Highlighted

CGAP / PGAP behaviour different if one node has a different Output Coordinate system (CSYS)

Phenom
Phenom

I believe I have found an "undocumented feature" of NX Nastran when doing a test model of a preloaded gap (for the purpose of pretensioning a beam "cable").  The analysis works perfectly as expected when all the nodes use the Basic Coordinate System (COORD 0) as the Output CSYS.  However, if the 2nd node of the GAP uses a different Output CSYS, then the analysis fails to converge - for no good reason.  The (failing) simple test model is below.  If GRID 2 is changed so its last entry is 0 instead of 3 (change Output CSYS to Basic Coord), then the analysis works as expected and correctly pre-tensions the "cable".

The workaround is to use COORD 0, but unfortunately when modelling typical cable-stayed structures it is often useful (necessary) to constrain cable ends in non-global coords.

Cheers,

 

SOL NLSTATIC
GEOMCHECK, NONE
CEND
TITLE = NX Nastran NL Static Analysis Set
ECHO = NONE
DISPLACEMENT(PLOT) = ALL
SPCFORCE(PLOT) = ALL
FORCE(PLOT,CORNER) = ALL
NLSTRESS(PLOT) = ALL
STRESS(PLOT,CORNER) = ALL
SPC = 1
NLPARM = 1
LOAD = 1
BEGIN BULK
$ ***************************************************************************
$ Written by : Femap with NX Nastran
$ Version : 11.4.1
$ Translator : NX Nastran
$ From Model : I:\gap_preload.modfem
$ Date : Thu Jul 26 17:41:16 2018
$ Output To : I:\
$ ***************************************************************************
$
NLPARM 1 100 5 25 PW NO+
+ .001 .001 1.-7 3 25 4 .2 .5+
+ 5 20. 20.
PARAM,PRGPST,NO
PARAM,LGDISP,1
PARAM,POST,-1
PARAM,OGEOM,NO
PARAM,AUTOSPC,YES
PARAM,K6ROT,100.
PARAM,GRDPNT,0
CORD2C 1 0 0. 0. 0. 0. 0. 1.+FEMAPC1
+FEMAPC1 1. 0. 1.
CORD2S 2 0 0. 0. 0. 0. 0. 1.+FEMAPC2
+FEMAPC2 1. 0. 1.
$ Femap with NX Nastran Coordinate System 3 : xzcv
CORD2R 3 0 1020. 0. 0.382.8111 0.796.4862+
+ 1557.587752.6221430.0698
$ Femap with NX Nastran Load Set 1 : Gap preload
FORCE 1 2 0 1. 1. 0. 0.
$ Femap with NX Nastran Constraint Set 1 : Fixture
SPC1 1 123456 1
SPC1 1 123456 3
$ Femap with NX Nastran Property 1 : Bar
$ Femap with NX Nastran PropShape 1 : 5,0,5.,0.,0.,0.,0.,0.
$ Femap with NX Nastran PropOrient 1 : 5,0,1.,1.,2.,3.,4.,-1.,0.,0.
PBEAM 1 178.53982490.8739490.8739 0.980.9212 0.+
+ -5. 0. 0. -5. 5. 0. 0. 5.+
+ YESA 1. +
+ .8861774.8861771
$ Femap with NX Nastran Property 2 : Gap for preload
PGAP 2 0. 8000. 10000. 1. 0. 0. 0.+
+ -1.
$ Femap with NX Nastran Material 1 : Steel
MAT1 1 200000. .3 0. 0. 0.
GRID 1 0-100000. 0. 0. 0
GRID 2 0 1020. 0. 0. 3
GRID 3 0 1010. 0. 0. 0
GRID 53 0 -89898. 0. 0. 0
GRID 54 0 -79796. 0. 0. 0
GRID 55 0 -69694. 0. 0. 0
GRID 56 0 -59592. 0. 0. 0
GRID 57 0 -49490. 0. 0. 0
GRID 58 0 -39388. 0. 0. 0
GRID 59 0 -29286. 0. 0. 0
GRID 60 0 -19184. 0. 0. 0
GRID 61 0 -9082. 0. 0. 0
CGAP 2 2 3 2 0. 1. 0.
CBEAM 52 1 1 53 0. 1. 0.
CBEAM 53 1 53 54 0. 1. 0.
CBEAM 54 1 54 55 0. 1. 0.
CBEAM 55 1 55 56 0. 1. 0.
CBEAM 56 1 56 57 0. 1. 0.
CBEAM 57 1 57 58 0. 1. 0.
CBEAM 58 1 58 59 0. 1. 0.
CBEAM 59 1 59 60 0. 1. 0.
CBEAM 60 1 60 61 0. 1. 0.
CBEAM 61 1 61 2 0. 1. 0.
ENDDATA

2 REPLIES 2

Re: CGAP / PGAP behaviour different if one node has a different Output Coordinate system (CSYS)

Phenom
Phenom

why would you want the nodes in a different Output CSYS?

Production: NX.CAE 10.0.2.6
Development: VB.NET (amateur level !)

Re: CGAP / PGAP behaviour different if one node has a different Output Coordinate system (CSYS)

Phenom
Phenom

As mentioned in the original post, this is for a cable-stayed structure, where the cables are typically never oriented in  basic rectangular directions.  For preloading cables, the gap must be "stabilised" so that the preload is definitely acting only in the cable direction - ie constrained to allow movement ONLY in the cable direction.  If COORD 0 were a suitable workaround, I wouldn't have posted this - but the problem is effectively a bug/limitation, as the documentation does not indicate that the gap behaviour should be dependent on the Output CSYS of the gap's 2nd node.