FEM: frd import, better comment in the regard of different node order

This commit is contained in:
Bernd Hahnebach 2016-09-25 19:36:34 +01:00 committed by Yorik van Havre
parent 1cc1d9e562
commit 86895f2bdd

View File

@ -149,11 +149,13 @@ def readResult(frd_input):
nd19 = int(line[83:93])
nd20 = int(line[93:103])
input_continues = False
# print((nd1, nd2, nd3, nd4, nd5, nd6, nd7, nd8, nd9, nd10,
# nd11, nd12, nd13, nd14, nd15, nd16, nd17, nd18, nd19, nd20)) # get the element node order of frd file
# CalculiX uses a different node order in input file *.inp and result file *.frd for hexa20 (C3D20)
# according to Guido (the developer of ccx)
# ccx (and thus the *.inp) follows the ABAQUS convention (documented in the ccx-documentation)
# cgx (and thus the *.frd) follows the FAM2 convention (documented in the cgx-documentation)
# FAM32 is from the company FEGS limited, maybe this company does not exist any more)
# elements_hexa20[elem] = (nd6, nd7, nd8, nd5, nd2, nd3, nd4, nd1, nd14, nd15,
# nd16, nd13, nd10, nd11, nd12, nd9, nd18, nd19, nd20, nd17)
# node order exported by ccx into frd is different than node order in inp file
# elements_hexa20[elem] = (nd6, nd7, nd8, nd5, nd2, nd3, nd4, nd1, nd14, nd15,
# nd16, nd13, nd18, nd19, nd20, nd17, nd10, nd11, nd12, nd9)
# hexa20 import works with the following frd file node assignment
@ -183,8 +185,9 @@ def readResult(frd_input):
nd14 = int(line[33:43])
nd15 = int(line[43:53])
input_continues = False
#elements_penta15[elem] = (nd5, nd6, nd4, nd2, nd3, nd1, nd11, nd12, nd10, nd8,
# nd9, nd7, nd14, nd15, nd13) # as for inp - do not work
# CalculiX uses a different node order in input file *.inp and result file *.frd for penta15 (C3D15)
# elements_penta15[elem] = (nd5, nd6, nd4, nd2, nd3, nd1, nd11, nd12, nd10, nd8,
# nd9, nd7, nd14, nd15, nd13) # order of the *.inp file
elements_penta15[elem] = (nd5, nd6, nd4, nd2, nd3, nd1, nd14, nd15, nd13, nd8,
nd9, nd7, nd11, nd12, nd10)
elif elemType == 6: