code to permit 'holes' to be put into the rack skin.
19/12/2012 The code for Holger's Step IV return yoke now has a switch to allow the new 50cm flanges to be added to the model. Although the flanges protrude past the component air they still appear to mesh ok. (The volume taken up by the flanges are removed from the Hall air as a precaution.)
18/12/2012 I have now got the location of the centre of ISIS wrt to the MICE Hall so I've been able to fix the East wall. It clearly has moved the wall slightly into the hall but this shouldn't have had any impact on any of the models studied so far. Also corrected a small bug on the East end of the trench (wrong material definition) but of no consequence to the model.
18/12/2012 To try and keep track of what models have been run with the compressor/NSW rack combination I have produced this table. I will fill this in as we run the models. A 'x' means the model won't be run, a '?' means that it is my intention to run these models. 'TBC' means the model is running but is yet 'To be Completed'.
13/12/2012 Committed comi code to Launchpad Repository via Bazaar. Revision 15 (Model 39)
12/12/2012 Committed comi code to Launchpad Repository via Bazaar. Revision 14 (Model 38)
12/12/2012 Closer inspection of Holger's return yoke has revealed a few discrepancies between his and my model. There is a geometry difference in the way that we have interfaced the shield to the Virostek Plates. Holger is now running a 12cm shield and this is modelled as a 10cm shield. I will make these corrections and re-run these models at a later date. Holger also has a concern about the on-axis field, he thinks it looks a little high. This needs investigating but my suspicion is that this is due to these models using the TRD magnet design at 240MeV/c - Note: This was later confirmed by Mike.
07/12/2012 Committed comi code to Launchpad Repository via Bazaar. Revision 13.
07/12/2012 I have noted that the step IV magnets and shield are centered at -2750mm but the Virostek plates are centrered around -2790mm. This 40mm discrepancy is because the magnets files are based upon the original TRD design, but the magnet dimensions have changed slightly since then. This means that the Step IV channel is not quite central wrt the Virostek plates.
06/12/2012 I've now removed the links to all of the test files prior to model 29. Some of the .op3s are still in the .op3 directory. If you want access to other files them please email me.
06/12/2012 The model solutions have been split so that the op3 file is in a different directory to the other files. The .op3 file is particularly large and backing them all up daily creates a significant overhead on the tape robot. Therefore the .op3 files are no longer backed up. However the .opc files are still backed up and these can regenerate the .op3 in the case of a disaster.