MD Nastran/MSC.Nastran Error List Update February 01, 2007 1-12864001 2007.0.0 Virtual Mass - vmopt=2 and resvec=no - Wrong Answers When vmopt=2 and resvec=no (or param,resvec,no) are both selected, the run will produce wrong answers. Avoidance: Do not deactivate residual vectors. 1-19126815 Optimization/Design Sensitivity - DRESP2/DCONSTR - Poor Message, sfm 3007 (dom10) When a DRESP2 entry is referenced by more than one DCONSTR entry (which is not allowed) the following message is issued: *** SYSTEM FATAL MESSAGE 3007 (DOM10) ILLEGAL INPUT TO SUBROUTINE DOM10G This message should be improved to be like the message below issued when this user error is encountered with the DRESP1 entry. *** USER FATAL MESSAGE 7143 (DOPR3H) DRESP1 ENTRY ID = # IS REFERENCED BY MORE THAN ONE DCONSTR ENTRY IN SUBCASE = #. 1-20216437 2007.0.0 Restarts - SFM 478 (GPFLIX) DATA BASE, HAS A NULL TIME STAMP Client ran restart job on Intel Linux and obtained a message indicating that the attached master file had a null timestamp. *** SYSTEM FATAL MESSAGE 478 (GPFILX) THE FOLLOWING LOGICAL NAME, PART OF THE LOCATED DATA BASE, HAS A NULL TIME STAMP. USER INFORMATION: LOGICAL NAME = SE1 ---------------------- Assign master='modes-r1.master' should have been Assign master='modes-r1.MASTER' (on linux you must specify the correct uppercase and lowercase spelling. The above SFM 478 is confusing and should indicate that the assigned file could not be found and to look toward possible misspelling of the file name and/or upper/lower case mismatch. 1-20275134 2007.0.0 Case Control - MAXMIN - User input errors are not trapped When the following incorrect case control commands are used: maxmin(def) disp maxmin(def) stress quad4 maxmin=all Nastran will fail without issuing a meaningful message. 1-20798063 Acoustics - IDENT, CP - UWM 6152, 6209 The IDENT, CP, and BW are 3 distinct fluid-structure interface methods. When IDENT is used, the METHOD field (CP or BW) should be ignored. However, if CP is entered on the METHOD field when the IDENT option is used, a default NORMAL of -.001 is passed to the IDENT method. This will cause the job to fail with UWM 6152 and 6209 stating that acoustic coupling matrix is not created for this model. Avoidance: Explicitly enter .001 for the NORMAL. 1-20891501 2007.0.0 DMAP - SOLVE - Wrong Answers The SOLVE module "invert" option (SYM=3) is broken. It only works for small problems; if FBS has to do multiple passes, it will produce wrong answers for the last pass. Avoidance: Generate an identity matrix, and use the identity matrix as the second input (rhs matrix). E.g. paraml kaa//'trailer'/2/s,n,NCOL $ matgen ,/ident/1/NCOL $ solve KAA,ident,,,/KAAINV1/ $ 1-21368688 Documentation/Installation - Mixed Masterkey and Concurrent Licenses Beginning with version v2005r2, both MSC.Nastran and MSC.Patran incorporate the new LAPI layer (v5.4.0) which introduces an additional environment variable, MSC_LICENSE_DISCRETE. It is necessary to set this variable when mixing Masterkey with concurrent license servers. This variable may be used to manipulate the default license path behavior, essentially instructing LAPI to reset the license search type between each server. The environment variable is necessary so that by default queuing is not inadvertently affected in homogeneous license installations. Subsequent releases of other MSC products, once they include the LAPI update, will have MSC_LICENSE_DISCRETE available to address the limitation as well. 1-22403461 2007.0.0 Aeroelastic Analysis - SWM 6065, DBCMSG Error Message 51, Subroutine DBCUGM In MSC.Nastran Version 2005.1, SOL 145 Flutter analysis fails to write to the xdb file and generates the following message: *** SYSTEM WARNING MESSAGE 6065 (DBC) *** DBCMSG ERROR MESSAGE 51 FROM SUBROUTINE DBCUGM MATRIX COLUMN CONSISTENCY CHECK FALUIRE; MATRIX 144 TABLE 210. This prevents Patran users from plotting interpolated mode shapes on the aerodynamic model. The same input file runs correctly In version 2005.0.0. 1-22614794 2007.0.0 Dynamics - PARAM,AUTOMSET and AUTOQSET - UFM 2101 and SFM 6143 Input decks that combine AUTOQSET and AUTOMSET may fail with *** USER FATAL MESSAGE 2101 (GP4) GRID POINT 1 COMPONENT 1 ILLEGALLY DEFINED IN SETS UM UB ... GRID POINT 3 COMPONENT 6 ILLEGALLY DEFINED IN SETS UM UB *** SYSTEM FATAL MESSAGE 6143 (UMRGE1) THE SIZES OF THE INPUT MATRICES AND THE DOF SETS ARE INCOMPATIBLE. SPECIFICALLY: The number of columns in KNN is not equal to the size of the COMP-set The number of rows in KNN is not equal to the size of the COMP-set User Information: Size of input matrices and DOF sets: KNN : Rows= 17136 by Cols= 17136 Set sizes: Major (N-set)= 17196 Set0 (COMP-set)= 17148 Set1 (Q-set)= 48 Avoidance: The fatal exit can be avoided by either replacing AUTOMSET by UM feature or replacing AUTOQSET by SPOINT, QSET1 or by applying the following alter: compile findm alter 'call zmce1' $ type parm,nddl,i,n,zuzri1 $ call dbstore zgpart,,,,//0/0/' '/0 $ zuzri1=newmsize $ compile augqset alter 'gp4 ','' TYPE PARM,NDDL,CHAR8,Y,automset $ GP4 CASE,GEOM4,eqexin,SIL,GPDT,BGPDT,CSTM,,,,,,,,/ RMG,Yg1,USET0,/ LUSET/S,N,NOMSET/0/S,N,NOSSET/0/S,N,NORSET/ nskip/0/0/0/0/SEID//sebulk/automset $ alter 'PARAML USET1'(,-1) call dbfetch /zgpart,,,,/0/0/0/0/0 $ umerge uset1,zgpart,/gparq/'g'//'q' $ type parm,nddl,i,n,zuzri1 $ MODUSET GPARq,USET1/tUSET/'NEWMSET'////zuzri1 $ equivx tuset/uset1/-1 $ ENDALTER 1-23241691 Superelements/Buckling - EXTSE - SFM 6551 (INPUTT2), during upstream buckling data recovery Beginning in Version 2005r3, a SOL 105 job with part external superelements may fail with the message below during buckling eigenvector data recovery for upstream superelements. SYSTEM FATAL MESSAGE 6551 (IINPUTT2) module in subDMAP EXTOTMIN 1-23504486 Documentation/QRG - PBEAML entry for DBOX error The figure for PBEAML type DBOX in the QRG has a typo. What is shown as DIM12 should be DIM2. 1-24063037 2007.0.0 Acoustic Analysis - STATSUB, MAT10 - Wrong Answers A model with a static subcase and fluid structure interaction fails to include the fluid effects and/or the preload effects. Avoidance: In Version 2005.1.0 apply the following alter: compile phase1b alter 'purgex /aa,aay,aaa,asf,','' dbstatus aaa//s,n,noaaa $ if ( noaaa<1 ) purgex /aa,aay,aaa,asf,/-1 $ Version 2004.0.6 compile phase0 alter 'FSCOUP=FALSE','' 1-24087243 2007.0.0 MDACMS - May be inefficient for enforced motion in a fluid-structure model The combined use of MDACMS and enforced motion in a fluid-structure model may consume large resources unnecessarily. The problem arises when there are no applied loads and no data recovery requested on any structural grid points. One indication of the problem is the large "A-set" compute requirement for MDACMS eigenvector recovery. Avoidance: There are two avoidances: 1. Apply a dummy load to, or request data recovery for, at least one structural grid point. OR 2. A DMAP alter avoidance is being tested. 1-24229901 2007.0.0 Iterative Solver - SMETHOD - Fails in statics with traceback If the input matrix contains null columns, the BIC iterative solver could fail with a meaningless traceback rather than producing an informative message. Avoidance: Ensure that the model is properly constrained using AUTOSPC or direct application of SPC. 1-24239871 2007.0.0 Plotting - Missing Title, Subtitle, and Label from VG xyplot The TITLE, SUBTI and LABEL information is missing from the flutter VG xyplots for version 2005.0 and 2005.1. TCURVE information appears in the xyplot output. Version 2004 and earlier releases are correct. Avoidance: A dmap alter is available on request. 1-24246501 2007.0.0 Elements - P-Element - SFM 3011 in VECPLOT P-Element jobs may encounter SYSTEM FATAL MESSAGE 3011 in VECPLOT. 1-24268315 2007.0.0 Parallel - SOL 101, AUTOMSET - SFM 3011 in FBS In SOL 101 and dmp>1 with PARAM,AUTOMSET,YES the run will terminate with: SFM 3011 (FBS) ATTEMPT TO WRITE A TRAILER ON FILE 201 WHEN IT HAS BEEN PURGED Avoidance: Insert the following DMAP alter: compile phase1dr alter 'call disph1b','' call disPH1b, CASES ,EDT ,EQEXINS ,GEOM4S ,GPLS , KGG ,PG ,RMG ,SILS ,USET0 , YG ,mgg ,spcpart ,cstms ,bgpdts , rxg ,yxg / GM ,USET ,KSF ,KFS ,KSS , dkff ,kmg ,dpf ,pss ,ys , mmg ,msf ,dmff ,rxa ,yxa , ym / error /nskip /nloads /semr $ compile disph1b alter 1,1 SUBDMAP disPH1b, CASES ,EDT ,EQEXINS ,GEOM4S ,GPLS , KGG ,PG ,RMG ,SILS ,USET0 , YG ,mgg ,spcpart ,cstms ,bgpdts , rxg ,yxg / GM ,USET ,KSF ,KFS ,KSS , kff ,kmg ,pf ,pss ,ys , mmg ,msf ,mff ,rxa ,yxa , ym / error /nskip /nloads /semr $ type db pym $ alter 'call sekr0','' CALL SEKR0 KGG,USET0,RMG,GEOM4S,EQEXINS,SILS,GPLS,spcpart, ,,cstms,bgpdts,cases,,yg,rxg,yxg/ GM,USET,knn,kmg,rxa,yxa,pym,ym/ ERROR/SEID/AUTOSPC/0/false/true/NSKIP/true $ alter 'endif \$ ncul=nloads or' paraml pym//'null'////s,n,nullpym $ if ( nullpym>-1 ) then $ add5 pg1,pym,,,/pgy//-1. $ equivx pgy/pg1/-1 $ endif $ nullpym>-1 ...from Mike Reymond. PJS 23feb SOL 101 DMP = 4 PARAM,AUTOMSET,YES approx 2 million dof's Using v2005.1, and v2005.5, job fails with *** SYSTEM FATAL MESSAGE 3011 (FBS) ATTEMPT TO WRITE A TRAILER ON FILE 201 WHEN IT HAS BEEN PURGED f04 shows it failed in the: MCE1S 12 FBS BEGN Using v2005.0, job runs successfully. 1-24400047 2007.0.0 MDACMS - EPOINT and TF - UFM 1126 MDACMS will terminate with UFM 1126 on data block XHSE if the model contains EPOINTs and transfer functions (TFL) Avoidance: Insert the following DMAP alter: compile get2pp subdmap get2pp cases,dits,usetd,rotor,rotmat,matpool,eqdyn, tfpool,bgpdts,cstms,kpp,bpp,mpp/ k2pp,b2pp,m2pp/ app/noue/lusetd/soltyp/nlheat/fs $ type parm,,i,n, nok2pp=-1,nob2pp=-1,nom2pp=-1,lusetd,noue $ type parm,,logi,n,freqdepr,acsyml,abfl,nlheat,fs $ type parm,nddl,char8,y,acsym $ type parm,,char8,n,app,soltyp $ TYPE PARM,,cS,n,cmu=(-1.,0.),cu=(1.,0.),cz $ PARAML cases//'DTI'/1/446//S,N,IRGYRO $ $ If no Campbell, do gyro terms here otherwise in sedceig PARAML cases//'DTI'/-1/495//S,N,ICAMP $ IF ( IRGYRO>0 and not(icamp=-1 and app='ceigen' and soltyp='direct') and (app='freqresp' or app='ceigen' or app='tranresp') ) then $ CALL GENGYRO cases,dits,USETD,ROTOR,ROTMAT/ MGPP,BGPP,KGPP,VGFDr/ APP/s,freqdepr/noue $ endif $ IRGYRO>0 call xmtrxin cases,matpool,eqdyn,tfpool,/ k2pp,m2pp,b2pp,,,/ 'x2pp '/lusetd/s,nok2pp/s,nom2pp/s,nob2pp/0/0/0 $ $ $ $ Add Matrices of Acoustic Infinite Elements $ DBSTATUS KPP,MPP,BPP//S,N,NOKPP/S,N,NOMPP/S,N,NOBPP $ IF ( NOKPP=1 ) THEN $ IF ( NOK2PP>-1 ) THEN $ CALL XADD5 KPP,K2PP,,,/K2PPX/CU/CU/CZ/CZ/CZ $ EQUIVX K2PPX/K2PP/ALWAYS $ ELSE $ EQUIVX KPP/K2PP/ALWAYS $ NOK2PP = 1 $ ENDIF $ ENDIF $ IF ( NOBPP=1 ) THEN $ IF ( NOB2PP>-1 ) THEN $ CALL XADD5 BPP,B2PP,,,/B2PPX/CU/CU/CZ/CZ/CZ $ EQUIVX B2PPX/B2PP/ALWAYS $ ELSE $ EQUIVX BPP/B2PP/ALWAYS $ NOB2PP = 1 $ ENDIF $ ENDIF $ IF ( NOMPP=1 ) THEN $ IF ( NOM2PP>-1 ) THEN $ CALL XADD5 MPP,M2PP,,,/M2PPX/CU/CU/CZ/CZ/CZ $ EQUIVX M2PPX/M2PP/ALWAYS $ ELSE $ EQUIVX MPP/M2PP/ALWAYS $ NOM2PP = 1 $ ENDIF $ ENDIF $ $ IF ( IRGYRO>0 and icamp<>-1 ) then $ PARAML MGPP//'NULL'////S,N,NOMGPP $ IF ( NOMGPP<>-1 ) THEN $ call xADD5 MGPP,M2PP,,,/M2PPX/cu/cu/cz/cz/cz $ EQUIVX M2PPX/M2PP/-1 $ NOM2PP= 1 $ ENDIF $ PARAML BGPP//'NULL'////S,N,NOBGPP $ IF (NOBGPP <> -1) THEN $ call xADD5 BGPP,B2PP,,,/B2PPX/cu/cu/cz/cz/cz $ EQUIVX B2PPX/B2PP/-1 $ NOB2PP= 1 $ ENDIF $ PARAML KGPP//'NULL'////S,N,NOKGPP $ IF (NOKGPP <> -1) THEN $ call xADD5 KGPP,K2PP,,,/K2PPX/cu/cu/cz/cz/cz $ EQUIVX K2PPX/K2PP/-1 $ NOK2PP= 1 $ ENDIF $ endIF $ IRGYRO>0 $ acsyml=(app='freqresp' and acsym='yes' and fs) $ IF ( SOLTYP='DIRECT ' AND NOT(NLHEAT) ) THEN $ $ PARAML MATPOOL//'PRESENCE'////S,N,NOMPOOL $ IF ( NOMPOOL>=0 ) THEN $ BMG MATPOOL,BGPDTS,CSTMS/ BDPOOL/ S,N,NOKBFL/S,N,NOABFL/S,N,MFACT $ ABFL = NOTL(NOABFL) $ IF ( ABFL OR NOTL(NOKBFL) ) THEN $ NOK2PP=1 $ MTRXIN, ,BDPOOL,EQDYN,,/ ABFL,KBFL,,,/ LUSETD/NOABFL/NOKBFL/0 $ call xADD5 ABFL,KBFL,K2PP,,/K2DPP/cmu/cu/cu/cz/cz $ EQUIVX K2DPP/K2PP/ALWAYS $ IF ( ABFL ) THEN $ NOM2PP=1 $ TRNSP ABFL/ABFLT $ ADD ABFLT,M2PP/M2DPP/MFACT $ EQUIVX M2DPP/M2PP/ALWAYS $ ENDIF $ ABFL ENDIF $ FL ENDIF $ NOMPOOL>=0 ENDIF $ SOLTYP=DIRECT AND NOT(NLHEAT) $ return $ end $ get2pp compile semfreq alter 'if ( nokpp=-1',''(,6) alter 'call fdrmgen'(,-1) call get2pp casem,dits,usetd,rotor,rotmat,matpools,eqdyn, tfpool,bgpdts,cstms,kpp,bpp,mpp/ k2pp,b2pp,m2pp/ app/noue/lusetd/soltyp/false/fs MATMOD K2pp,b2pp,m2pp,,,/Vpr,xx/25/S,N,NOvpr/3 $ upartn usetd,vpr/vpr1,,,/'p'/'g'//1 $ MATMOD vpr1,,,,,/vpr2,/47 $ diagonal vpr2/vpaie/'whole'/0.0 $ compile semceig alter 'call modefsrs','' call get2pp casem,dits,usetd,rotor,rotmat,matpools,eqdyn, tfpool,bgpdts,cstms,,,/ k2pp,b2pp,m2pp/ app/noue/lusetd/soltyp/false/fs Matmod k2pp,b2pp,m2pp,,,/vy1,/12/s,n,nonull/3/-1 $ upartn usetd,vy1/vy2,,,/'p'/'g'//1 $ diagonal vy2/v01p/'whole'/0.0 $ type parm,nddl,logi,n,seindr $ seindr=true $ IF ( DOPH2M ) CALL MODEFSRS, MR ,USET ,Dar ,casem ,DYNAMICS, MAA ,mKAA ,GPLS ,SILS ,EED , EQEXINS ,VAFS ,pg ,gm ,got , ksf ,yg ,lll ,kll ,drg , bgpdts ,cstms ,vgq ,goq , , bjj ,rvg ,cee ,mee ,gge , bec ,mec ,bsf ,msf ,pmyg , pbyg ,pkyg ,pvt0 ,mgg ,vgfs , , , , , , , , , , , ,v01p , , , / PHA ,LAMA ,PHFA ,LAMAF ,PHSA , LAMAS ,mechg ,lammats ,lammatf ,mis , mif ,meh , , , , , ,efmfacf ,efmfacs ,mdphiq , mdgot ,mdgoq ,mdvft ,mdmap ,mdgrdrm , mdgotmap, , / app /NORSET /NOASET /FS /NOFASET / NOSASET /METH /METHF /TRUE /NOQSET / 1 /0 /'no ' $ delete /v01p,,,, $ compile mdddtohh alter 'PARTN PHDSH,,VDE','' type parm,,i,n,nohset $ PARAML phdsh//'TRAILER'/1/S,N,NOHeSET $ nohset=noheset-noue $ MATGEN ,/VHE0/6/NOHESET/NOHSET/NOUE $ PARTN PHDSH,vhe0,VDE/PHDSH0,,,/1 $ alter 'TRNSP XEHST\/XHSE','' TRNSP XEHST/Xesh $ endalter 1-24507921 2007.0.0 Riged Elements - RBE2GS with IIRBE2 option fails with segmentation violation in sep1x RBE2GS with IIRBE2 option fails with segmentation violation in the SEP1X module 1-24510816 2007.0.0 Documentation - PCOMP - Composite Elements with Temperature Dependent Material Properties For layered composites, neither the TREF specified on the material entries or TEMP(INIT) or TEMP(MATE) are used to determine ply reference temperature. The TREF on the PCOMP or PCOMPG entries is used for all plies of the element. This is true for both linear and nonlinear analysis. SOL 106 looks up a new ply temperature at each incremental load step. PARAM,COMPMATT,YES is for shape memory which is only in SOL 106. 1-24618141 2007.0.0 Aeroelastic Analysis - UWM 7343 (MRBE3D THE ERROR VALUE FOR MONDSP1 ENTRY EXCEEDS 1.0e-10 The following message: *** USER WARNING MESSAGE 7343 (MRBE3D) THE ERROR VALUE FOR MONDSP1 ENTRY D_NG_UPR IS 1.126E-09, WHICH EXCEEDS THE TEST CRITERION OF 1.000E-10. may be encounted as the result of too tight a tolerance on a matrix inversion. The code will be changed to a value of 1.e-06. 1-24674065 Eigenvalue Analysis - READ, MFLUID Virtual Mass - Access Violation A simple model with MFLUID virtual mass elements fails in the READ Module with an Access Violation. Avoidance: Do not use "extreme" reordering, use metis by setting SYS206=8. 1-24741081 2007.0.0 Restarts, Optimization/Design Sensitivity - MAT2 - Wrong Answers There is an error in the way dependencies for Gij terms on the MAT2 entry are defined with the result that restarts that only change this data will not compute new element stiffness matrices. If these are the only entries that are changed, the results for the original submittal are likely to be repeated. This has implications in SOL 200. If the only designed properties are MAT2 terms and SCR=NO is used, the analysis will not use the updated properties and the analysis results will be unchanged from a previous iteration. Avoidance: For SOL 200, the avoidance is to use SCR=yes. For other solution sequences, one avoidance would be to make an addtional change in some other input value that will trigger computation of new stiffness matrices. 1-24850088 2007.0.0 Elements - PCOMP - Incorrect answers for non structural mass When ifp processes NSM, NSM1, NSML, and NSML1 entries it is supposed to replace: 1) PCOMP with PSHELL 2) PBCOMP with PBEAM 3) PBEAML with PBEAM 4) PBARL with PBAR 5) PRAC2D with PDUM8 in the TYPE field For the NSM and NSML entries it fails to replace PCOMP with PSHELL (item 1). Items 2-5 are done correctly for NSM, NSM1, NSML, and NSML1 entries 1-24917119 2007.0.0 Cyclic Modes - STATSUB, HARMONIC=m, HOUTPUT=n - No Preload Effects If cyclic symmetry eigenvalue analysis, SOL 115, is run with a static preload subcase, STATSUB, and a request for HARMONIC and HOUTPUT using a set id is included, then the preload effect of the STATSUB is missing and the eigenvalues calculated are those with no preload. To avoid use either request for ALL on the HARMONIC and HOUTPUT requests or use the following DMAP alter: compile cycmode alter 'call statcyc' equivx uxv/cyul/-1 $ 1-24978987 2007.0.0 Optimization/Topology - TOPVAR - SFM 4276 (GREAD) in DOPR1 for TOPVAR used with Beta function Topology optimization using function BETA on the DRESP2 entry fails with the message: SYSTEM FATAL MESSAGE 4276 (GREAD) ERROR CODE 3001 PID= 0 in DOPR1. 1-24992231 2007.0.0 Optimization/Design Sensitivity - Labeling errors for unfomatted design sensitivity prints If param optexit 4 is used to produce prints of unformatted sensitivities that include eigenvector responses, the following erroneous labeling is produced: ----- STATICS RESPONSES ----- -------------------------------------------------------------------------------- ---------------------------------------- COLUMN DRESP1 RESPONSE GRID/ELM VIEW COMPONENT SUB PLY/GRID NO. ENTRY ID TYPE ID ELM ID NO. CASE NO. -------------------------------------------------------------------------------- ---------------------------------------- 3 302 DISP 302 1 1 1 4 302 DISP 302 1 2 1 There are a number of things wrong with this output: a. eigenvector sensitivity does not support composites, so the last column is unnecessary b. View elm id is a p-element output and should the number should be blank unless this is a stress, force or strain response on a p-element c. The Mode id is printed in the view elm id slot d The mode id should be printed in the last slot and MODE should take the place of PLY/GRID User Action: Ignore the "Static Responses" when these are "Eigenvector Response" and obtain the Mode id from the View Elm ID column. 1-25080130 Documentation/QRG - PARAM description does not indicate that they can be used in rc and ini files The Parameter section in the QRG currently says: "Parameters are used extensively in the solution sequences for input of scalar values and for requesting special features. Parameters values are specified on PARAM Bulk Data entries or PARAM Case Control commands. The PARAM Bulk Data entry is described in the “Bulk Data Entries” on page 851. The PARAM Case Control command is described in the “Case Control Commands” on page 177. A complete alphabetical list of PARAMeter names and their functions is given in this section." It should be changed as follows to note the fact that it is often very useful to set parameters in an rc or ini file. "Parameters are used extensively in the solution sequences for input of scalar values and for requesting special features. Parameters values are specified on PARAM Bulk Data entries or PARAM Case Control commands. The PARAM Bulk Data entry is described in the “Bulk Data Entries” on page 851. The PARAM Case Control command is described in the “Case Control Commands” on page 177. PARAMs may also be used in .rc and .ini files as described in the MSC Nastran Installation and Operations Guide, Section 3.8 "Specifying Parameters". A complete alphabetical list of PARAMeter names and their functions is given in this section. 1-25202690 2007.0.0 Case Contol - REPCASE- Missing Punch Output A job writes no punch if repcases are present and subcase has sort2 output request 1-25259262 2007.0.0 Case Control - K2GG, M2GG - Error reading multiple DMIG entries K2GG and M2GG entries in the case control that span more than 72 characters are not read correctly. Avoidance: The avoidance is to place the DMIG names in a SET and reference the SET on the K2GG and M2GG entries. As shown SET entries that continue to the next line should end in a comma before column 72. SOL 101 CEND $-------2-------3-------4-------5-------6-------7-------8-------9-------0------- set 100 = KA1,KA2,KA3,KA4,KA5,KA6,KB1,KB2,KB3,KB4,KB5, KB6,KC1,KC2,KC3,KC4,KC5,KC6 K2GG = 100 set 200 = MA1,MA2,MA3,MA4,MA5,MA6,MB1,MB2,MB3,MB4,MB5, MB6,MC1,MC2,MC3,MC4,MC5,MC6 M2GG = 200 1-25264891 2007.0.0 Rigid Elements - Incorrect SPCFORCES are produced if load is applied to dependent DOF Wrong SPCFORCES occur if load is applied to dependent degrees of freedom. Avoidance: Use "param,sparsedr,no" Or use the following alter inversion 2005+ which is much more efficient. compile spdr1 alter 'if ( novspcg'(3) upartn uset,vpg/vpm,,,/'g'/'m'//1 $ paraml vpm//'null'////s,n,nlvpm $ if ( nlvpm>-1 ) then $ upartn uset,vspcg/vspcn,,,/'g'/'n'//1 $ partn gm,vspcn,vpm/,,,xxx/1 $ paraml xxx//'null'////s,n,nlxxx $ if ( nlxxx>-1 ) then $ matmod xxx,,,,,/xxx1/6 $ diagonal xxx1/xxx2/'whole'/0. merge ,,xxx2,,,,vpm/vpm1/1 $ umerge1 uset,vpm1,,,/vpmg/'g'/'m'//1 $ call xadd5 vspcg,vpmg,,,/vyx/cu/cu/cz/cz/cz $ equivx vyx/vspcg/-1 $ endif $ nlxxx>-1 endif $ nlvpm>-1 endalter 1-25272631 2007.0.0 SSSALTER - SEGYRO alter is missing in sssalter library since Nastran Version 2004 EGYRO alter is missing in sssalter library since Nastran Version 2004. This alter is required for calculating flexible rotating structures like helicopter rotor blades. The rotor dynamic feature doesn't support this. 1-25287241 2007.0.0 Frequency Response - Summary Table in FRRD1 prints zeroes for 51st and subsequent frequencies If there are more than 50 frequencies in the job, the Frequency Response Summary Table in FRRD1 prints the correct values only for the first 50 frequencies. After that, all frequencies are printed as 0.0. This does NOT effect the results. It is just a cosmetic error in the Summary Table. 1-25306330 2007.0.0 Axisymmetric Analysis - AXIC Bulk Data entry - Wrong answers If the AXIC entry is present and a limited set of ouput is requested such that sparse data recovery is activated then output will be wrong. Avoidance: Insert PARAM,SPARSEDR,NO or specify ALL on at least one Case Control output command. 1-25307074 2007.0.0 Axisymmetric Analysis - SFM 6406 (G4SPC), AXIC and SPCAX entry higher than 100 and SPCAX An axisymmetric analysis with AXIC and SPCAX defined greater than 100 leads to the following fatal message: *** SYSTEM FATAL MESSAGE 6406 (G4SPC) GRID POINT ID= 102000001 ON AN SPCP/SPCDP ENTRY HAS COMPONENT= 23 GREATER THAN MAX. NDOF= 6 USER ACTION: CONTACT MSC.SOFTWARE CLIENT SUPPORT. PROGRAMMER ACTION: CHECK SPCP/SPCDP ENTRIES CREATED IN GP0 MODULE. *** SYSTEM FATAL MESSAGE 6406 (G4SPC) GRID POINT ID= 101000001 ON AN SPCP/SPCDP ENTRY HAS COMPONENT= 23 GREATER THAN MAX. NDOF= 6 USER ACTION: CONTACT MSC.SOFTWARE CLIENT SUPPORT. PROGRAMMER ACTION: CHECK SPCP/SPCDP ENTRIES CREATED IN GP0 MODULE. *** USER INFORMATION MESSAGE 4110 (OUTPX2) 1-25476171 2007.0.0 Case Control - SUBSEQ - UWM 2092, Missing Output If a SUBSEQ entry contains a factor for each of the preceding subcases but ends in a comma, UWM 2092 may be issued and the output for the SUBSEQ may be missing. 1-25479301 2007.0.0 Data Recovery - PARAM,SPARSDR,NO and PARAM,POST,-1 - inefficient If PARAM,SPARSDR,NO and PARAM,POST,-1 are specified in a run where SORT2 output is computed, the run will perform unnecessary operations. compile sedrcvr alter 'if ( not(sort2) or post','' if ( not(sort2) and not(POST=-1 OR POST=-2 OR POST=-6) ) then $ alter 'IF ( sort2mm and (POST=-1','' IF ( (sort2mm or app1='mmreig') and (POST=-1 OR POST=-2 OR POST=-6) and not(acms) and fullsedr='no' ) THEN $ delete /opg1,,,, $ endalter 1-25697761 2007.0.0 SOL 600 - MARCIN written in small field format is not translated into marc.dat file correctly MARCIN written in small field format is not translated into marc.dat file correctly. For example: If marcin entry below is in the input deck, "feture,6701" should be in the parameters section of marc.dat file, but it does not. marcin -1 0 feature,6701 If marcin entry below is in the input deck, "feture,6701" is in the parameters section of marc.dat file. marcin,-1,0,feature,6701 Avoidance: Use free field format. 1-25716447 2007.0.0 Rotordynamics - Linear Gap Constraing - SFM 3001, RBMG4, SOL 101 Attempting to run a SOL 101 with the rotordynamic and linear gap constraint will result in the following fatal message: *** SYSTEM FATAL MESSAGE 3011 (RBMG4) ATTEMPT TO WRITE A TRAILER ON FILE 201 WHEN IT HAS BEEN PURGED Avoidance: Include one of the two following DMAP alters. $ for Version 2005.5.0 compile sestatic alter 'call phase1c','' $ Version 2005.5.0 IF ( nouset>0 and not(mdstat) and ( (dbrcvx='scratch' and not(pardcmp)) or (pardcmp and hlpmethx=1 and procid=1) ) ) CALL PHASE1C KAA ,USET ,SILS ,EQEXINS ,maa , PA , , ,cases , , drg , , , / KLL ,LLL ,Dar ,ULL ,mr , mlr ,mll ,mrr ,QR ,PL , LRSEQ , , , , , / SEKR /SEMR /SELR /NOLSET /NORSET / true /ERROR /FIXEDB /NOQSET /NOTSET / ' '/TRUE /false /false /noa / false $ $ for versions 2005.1.0 and 2005.0.0 compile sestatic alter 'call phase1c','' $ tested on version 2005.1.0 and 2005.0.0 IF ( nouset>0 and not(mdstat) and ( (dbrcvx='scratch' and not(pardcmp)) or (pardcmp and hlpmethx=1 and procid=1) ) ) CALL PHASE1C KAA ,USET ,SILS ,EQEXINS , , PA , , ,cases , , drg , , / KLL ,LLL ,Dar ,ULL ,mr , mlr ,mll ,mrr ,QR ,PL , LRSEQ , , , , / SEKR /SEMR /SELR /NOLSET /NORSET / true /ERROR /FIXEDB /NOQSET /NOTSET / ' '/TRUE /false /false /noa / false $ Test file place on /msc/app/etl as q1-25716447.dat 1-25721381 2007.0.0 DMAP - DMAP compiler issues UWM 1 instead of UFM 12 A DMAP compile may issue UWM 1 instead of UFM 12: # *** USER FATAL/WARNING MESSAGE 12 (XGPIDG) ERROR IN DMAP INSTRUCTION NO. %1, ILLEGAL CHARACTER IN DMAP INSTRUCTION NAME. 1-25750061 2007.0.0 Connector Elements - PFAST with MCID=-1 and MFLAF=1 may yield strange element coordinate system PFAST entry with MCID=-1 and MFLAG=1 may yield strange elemet coordinate system for force recovery. If MCID=-1, MFLAG should always be blank or 0. The fix will set MFLAG to if MCID=-1 regardless of value specified. 1-25765441 2007.0.0 Connector Elements - CFAST, CWELD may not find projection and may fail to connect If gridA and gridB are specified on a CFAST or CWELD and grid GS is left blank for the attached model the algorithm may be unable to project. 1-25773137 Superelements - PARAM,SESDAMP,YES does not work correctly when PARAM,AUTOQSET,YES is also specified. Attempting to apply different damping to different superelements using PARAM,SESDAMP,YES combined with PARAM,AUTOQSET,YES leads to wrong answers. Avoidance 1: Use manual Q-set entries instead of employing PARAM,AUTOQSET,YES Avoidance 2: Employ the DMAP alter given below when using PARAM,AUTOQSET,YES compile phase1b alter 'call semrb'(9) np=noqset-xnoq $ delete /vq,,,, $ matgen ,/vq/6/noqset/xnoq/np $ matgen ,/vq1/6/noqset/np/xnoq $ partn mbqq,vq,/mbqq0,,, $ delete /mbqq,,,, $ merge ,,,,mbqq0,vq1,/mbqq $ 1-25775091 2007.0.0 Aeroelastic Analysis - MONPNT2 - SFM 3008 (ILMP1) Insuficient entry _MONPNT2 The message: SYSTEM FATAL MESSAGE 3008 (ILMP1) Insuficient entry _MONPNT2 can be produced if a MONPNT2 bulk data entry invokes an element that is connected to the highest numbered grid point in the model. Avoidance: Do not select an element on a MONPNT2 that is connected to the highest numbered grid point in the model. Else renumber the grids associated with the element so that the grid is not the highest in the model. 1-25801752 Postprocessing - DBC, PARAM,POST,0 - SFM 6062 (DBC), Poor Message If the XDB size is exceeded the following fatal message is issued. *** SYSTEM FATAL MESSAGE 6062 (DBC) *** DIOMSG ERROR MESSAGE 6 FROM SUBROUTINE WRTLST OVERFLOW OF DICTIONARY PRIMARY INDEX FOR DATA BASE UNIT 1. This message should be improved to instruct the user to increase the default RECL for the xdb file on an ASSIGN statement. 1-25860601 SOL 700 - Does not work on SunOs - duplicate see 1-25836331 opened by Ani SOL 700 does not work on Solaris and Primepower in MD Nastran 2006r1. There is no avoidance. The f06 file will contain the message: Running MSC.Dytran LS-DYNA step 1 *** Elapsed Time - step 1: 0.0u 0.0s 0:00 0% 0+0k 0+0io 0pf+0w Error(s) occured during step 1 - See file: jid.dytr_prep.out - Search for string: *** Error *** ISHELL PROGRAM '/local/msc2006t2/dyna/solaris/run_dytran' COMPLETED *** The jid.dytr_pre.out file will contain the message: ld.so.1: dytran-lsdyna: fatal: /local/msc2006t2/dyna/solaris/libfui.so.2: wrong ELF class: ELFCLASS32 Job cre.5362 on delagoa: received signal KILL. 1-25925548 2007.0.0 Elements - PCOMP yields Incorrect results for the BEND option When the BEND option is added to the pcomp entry, the generated mat2 entry has the Gij terms for membrane instead of the Gij terms for bending . 1-25986051 2007.0.0 Superelements - SFM 3007 (BNDSPC), BNDSP3 A job with many superelements may fail with the following error message: *** SYSTEM FATAL MESSAGE 3007 (BNDSPC) ILLEGAL INPUT TO SUBROUTINE BNDSP3 Avoidance: Possible DMAP avoidance is to test for the presence of coordinate systems and/or SPC's on boundary points of the superelement. If neither of these exists, the call to the BNDSPC module may be skipped with DMAP as neither YS nor USET is updated. 1-26047321 2007.0.0 Optimization/Design Sensitivity - UFM 7544 (DSAD13) Issued in error The following error message may be issued in error when a DRESP1 requests a CFAILURE response with item code 7 (interlaminar shear) and this is the last response applied to the element. *** USER FATAL MESSAGE 7544 (DSAD13) DRESP1 ID 7 IS A CFAILURE RESPONSE WITH ITEM CODE 7: INTERLAMINAR STRESS/STRAIN. USER INFORMATION: THIS RESPONSE IS BEING APPLIED TO PLY NUMBER 7, WHICH IS THE FINAL PLY. INTERLAMINAR STRESS/STRAIN RESPONSES ARE NOT AVAILABLE FOR THE FINAL PLY USER ACTION: DO NOT APPLY ITEM CODE 7 TO THE FINAL PLY User Information: Responses are ordered by element id, ply no and item code. If the last cfailure response for an element has item code 7, the above fatal message is produced regardless of the ply no. it is associated with. The message is only correct when the ply is the highest numbered ply for the element. Avoidance: An avoidance is to request an additional CFAILURE response for the element that has a higher item code (8 or 9) or is applied to a higher numbered ply than the item code 7 request. 1-27318160 2007.0.0 Elements - PBMSECT - Arbitrary Section, Incorrect direct shear stresses The direct shear stresses are off by a factor of the area. The forces sent to the routines that calculate cross section stresses are inadvertantly divided by the area prior to the stress calculation. 1-28310299 2007.0.0 Superelements - EXTSEOUT, acoustic cavity, AUTOQSET- SFM 6143 When creating an external superelement with an acoustic cavity and PARAM,AUTOQSET,YES the run may fail with SFM 6143. For example: SYSTEM FATAL MESSAGE 6143 (UMRGE1) THE SIZES OF THE INPUT MATRICES AND THE DOF SETS ARE INCOMPATIBLE. SPECIFICALLY: The number of columns in EXTDB is not equal to the size of the A-set User Information: Size of input matrices and DOF sets: EXTDB : Rows= 248 by Cols= 33 Set sizes: Major (G-set)= 278 Set0 (A-set)= 248 Set1 (COMP-set)= 30 Avoidance: Insert the folowing DMAP alter: compile extout alter 'MPYAD INTEXTA,AA,'(,3),'' umerge1 usetx,extdb,,,/temp/'g'/'a'//1 $ 1-28316191 2007.0.0 Aeroelastic Analysis - MONDPS1 - Output missing in SOL 144 runs The MONDPS1 feature was added in Version 2005R3. Due to a dmap error, the prints associated with this entry are not produced in a static aeroelastic run. User Avoidance: Insert the following dmap alter in the run: compile mnpntdr list alter 'aero'(3,-1) IF ( DSP1=-1 and (statics or app='reig') ) THEN $ MPYAD SDMPIVa,Ua,/UMONDISP/1 $ mppstatc casedr,dispmon,umondisp,//'MONDISP' $ ENDIF $ DSP1=-1 existing tpl deck msimpms.dat shows the affect of this fix. EHJ 5/24/2006 alter 'umondisp'(1,-1),'umondisp'(1,3) 1-31110515 Topology Optimization - MAT2 - Poor Message In topology optimization, when a PSHELL entry references a non-supported material such as a mat2 entry which is not, UFM 2042 is issued from routine ifp10e and indicates an undefined material property id. The message should indicate that the material is not supported in topology optimization. 1-33238331 Optimization - mattx associated with a matx referenced by a dvmrel1 - Wrong Sensitivies If a dvmrel1 (or dvmrel2) references a matx entry (e.g., a mat1) that has an affiliated mattx entry (e.g., a matt1) the senstivities will be wrong whenever the value determined from the mattx entry differs from that given on the matx entry. User information: It has never been possible to design a material property that varies as a function of temperature provide by a tablem1 entry. However, it should be possible to support desiging a material property that is not a function of temperature when other properties are a function of temperature. For example, if the Young's module is a function of temperature, but the density is not, it should be possible to design the density. Because of a bug, the sensitivities in this case will be incorrect. User avoidance: There is no avoidance except to not use temperature dependent properties for any material id that has designed properties. 1-33793531 2007.0.0 SOL 600 - Translation of BCBODY LINE section into CONTACT option is wrong if 8 or more points When there are 8 or more points in the LINE section of a BCBODY the translation of the BCBODY LINE section into the CONTACT option is wrong. 1-34278877 2007.0.0 Optimization/Design Sensitivity - PBARL, PBAR, PBEAM, PBEAML - Missing entries in punch file When an optimization job includes both PBARL and PBAR entries and PBAR properties are being designed, no updated PBAR entries are written to the punch file at the end of the design. This shows up either with the DESPCH/DESPCH1 request or the ECHO=PUNCH(NEWBULK) request. Avoidance: Add a dummy pbrsect entry if the input deck has PBARL (or dummy PBMSECT if has PBEAML). An example as follows POINT 80001 0.0 0.0 POINT 80002 1.0 0.0 set1 80003 80001 80002 pbrsect 80004 1 OP OUTP=80003,t=1.0 1-34831285 2007.0.0 Aeroelastic Analysis - Labelling for inertial derivatives in static aeroelasiticity is incorrect The trim results for rotational accelerations currently show units of length/s/s when PARAM ANITS=1.0 (default) is used. This should be radians/s/s When PARAM AUNITS has some value other than 1.0, the rotational accelerations are labled as LOAD FACTOR. This should be radians/s/s_per_g 1-34885375 2007.0.0 SOL 700 - CTETRA10 - Not translated correctly CTETRA10 is not supported in SOL700. If inlcuded the translation will cause an error in the explicit solver. 1-34974981 2007.0.0 SOL 400 - QUAD4, PARAM,LGDISP - UFM 4551 or Floaing Point Overflow in xqd4pd SOL 400 with CQUAD4 elements and PARAM,LGDISP may fail with UFM 4551 or a floating point overflow in xqd4pd. 1-34997021 2007.0.0 Dynamics - PARAM,AUTOQSET,YES - SFM 3001 in TA1 If PARAM,AUTOQSET,YES is requested with nonstructural mass specified on the NSM* entry then the run will terminate with: *** SYSTEM FATAL MESSAGE 3001 (TA1) THE OUTPUT DATA BLOCK IN POSITION 9 DOES NOT EXIST. Avoidance: Insert the following DMAP alter: compile phase1dr alter ' ta1 ','' TA1 mpts,ECTS,EPTS,BGPDTS,SILS,ETT,CSTMS, dit,ectas,eht/ EST,,GEI,GPECT,,vgfd,ditid,,nsmest/ LUSETS/-1/S,N,NOSIMP/1/S,N,NOGENL/0/// s,n,freqdeps/shldamp/s,n,mge/nsm $ 1-35331198 2007.0.0 Superelements/Acoustics - PARAM FLUIDSE, c-set - SFM 5423 A model with PARAM,FLUIDSE and some or all of the fluid superelement's boundary is in the c-set then the run will fail with: *** SYSTEM FATAL MESSAGE 5423 (MPYAD1) ATTEMPT TO MULTIPLY INCOMPATIBLE MATRICES FILE# COLS ROWS FORM TYPE NZWD DENS TRAILER FOR MOT1 IS 101 313 606 2 2 1212 9665 TRAILER FOR GOQ IS 102 10 919 2 2 1838 10000 TRAILER FOR (NONE) IS 0 0 0 0 0 0 0 TRANSPOSE FLAG IS 1 Avoidance: Insert PARAM,MHRED,NO or insert the following DMAP alter: compile semr3 alter 'if (mhred='(2),'' IF ( MHRED='YES' AND NOT(FLUID) ) THEN $ endalter 1-35436091 2007.0.0 SOL 600 - BCBODY to CONTACT is wrong for 2nd rigid body represented by NURBS The 2nd (and perhaps all except the 1st) rigid body represented by NURBS is wrong. The Nastran-to-Marc translator does not write the "data" section from BCBODY to CONTACT. Note that the translation of the "application region" section (i.e., NURBS section) is correct for all the rigid bodies. 1-35546885 2007.0.0 Eigenvalue Analysis - OMODES, RESTART - Wrong Answers The OMODES Case Control command is ignored on restart. Avoidance: Insert the following DMAP alter: compile super3 alter 'call sedisp'(,-1) delete /UG,QG,QMG,OL1S,gpke $ endalter 1-35665271 2007.0.0 Aeroelastic Analysis - MONPNT2 results missing when AECONFIG request is present - UFM 2170 (PARTN2) A MONPNT2 bulk data entry in an input file that also contains an AECONFIG case control command produces the following error message: *** USER FATAL MESSAGE 2170 (PARTN2) BOTH THE ROW AND COLUMN PARTITIONING VECTORS ARE PURGED AND ONLY ONE MAY BE. 0FATAL ERROR User Information: This is caused by a datablock qualifer not being set. User Avoidance: The easiest avoidance is to omit the AECONFIG case control command so that the default value of AECONFIG=AERO2GSD is used. : 1-35683049 2007.0.0 Nonlinear Transient Analysis with rotordynamics - RGYRO, UNBALNC - SFM 3001 If there are multiple subcases in SOL 129 with RGYRO command and UNBALNC loads then the run may fail with SFM 3001 for UNBVECD. Avoidance: Insert the following DMAP alter: compile nltran alter 1 file UNBVECD=save $ alter 'call gyrotran' DELETE /UNBVECD,,,, $ endalter 1-35917921 2007.0.0 Case Control - Scale Factor applied to P2G lead to UWM 2070, followed by UFM 300 Attempting to apply a scale fact to P2G, (i.e. P2G = 1.25*PAX) may lead to user fatal message 300 in the case control section. 1-37227076 2007.0.0 SOL 600 - Loads in SOL 600 subcase get incorrectly applied For SOL 600, If the same load is applied in two subcases you must use PARAM,MRDISCMB,1 At the request of Marc development, this parameter was introduced together with an algorithm to combine distributed loads with the same magnitudes. This algorithm fails in problems where just one distributed load of the same magnitude is applied in two subsequent load cases. The default for PARAM,MRDISCMB will be changed to zero for such loadings. The avoidance is to apply slightly different distributed loads. It is unlikely likely that In a practical problem with many loads of varying magnitude that this problem will be encountered. 1-37590653 2007.0.0 Elements - CBAR, PBARL - Incorrect mass with PBARL With Nastran 2005.5, also known as 2005r3, if a PBARL is defined with non structural mass, the mass of the element doesn't include the non structural mass. Avoidance: Activate the previous element formulation of pbarl with nastran twbrbml=1 1-37797311 2007.0.0 Distributed Memory Parallel - SOL 103 ACMS with DMP does excessive MPI - Poor Performance Verified by running the "vlaw" benchmark with SOL 103 and asking for DISP(PLOT)=ALL. PJS 12/12/06 DMP performance will be degraded when using ACMS with SOL 103 due to an MPI broadcast of the full (A-set size) PHI matrix from Master to all Slave processes. This broadcast should not be necessary. Avoidance: Insert the following DMAP alter: compile modefsrs alter 'if ( (acms or mdacms) and nproc','' if ( (acms or mdacms) and nproc>1 and not(fdmodes or app='reig') ) then $ alter 'if ( (acms or mdacms) and nproc'(2),'' if ( (acms or mdacms) and nproc>1 and not(fdmodes or app='reig') ) then $ compile semodes alter 'call modefsrs' if ( procid>1 and mdacms ) then $ CALL PRTSUM /RSTJUNK $ exit $ endif $ endalter 1-37948581 2007.0.0 DMAP - MATMOD Option 16 to process a KAA matrix gives DMIG output with an invalid grid point ID Use of MATMOD Option 16 to process a KAA matrix gives DMIG output with an invalid grid point ID, making it impossible to subsequently use this DMIG output properly. 1-38453316 Documentation/QRG - CBUSH - SOL 106, missing CBUSH forces with LGIDSP In sol 106, nonlinear, with PARAM,LGDISP,+1 cbush forces are missing from the output. Avoidance: Request STRESS and the forces will be in the nonlinear stress block. A comment should be added to the Quick Reference Guide: 10. In SOL 106 with PARAM,LGDISP,1 the force recovery is part of the nonlinear stress recovery. This means that the FORCE request will not provide CBUSH forces; instead, appropriate STRESS/NLSTRESS are required to obtain the nonlinear force and stress in bush elements. 1-38620791 2007.0.0 Superelements - Restarts of external SE assembly runs fail with SFM 4276 in the GP1 module Restarts of assembly runs employing external SEs (using the EXTSEOUT feature) fail with SFM 4276 in the GP1 module. Avoidance: Insert the following DMAP alter in the restart run: compile phase0 alter 'if ( not(noxsort) and not(selist)','' if ( false ) then $ endalter 1-38982011 2007.0.0 Optimization/Design Sensitivity - ANALYSIS=BUCK - SFM 3001 CASEXX does not exist If SOL 200 runs with a single subcase the ANALYSIS = BUCK, the following system fatal message is produced: *** SYSTEM FATAL MESSAGE 3001 (SEP3) THE INPUT DATA BLOCK NAMED CASEXX AT POSITION 1 DOES NOT EXIST. This is caused by the fact that there is no subcase with ANALYSIS=STATICS to obtain the static deflection to drive the buckling analysis. User Information: In SOL 105, a buckling subcase without a statics subcase produces the somewhat more meaningful message: *** USER FATAL MESSAGE 4683 (LNNRIGL) KDIF MATRIX NEEDED FOR EIGENVALUE ANALYSIS User Avoidance: Add a static subcase to the input deck 1-39271221 2007.0.0 Installation - Install notes have two errors The "Install Notes" have two errors: 1. The notes state that linux does not support SMP. Linux does support SMP 2. The notes state that 2005.5 runs on RedHat 7.3, RH 8, and SuSE 8. Please change to it has been run on RedHat 9, RH 3, RH 4, SuSE 9, Fedora Core 2 1-39297485 2007.0.0 Superelements - Superelement with zero mass in residual elements leads to UFM 9088 (SELG) In superelement analysis, if there is zero mass in the residual elements, the following error will be reported (attachment internal.dat) ^^^ USER FATAL MESSAGE 9088 (SELG) ^^^ PARAM,INREL,-2 IS PRESENT BUT NO MASS IS DEFINED. Avoidances: 1) add a dummy grid to the residual with low mass. or, 2) use param,inrel,-1 and SUPORT entry or, 3) use the following dmap in V2005.5.0: compile selg alter 76,87 $ v2005.5.0 IF ( NULLMJJ=-1 ) THEN $ MESSAGE //' USER'/WRN/'MESSAGE 9088 (SELG)' $ MESSAGE //' PARAM,INREL,-1 or PARAM,INREL,-2 IS PRESENT BUT'/ ' NO element MASS IS DEFINED for SEID='/seid $ MESSAGE //' USER INFORMATION: The loads will pass'/ ' to the assembled residual for final processing and '/ ' inertia relief calculations' $ ELSE $ compile statrs alter 'nullqrr' if (nullqrr=-1) then message//'no mass present in residual processing of inertia relief' message//'-- fatal' exit endif $ 1-39313341 2007.0.0 Acoustic Analysis - Exterior acoustics with DMP - ACFPMRESULT - Missing Output If DOMAINSOLVER FREQ is requested with exterior acoustic analysis in sol108 then ACFPMRESULT data recovery will be missing output. 1-39507540 2007.0.0 Frequency Response - DMP and ACMS - frequency dependent elements - SFM 5423 If DMP and ACMS are requested in frequency response with frequency dependent elements, then the run may fail withSFM 5423. Avoidance: Insert the following DMAP alter: compile freqrs alter 'frrd1' alter 'merge ,,,ppf,,' merge ,,,qsgf,,expand,/qsgfe/1 $ merge ,,,qmgf,,expand,/qmgfe/1 $ alter 'DISUTIL ppfe,,,' DISUTIL qsgfe,,,/qsgf0///-8 $ DISUTIL qmgfe,,,/qmgf0///-8 $ alter 'DISUTIL ypf1,,,' DISUTIL qsgf,,,/qsgf0///9 $ DISUTIL qmgf,,,/qmgf0///9 $ alter 'DISUTIL ypf10,,,' DISUTIL qsgf0,,,/qsgf2///3 $ DISUTIL qmgf0,,,/qmgf2///3 $ alter 'equivx ypf10' equivx qsgf0/qsgf/-1 $ equivx qmgf0/qmgf/-1 $ alter 'equivx ypf12' equivx qsgf2/qsgf/-1 $ equivx qmgf2/qmgf/-1 $ endalter 1-39801126 Documentation/QRG - SORT1 vs SORT2 description in QRG The description of SORT1/SORT2 in Remark2 of the QRG does not describe that for Statics, ANY SORT2 request will result in all output in SORT2. 1-40036154 2007.0.0 MCFRACTION - Missing output or unexpected output when degree of freedom not specified. The MCFRACTION Case Control command produces unexpected results if the grid point's degree of freedom is not specified as required in the set. For example: SET 1000=92 MCFRACTION(sort=absa,print)=1000 produces modal contribution factor output for grid 9 degree of freedom 2. Specifying a grid 97 will produce no output without a warning message. 1-40041100 2007.0.0 Buckling - STATSUB(BUCKLING) - Missing Output In sol 105 buckling analysis contains two buckling subcases that have different preloads but the same buckling load, the second buckling subcase will be skipped. 1-40102391 2007.0.0 Superelements - AUTOMSET - Possible wrong results for AUTOMSET=YES If rigid elements/MPCs connect DOF on the boundary of a superelement with AUTOMSET=YES then wrong answers may be produced. Avoidance: The following alter can be used to avoid the error --- compile mce1s $ alter 'imm,ym','' $ mpyad imm,ym,/rinvym//-1 $ alter 'i11,yy','' $ mpyad i11,yy,/y//-1 $ The attached input data file demonstrates the error. 1-40265587 Superelements - Incorrect Modal Complex Eigenvalues Modal complex eigenvalue analysis using superelements produces incorrect damping coefficients. 1-40401601 2007.0.0 Matrix Methods - Sparse decomp may hang during the reordering in a specific memory range It is possible for the sparse decomp to hang during the reordering phase provided that all of the following conditions occur: 1) Either BEND or AMF reordering is specified (BEND is the default) 2) Grid compression is feasible (that is, the EQXIN, SILS, and USET tables are available). 3) Enough memory is available for the grid compression 4) Not enough memory is available for BEND or AMF The offending routine is DFMSYN, which, under these conditions, and only under these conditions, tries to perform MMD reordering by calling DFMMD1 with the wrong data structures. Usually the job will hang, but it may also produce a segmentation fault or an error message like: *** SYSTEM INFORMATION MESSAGE 6916 (DFMSYN) DECOMP ORDERING METHOD CHOSEN: MMD, ORDERING METHOD USED: MMD *** SYSTEM WARNING MESSAGE 6705 (DFMAGC) LOGIC ERROR 2 IN SYMBOLIC PHASE OF SPARSE DECOMPOSITION USER INFORMATION: PERFORMANCE MAY BE DEGRADED AND/OR MORE MEMORY MAY BE REQUIRED IF THE MATRIX HAS NOT BEEN RESEQUENCED PROPERLY This error can be reproduced in released systems starting with v2001. Using the tpl deck bst80a, the error can be reproduced in v2001 and v2004 by specifying mem=1754kw, in v2005 by specifying mem=1865kw, and in v2006R1 by specifying mem=2000kw. The amount of memory needed to cause the error can be estimated from the diagnostic output produced by setting sys166=2. For bst80a, one sees the following message in the f06 file: GRID MATCHING COMPRESSION USED: NUMBER OF COLUMNS IN MATRIX = 32015 NUMBER OF NONZEROES IN FULL MATRIX = 654646 NUMBER OF COMPRESSED COLUMNS IN MATRIX = 6403 NUMBER OF COMPRESSED NONZEROES = 50252 If we let IWFR = NUMBER OF COMPRESSED NONZEROS N = NUMBER OF COLUMNS IN MATRIX NEWN= NUMBER OF COMPRESSED COLUMNS IN MATRIX then if specify memory so that the USER OPENCORE (as reported in the f04 file) is approximately 15*NEWN + IWFR + 9*N the error should occur. This is not an exact formula, so it may be necessary to alter the value a bit. The avoidance is to specify enough memory so that USER OPENCORE (as reported in the f04 file) is at least 17*NEWN + IWFR +10*N . Another avoidance is to specify a reordering method besides BEND or AMF, by setting sys206=8, or setting sys206=9, or setting sys206=2. 1-40506427 2007.0.0 Eigenvalue Analysis - RIGID=LAGRANGE - SFM 3034 Lanczos eigenvalue problems with rigid=lagrange may fail with the message: *** SYSTEM FATAL MESSAGE 3034 (LNNHERR) INTERNAL FAILURE IN THE LANCZOS PROCEDURE: M-ORTHOGONAL QR PROCEDURE FAILED TO CONVERGE. PROBABLE CAUSE: MASS MATRIX IS INDEFINITE (MODES) OR STIFFNESS MATRIX IS INDEFINITE (BUCKLING). USER ACTION: CONTACT MSC CLIENT SUPPORT. 1-40535203 2007.0.0 Optimization/Design Sensitivity - BUCKLE and SCR=NO - Wrong answers If running SOL 200 with ANALYSIS=BUCKLE and SCRATCH = NO, then wrong answers occur. This is only detected by running a buckling solution using the output from PUNCH NEWBULK feature. The eigensolution is different between the final SOL 200 and NEWBULK SOL 105. Avoidance: either run the SOL 200 with SCRATCH = YES or insert this dmap alter compile sekdr alter 'DBSTATUS KDJJ,'(,-1) delete /kdjj,,,, $ endalter 1-40553852 2007.0.0 SOL 700 - KU field of MATDS08 casuses error in d3hsp KU field of MATDS08 casuses an error in d3hsp. %E *** Error DYTRAN Input 42 MATDS08 1 2 0. 1. MATDS08 (MID = 1) - has wrong KU value 0. 1-40553875 2007.0.0 SOL 700 - PSHELL - Shell thickness cannot differ for each nodes If the thickness specified on a PSHELL entry is not same for all nodes of a shell element element, it causes an error in d3hsp. 1-40558687 2007.0.0 Optimization/Design Sensitivity - ADS9P4 - SFM 3008 (UTMCOR), Insufficient Core The following message should be improved to indicate the additional memory required: SYSTEM FATAL MESSAGE 3008 (UTMCOR) MODULE = DOM9 INSUFFICIENT CORE AVAILABLE FOR SUBROUTINE ADS9P4 PROGRAMMER INFORMATION: LOCATION KEY = ADS9P4 FATAL ERROR Avoidance: Request more memory or use OPTCOD=DOT on the DOPTPRM entry 1-40726454 2007.0.0 Optimization/Design Sensitivity - GPFORCE, ESE sensitivities incorrect with TEMP(LOAD) GPFORCE and Element Strain Energy response sensitivities are incorrect in the presence of thermal loads. 1-40803671 2007.0.0 Heat Transfer Analysis - SOL 101 153 or 159 - Rigid Elements result in UFM 5419 (CHKRGD) If a heat transfer analysis contains rigid elements, the run will fail in v2005.5+ with the message: *** USER FATAL MESSAGE 5419 (CHKRGD) RBE2 ELEMENT ID = 0, REFERENCES FLUID GRID POINT ID = XX. Avoidance: The only avoidance is to replace the rigid connections with MPC equivalents. 1-40966401 2007.0.0 Dynamics - PARAM,AUTOQSET - Poor performance Poor performance may be experienced when PARAM,AUTOQSET,YES is specified and structural (e.g., GE on MATi) or viscous damping (e.g., CDAMPi, CVISCi) are present. Avoidance: Insert the following DMAP alter: compile phase1b alter 'smpyad.*rgt,bgg ,rgq,,,','smpyad.*rgq,k4gg,rgq,,,\/'(,1) matmod k4gg,,,,,/nzk4,/12/s,n,nzk4/1 $ if ( nzk4>0 ) then $ partn k4gg,nzk4,/k4xx,,, $ remove null columns/rows partn rgq,,nzk4/rxq,,,/1 $ and corresponding rows partn rgt,,nzk4/rxt,,,/1 $ smpyad rxq,k4xx,rxq,,,/k4qq/3////1////6 $ smpyad rxt,k4xx,rxq,,,/k4tq/3////1 $ delete /rxt,rxq,,, $ else $ smpyad rgq,k4gg,rgq,,,/k4qq/3////1////6 $ smpyad rgt,k4gg,rgq,,,/k4tq/3////1 $ endif $ if ( not beigrd ) then $ matmod bgg,,,,,/nzb,/12/s,n,nzb/1 $ if ( nzb>0 ) then $ partn bgg,nzb,/bxx,,, $ remove null columns/rows partn rgq,,nzb/rxq,,,/1 $ and corresponding rows partn rgt,,nzb/rxt,,,/1 $ smpyad rxq,bxx,rxq,,,/bqq/3////1////6 $ smpyad rxt,bxx,rxq,,,/btq/3////1 $ delete /rxt,rxq,,, $ else $ smpyad rgq,bgg,rgq,,,/bqq/3////1////6 $ smpyad rgt,bgg,rgq,,,/btq/3////1 $ endif $ endif $ not beigrd dbstatus mbqq//s,n,nombqq $ if ( nombqq>0 ) then $ if ( kdampx>-1 ) then $ add5 bqq,mbqq,,,/mbqq1 $ equivx mbqq1/bqq/-1 $ else $ add5 k4qq,mbqq,,,/mk4qq1 $ equivx mk4qq1/k4qq/-1 $ endif $ endalter 1-40966515 2007.0.0 Elements - Multiple element types with many property entries - Poor performance in TA1 If there are multiple element types and at least one element type has many property entries then poor performance may be observed in the TA1 module executed in subDMAP PHASE1A. This error can occur if there is a TOPVAR entry present or there is a unique property entry assigned to each element. Avoidance: Insert the following DMAP alter: compile phase1a alter ' ta1 '(2,-1) putsys(1,56) $ alter ' ta1 '(2) putsys(0,56) $ endalter 1-41130517 2007.0.0 Optimization/Design Sensitivity - SFM 3007 (DSAD) ILLEGAL INPUT TO SUBROUTINE DSAD6, UFM 6553 The message: SYSTEM FATAL MESSAGE 3007 (DSAD) ILLEGAL INPUT TO SUBROUTINE DSAD6 appears when a scaled weight constraint is invoked in a subcase other than the first. This occurs in combination with the message: USER FATAL MESSAGE 6553 (DOPR3R) A WEIGHT RESPONSE HAS BEEN INVOKED BY THE DESSUB COMMAND IN SUBCASE 22, WHICH IS NOT THE FIRST SUBCASE. It is possible that this UFM will not cause the run to stop due to a programming bug. Avoidance: Constraints on weight or volume must be placed above the subcase level or in the first subcase 1-41330851 2007.0.0 Optimization/Design Sensitivity - tpl test dtrustr2 fails in DOM10 - SFM 4276 on AIX An optimization job with an active Trust Region request may fail in DOM10 with SYSTEM FATAL MESSAGE 4276. Avoidance: use the dmap alter below: compile desopt alter 'if ( trustR ) D','' if ( trustR ) THEN DBEQUIV DATABLK=(XINIT/XINITR,DESTAB/DESTABR, OPTPRMG/OPTPRMGR,R1VALRG/R1VALRGR,R2VALRG/R2VALRGR, RSP2RG/RSP2RGR,R3VALRG/R3VALRGR, R1TABRG/R1TABRGR,CNTABRG/CNTABRGR,DSCMG/DSCMGR, OBJTBG/OBJTBGR,RR2IDRG/RR2IDRGR, RESP3RG/RESP3RGR,RQATABRG/RQATBGR, DRSTBLG/DRSTBLGR,WMIDG/WMIDGR, MODRPRG/MODRPRGR,GEOM1/GEOM1R,GEOM2/GEOM2R, GEOM1QS/GEOM1QSR,GEOM2QS/GEOM2QSR, GEOM4QS/GEOM4QSR,FRQRPRG/FRQRPRGR, R1MAPR/R1MAPRR,R2MAPR/R2MAPRR,MTRAK/MTRAKR), WHERE (DESITER=DESCYCLP AND WILDCARD) , CONVERT (DESITER=DESCYCLP+1) OVRWRT $ DELETE /RSP2RG,CNTABRG,R1TABRG,DRSTBLG,FRQRPRG $ DELETE /R1VALRG,R2VALRG,CVALRG,wmidg,r3valrg $ DELETE /RESP3RG,rr2idrg,wgtmg,RQATABRG,modrprg $ DELETE /DSCMG,DELW,DELV,DELB,DELF $ endif 1-41392631 2007.0.0 SOL 600 - MARCPRNH and MARCPR99 cannot correctly converted In SOL600, MARCPRNH and MARCPR99 cannot BE correctly converted to PRINT ELEMENT by the CQUAD4 element. For example the input cquad4.bdf with: --- param,marcprnh,1 param,marcpr99,-1 --- cquad4.marc.dat: --- PRINT ELEMENT STRESS CAUCHY STRAIN PLASTIC PR999999 1 to 1 1 to 1 --- This input file outputs only point1 and layer1. 1-41458341 2007.0.0 File Management Section - APPEND - Output datablock via dbview fails The following simple deck fails in gnfist die to an append datalock geom1 that using the alias dbviewname VIEW0001 ASSIGN inputt2='../complex_107.op2',unit=25 status=old acquire nddl sol mysol compile mysol list subdmap mysol TYPE PARM,,I,Y,IUNIT=25 $ INPUTT2 /,,,,/-1/IUNIT $ TYPE DB GEOM1,GEOM168 $ FILE GEOM1=APPEND $ DBVIEW VIEW0001=GEOM1 WHERE(AUXMID=1;AFPMID=2;DESITER=3;HIGHQUAL=0) INPUTT2 /GEOM168,,,,/0/IUNIT////////'GEOM168' $ MAKENEW GEOM168,,,,/VIEW0001,,,,/'GEOM1'/////'GEOM1' $ IFPINDX /VIEW0001 $ DBDIR $ end cend *** USER FATAL MESSAGE 1126 (GNFIST) DMAP MODULE IFPINDX IS ATTEMPTING TO OUTPUT DATABLOCK 201 NAME = GEOM1 , WHICH ALREADY EXISTS. USER ACTION: DELETE THE DATABLOCK, OR USE FILE GEOM1=OVRWRT 1-41729401 2007.0.0 Documentation/SOL 600 - MARCOUT Defalt The code was revised to make the default t16 file smaller, but the QRG was not revised accordingly. MD Nastran Quick Reference Guide ch.8 MARCOUT Remarks 6, - 6. If this entry is not used, the following defaults are entered automatically: E1, E11, E18, E27, E301, E321, E341, E401, E47, E48, N1, N2, N5, N35, N37, N39. In MD Nastran 2006, if this entry is not used, only E301, E341, E47, N1, N2, N35, N37, N38 and N39 are output. 1-41737627 2007.0.0 SOL 600/Documentation - QRG incorrect for PARAM, MRFOLLOW The electronic 2005r3 QRG indicates: MRFOLLOW Integer, Default = 0, MSC.Nastran Implicit Nonlinear (SOL 600) only MRFOLLOW=1, FORCE1, FORCE2, MOMENT1, MOMENT2 will act as non-follower forces. This option must be entered if MSC.Marc versions prior to 2003 r1 are to be used. MRFOLLOW=0, follower forces entered using FORCE1, FORCE2, MOMENT1, MOMENT2 will be mapped to MSC.Marc’s new follower force option available starting with MSC.Marc version 2003r1. MRFOLLOW=-1, follower forces will be turned off even if requested to be on using other options. This is sometimes necessary for multiple load cases where pressures are applied to different elements in the different load cases. ========================================== This is incorrect, the doucmentation should indicate: ========================================== MRFOLLOW Integer, Default = 0, MSC.Nastran Implicit Nonlinear (SOL 600) only MRFOLLOW=0, FORCE1, FORCE2, MOMENT1, MOMENT2 will act as non-follower forces. This option must be entered if MSC.Marc versions prior to 2003 r1 are to be used. MRFOLLOW=1, follower forces entered using FORCE1, FORCE2, MOMENT1, MOMENT2 will be mapped to MSC.Marc’s new follower force option available starting with MSC.Marc version 2003r1. MRFOLLOW=-1, follower forces will be turned off even if requested to be on using other options. This is sometimes necessary for multiple load cases where pressures are applied to different elements in the different load cases. ========================================== The actual action is opposite what the manual indicates. 1-41808171 2007.0.0 Aeroelastic Analysis - FLFACT - Floating point exception, divide by zero, subroutine FASORT A divide by zero fatal message will occur for the pk or pknl methods when two consecutive dynamic pressures (= 1/2 density * velocity**2) are the same. User Avoidance: adjust the flfact input for densities and velocities to avoid the duplicate dynamic pressure. 1-41822251 2007.0.0 Documentation/QRG - SOL 600 - PARAM,MARCT16 output The description of Quick Reference Guide(MARCT16) is not correct. It has been updated for the MD Nastran R2 and MSC.Nastran 2007 releases. 1-41822276 2007.0.0 Documentation/QRG - SOL 600 - MARCT16 document The QRG needs to be modified to indicate that MARCOUT should not be entered when PARAM,MARCT16 is used. 1-41832822 2007.0.0 Documentation/QRG - SPC Case Control Command missing reference to SPC2 The QRG description of the SPC case control command needs to be updated to indicate that it can reference a SPC2 bulk data entry. 1-41883251 2007.0.0 Matrix Methods - SMETHOD - CASI solver on ILP-64 AIX Fails The element based iterative solver (CASI) will fail on IBM AIX platforms with a Traceback similar to the one reproduced below. Note that this only occurs with the ILP-64 version ("mode=i8"). Avoidance: A possible workaround is to rerun the job with mode=i4. Signal received: SIGSEGV - Segmentation violation Traceback: Offset 0x00000030 in procedure casiFree Offset 0x00000054 in procedure deleteFMatrix_159_128 Offset 0x00000ca4 in procedure pcgsl Offset 0x000003d0 in procedure pcglss Offset 0x00000078 in procedure xqcasi_ Offset 0x00001010 in procedure itscas_ Offset 0x000008d8 in procedure solvit_ Offset 0x00000670 in procedure xsem07_ Offset 0x00000198 in procedure xsemdr_ Offset 0x00000064 in procedure nastmon_ Offset 0x00000a78 in procedure mainal --- End of call chain --- 1-41884101 2007.0.0 Matrix Methods - SMETHOD - CSEAM element not recognized by Casi solver The casi element-based iterative solver interface does not currently allow the CSEAM element. 1-41927461 2007.0.0 Documentation/QRG - SOL 600 - BRKSQL Remarks 7 Remark 7 of the Quick Reference Guide BRKSQL entry description has been updated to reflect the actual code. In MD Nastran 2006, if METH =1 and ICORD=0 or 1, an MSC.Marc t19 file will automatically be produced. If METH =0 and ICORD=0 or 1, an MSC.Marc t19 file not generate. 1-42002359 2007.0.0 Loading - TEMP(INIT), TEMP(BOTH) - Wrong Answers Specifying both TEMP(INIT) and TEMP(BOTH) can lead to wrong answers as follows: 1) if TEMP(INIT) and TEMP(BOTH) -> WRONG ANSWERS (Delta-T is wrong) 2) TREF will NEVER perform table lookup 3) If TEMP(BOTH) and TEMP(LOAD) are specified, only the last request will be honored 4) In SOL 105 ONLY: If TEMP(BOTH) is specified in the subcase, it does not perform This condition will be trapped with a new fatal message beginning in version 2007. USER FATAL MESSAGE XXX (IFP1D) MULTIPLE TEMP(BOTH) OR COMBINATIONS OF TEMP(BOTH) AND TEMP(INIT) ARE NOT ALLOWED. 1-42059473 2007.0.0 Matix Methods - SMETHOD - Casi iterative solver does not support K2GG input matrix data The casi element-based iterative solver interface does not does not support K2GG input matrix data. 1-42074241 2007.0.0 Documentation/QRG - DOBJ2, DPMIN, FSDALP and FSDMAX missing from DOPTPRM description The 2005R3 QRG inadvertantly deleted fsdmax, fsdalp,dobj2 and dpminfrom the DOPTPRM entry. User Information: Refer to the 2005 Quick Reference Guide for information on these parameters. 1-42173286 Postprocessing - POST TOFILE - op2 file cannot be read by Patran When POST TOFILE is being used to control the results output to the op2 file, Patran or other postprocessors cannot read the geometry. This error is caused by CASECCR being written before the results are output. Avoidance: Apply the following dmap alter: compile super3 $ alter 'OUTPUT2.*CASECCR','' $ 1-42236771 2007.0.0 SOL600 - MATHE-K, MOONEY - Not correctly converted The values of MATHE-K(bulk modulus) is converted to the "coefficient of thermal expansion" and the "bulk modulus" of MOONEY in the marc data. The value of MATHE-Texp is disregarded. [nastran.dat] MATHE, 2, MOONEY, , 33.0E6, .0003, 2.5E-6, 70.5, .015 , 5.5E6, 1.0E6 , 4.0E6, 0.8E6 , 2.0E6 [marc.dat] MOONEY COMMENT, MATHE 2 1, 2, 5.500000E+06, 1.000000E+06, 7.770000E-07, 3.300000E+07, 8.000000E+05, 4.000000E+06, 2.000000E+06, 3.300000E+07, 1-42267596 2007.0.0 DMAP - MATPRN, DUPVH - SFM 4276, EC 3021 The following MATPRN caused EMG to fail with SYSTEM FATAL MESSAGE 4276 (READ) ERROR CODE 3021 compile NONLIN LIST alter 'EMG .*ESTNL.*UGNIX'(1,-1) $ MATPRN DUPVH// $ 1-42425397 2007.0.0 Nonlinear Analysis - CGAP, TEMP(LOAD) - UFM 2044, UFM 3007 Nonlinear analysis with CGAP entries and TEMP(LOAD) may fail with the following fatal messages: *** SYSTEM INFORMATION MESSAGE 6916 (DFMSYN) DECOMP ORDERING METHOD CHOSEN: BEND, ORDERING METHOD USED: BEND *** USER FATAL MESSAGE 2044 (NLCOMB) UNDEFINED TEMPERATURE SET 104 *** SYSTEM FATAL MESSAGE 3007 (NLCOMB) ILLEGAL INPUT TO SUBROUTINE NLCETT 0FATAL ERROR 1-42507691 2007.0.0 SOL 600 - SPC Incorrectly Translated In Versions 2005.5.2 and 2006.0.1 enforced displacement of SPC is not translated correctly, though SPCD+SPC is correctly translated. 1-42792171 2007.0.0 SOL 700 - CBEAM - Orientation vector is not correctly translated if BEAMOR is used to define it. If BEAMOR is used to define the orientation vector of a CBEAM, it is not translated correctly. For example, BEAMOR 1. 0. 0. CBEAM 1910002 1000009 1000044 1000045 are translated as shown below into dyt.dat file. CBEAM 1910002 1000009 1000044 1000045 1.00000 and it casuses Error in dytran-lsdyna as shown below. %E *** ErrorMD Nastran Input 153733 CBEAM 1910002 1000009 1000044 1000045 1.00000 invalid integer (1.00000) in field 6 The correct translation for this example is: CBEAM 1910002 1000009 1000044 1000045 1.00000 0.0 0.0 1-42803355 2007.0.0 Matrix Methods - DECOMP, METIS - UFM 6413 (DFMSYN) If metis reordering is specified by sys206=8, and more than 2gb of memory is given, the job will always fail with a message similar to: *** USER FATAL MESSAGE 6413 (DFMSYN) INSUFFICIENT MEMORY FOR THE SYMBOLIC PHASE OF SPARSE DECOMPOSITION (METIS). USER ACTION : INCREASE MEMORY BY AN ESTIMATED 524680801 WORDS. 1-42936851 2007.0.0 Optimization/Design Sensitivity - DRESP1 - SFM 3007 (DSAD) writh RMS or PSD responses The message "SYSTEM FATAL MESSAGE 3007 (DSAD)" can occur when a SOL 200 job contains DREPSP1 entries with RTYPE = PSDXXX and/or RMSxxx and all of the computed response are zero. This typically indicates an error in the user input. Avoidance: Review the specification of the DRESP1 entries with RTYPE = PSDxxxx and/or RMSxxxx to make sure that the requested quantities are non-zero. 1-43084871 2007.0.0 SOL 600 - CHEXA (with PSOLID and PLSOLID) is not correctly converted If a model contains CHEXA elements that reference PSOLID entries and others that reference the PLSOLID entry, the input is not translated correctly and the following message is issued: --error message-- ** Severe Warning in SOL 600 trnslator - CHEXA must have 8 or 20 nodes to be used in SOL 600 1-43300972 2007.0.0 Nonlinear Analysis - RFORCE and LOAD(with negative scale factor) - NaN or UFM 4296 (NQD4D) If a SOL 106 run has a LOAD bulk data that references an RFORCE entry with a negative scale factor, either a "NaN" will occur during the first nonlinear iteration or a fatal message will be issued for every quad element in the model. Avoidance: Avoid negative scale factors on the LOAD entry for the RFORCE id. 1-43365991 2007.0.0 SOL 600 - CBUSH1D element is not supported in SOL 600 The CBUSH1D entry is not supported by SOL 600. This entry may cause the sol 600 translator to crash without issuing an error message. 1-43448777 2007.0.0 Nonlinear Transient Analysis - CMS - UFM 1126 Module UMERGE1 outputtingt GOA which already exists In V2005+, a SOL 129 model with CMS (QSETi) may fail with: *** USER FATAL MESSAGE 1126 (GNFIST) DMAP MODULE UMERGE1 IS ATTEMPTING TO OUTPUT DATABLOCK 201 NAME = GOA , WHICH ALREADY EXISTS. USER ACTION: DELETE THE DATABLOCK, OR USE FILE GOA=OVRWRT Also, if there are unused QSETi degrees-of-freedom then the run may fail with: *** USER WARNING MESSAGE 6137 (UDSNRD) INPUT MATRIX IS RANK DEFICIENT, RANK = 24 USER ACTION: CHECK MODEL 0*** UNSUCCESSFUL MATRIX DECOMPOSITION. *** SYSTEM FATAL MESSAGE 3007 (NLTRD2) ILLEGAL INPUT TO SUBROUTINE NLTRD2 Avoidance: Insert PARAM,SMALLQ,1.e-10 and the following DMAP alter compile nltran alter 'call segoa'(,6) delete /goa,,,, $ 1-43465661 2007.0.0 SOL 600 - MATF - Incorrect maximum compressive stress failure criteria The maximum compressive stress in the x and y directions as specified on the MATF entry which defines the failure criteria are translated incorrectly. 1-43537471 2007.0.0 Aeroelastic Analysis - SPLINE6 - UWM 328, Excessive Output SPLINE6 entries that do no explicitly define a continuation entry will produce UWM 328 for each entry. This may lead to excessive output. 1-43604151 2007.0.0 Nonlinear Analysis - UWM 3057, UFM 3007, for MATS1, NLELAST, TABLES1 w/o origin defined For a material nonlinear run with MATS1 NLELAST and TABLES1, if the origin is not defined on the TABLES1, then the model may go several decomps before finally failing with the following message. *** USER WARNING MESSAGE 3057 (NLITER) MATRIX IS NOT POSITIVE DEFINITE. *** SYSTEM FATAL MESSAGE 3007 (NLITER) ILLEGAL INPUT TO SUBROUTINE NLINIT This should be trapped as a user input error. The rule is defined in the Quick Ref Guide, MATS1 Remark 3, bullet 2. 1-43735881 2007.0.0 SOL 600 - BCMOVE limitation when BCTABLE with 2 contacts The BCTABLE implementation in Nastran Version 2006.0.1 and earlier releases does not allow no contact to be defined between a rigid and flexible body. This prevent modelling the contact force from being released gradually. 1-43778881 2007.0.0 Transient Analysis - GPSTRESS, SORT2, PUNCH - Incorrect time label. Punched GPSTRESS SORT2 output from a transient analysis may contain incorrect time labels. 1-43779004 Response Spectrum Analysis - RSPECTRA, MEFFMASS - UFM 6144 (MERGE1) Response spectrum analysis stopped with UFM 6144 (MERGE1) when the client used "MEFFMASS" entry. Nastran issued following messages. Ver2001,2004 and 2004r2 was no problem with same input data(I attached the input data.). Let me know if it has some avoidance. Thank you for your help. *** SYSTEM FATAL MESSAGE 6144 (MERGE1) THE SIZES OF THE INPUT MATRICES AND PARTITIONING VECTORS ARE INCOMPATIBLE. SPECIFICALLY: The number of rows in EFMFACS0 is not equal to the number of non-zeros in partitioning vector PARTVECS User Information: Size of input matrices and partitioning vectors: EFMFACS0: Rows= 4 by Cols= 123 PARTVECS: Rows= 6 no. of non-zeros= 6 no. of zeros= 0 0FATAL ERROR 1-43834511 Optimization/Design Sensitivity - Missing output for DSAPRT(export,by=1) When keywords export and by=n (n>1) are specified on a DSAPRT Case Control command , the sensitivity data only at the last design cycle is output to the external binary file. Avoidance: use the following dmap alter compile exitopt list alter 'OUTPUT2 D'(1,-1) type parm,nddl,i,n,zuzri1 $ if ( dsexport) then MESSAGE //'DESIGN CYCLE = '/DESCYCL1 $ if (zuzri1 = 0) then MESSAGE //'=========DSAPRT START =========='/ $ $ alter 'OUTPUT4 D' $ $ Output DSCMCOL And DSCM2 : $ OUTPUT2 DSCMCOL,,,,//-1/IUNIT/'DSCMCOL'/OMAXR $ $ OUTPUT4 DSCM2,,,,//-2/IUNIT $ zuzri1 = 1 else if (zuzri1 >0 and not(lastcyc)) then MESSAGE //'=========DSAPRT MIDDLE =========='/ $ OUTPUT2 DSCMCOL,,,,//0/IUNIT/'DSCMCOL'/OMAXR $ OUTPUT4 DSCM2,,,,//0/IUNIT $ else if (zuzri1 >0 and lastcyc) then MESSAGE //'=========DSAPRT END =========='/ $ OUTPUT2 DSCMCOL,,,,//0/IUNIT/'DSCMCOL'/OMAXR $ OUTPUT4 DSCM2,,,,//-2/IUNIT $ endif $ For QA test, need to ensure the sensitivity data is written out at each design cycle (or given design cycle). 1-43943079 2007.0.0 Matrix Methods - Singular matrices not flagged with param,asing,-1 PARAM,ASING,-1 should cause nastran to issue a fatal message when a singular matrix is encountered. However, when param,asing,-1 is specified along with autospc=no, a high max ratio may be reported but the execution continues and the offending degrees of freedom are not identified. 1-44040531 Documentation/SOL 700 User's Guide - Description of WALL/WALLGEO is incorrect. The description of WALL/WALLGEO in the SOL 700 User's Guide is incorrect. Avoidance: Refer to the MD Nastran Quick Reference Guide for the correct description. 1-44225671 2007.0.0 Documentation/QRG - RBE2/RBE2D description is wrong. The remark 4 of the RBE2D entry in the Quick Reference Guide shown below is wrong. RBE2 without the FULLRIG option can simulate rotations. 4. For modeling a nodal rigid body that undergoes rotations, use RBE2D or RBE2 with the FULLRIG option. Note that the RBE2 without the FULLRIG option can not properly simulate rotations. 1-44398314 Nonlinear Analysis - LOAD, Follower Forces, PLOAD4 - Wrong Answers In SOL 106, if a LOAD bulk data entry references a follower-force type loading (e.g. PLOAD4) and another load set ID that does not exist, AND that missing load set ID is LOWER than the follower force load set's ID, then the follower force loading acts like a linear load and no longer follows the structure as it deforms. If the missing load set ID is higher than the follower force's, then the follower force acts as expected. 1-44744683 2007.0.0 Optimization/Design Sensitivity - MATT9 - Wrong Answers MATT9 entries may lead to incorrect sensitivities. 1-44900725 2007.0.0 Installation - Unable to find library 'liblapi.sl'. In Version 2005.5 with a MINIMUM install, the following error might occur when attemting to run Nastran: MSC.Nastran started Wed Oct 25 09:50:52 CDT 2006 /usr/lib/pa20_64/dld.sl: Unable to find library 'liblapi.sl'. time: command terminated abnormally. Avoidance: Select a full install 1-45244091 2007.0.0 SOL 700 - PSHELLD* causes UFM 6440 (MODGM2) PSHELLD* causes the fatal error shown below in the MODGM2 module. *** USER FATAL MESSAGE 6440 (MODGM2) ELEMENT 2620121 REFERS TO AN INVALID PROPERTY ENTRY. USER ACTION: SPECIFY APPROPRIATE PROPERTY ENTRY. 0FATAL ERROR Avoidance: Use the dmap alter below to skip MODGM2 module in SOL 700 MODGM2 IEPT.11,IGEOM2.6,IGEOM1.5,IGEOM4,XBGPDT,XCSTM,IMPT.10/ IGEOM2,IGEOM1Q,IGEOM2DL,IEPT,IMPT/ /OSWPPT/OSWELM/S,N,NSWPPT/S,N,NSWELM/S,N,SWEXIST/ S,N,NOGOMGM2/S,N,RGDEXIST/RIGID/ORIGID/S,N,NLRIGID/ LMFACT/PENFN/NONLNR/CWRANDEL/CWDIAGP/CFRANDEL/CFDIAGP/ S,N,NOEPT/S,N,NOMPT/S,N,NOGEOM1/SOFFSET/CSRANDEL/CSDIAGP $ IF ( NOEPT=-1 ) EQUIVX IEPT.11/IEPT/-1 $ IF ( NOMPT=-1 ) EQUIVX IMPT.10/IMPT/-1 $ IF ( NOGEOM1=-1 ) EQUIVX IGEOM1.5/IGEOM1Q/-1 $ IF ( NOGO=0 ) THEN $ IF ( NOGOMGM2 ) THEN $ SPOT WELD CHECK RUN CALL PRTSUM /RSTJUNK $ EXIT $ ENDIF $ NOGOMGM2 ENDIF $ NOGO=0 1-45244156 2007.0.0 SOL 700 - MATD003 can not be used with PBEAMD ELFORM=3 while it is allowed in LS-Dyna. MATD003 can not be used with PBEAMD ELFORM=3 else the following error is output to the dytr_prep.out file. %E *** Error MD Nastran Inpu pbeamd (= 999999) references wrong material type (= 3) This combination is allowed in LS-Dyna. 1-45605382 2007.0.0 Complex Eigenvalue Analysis - CEAD Module UFM 1126 A complex eigenvalue analysis to obtain Campbell diagram data with the keyword scr=no, fails in the CEAD module with: *** USER FATAL MESSAGE 1126 (GNFIST) DMAP MODULE CEAD IS ATTEMPTING TO OUTPUT DATABLOCK 202 NAME = CLAMA , WHICH ALREADY EXISTS. USER ACTION: DELETE THE DATABLOCK, OR USE FILE CLAMA=OVRWRT Avoidance: Set scr=yes and the attached job runs. Or insert the following DMAP alter: compile sedceig alter 'if ( gcase\>1 or ccase\>1','' if ( gcase>1 or ccase>1 or ddval>0. ) delete /CPHD,CLAMA,lphd,, $ alter 'call super3' if ( ddval>0. ) dbdelete datablk=(cphp,cphqp,cphqmp) , where(wildcard) $ 1-45630951 2007.0.0 Miscellaneous - RSEQCONT or SYSTEM(357) - UFM 4392 The NASTRAN keyword RSEQCONT=1 (a.k.a. NASTRAN SYSTEM(357)=1) should cause Nastran to ignore all continuation fields. This is very advantageous when you have lots of input decks or pieces of decks coming from different sources. For the most part this feature works. However, in the case where this is an include file, for instance a file with only MAT8’s and the second row is missing (generally left out as it is often all zero) Nastran will issue a USER FATAL MESSAGE 4392. To message doesn’t list the MAT8. It will indicate the problem is with the next entry it finds.. Avoidance: Explicitly enter the continuation entry as follows MAT8 1020009 2.0600+71.1300+6.340000 580000.00.0 0.0 .0 +000001 MAT8 1020014 1.0030+71.0000000.0 1.0000000.0 0.0 .0 +000003 needs to be changed to: MAT8 1020009 2.0600+71.1300+6.340000 580000.00.0 0.0 .0 +000001 +000001 0.0 0.0 MAT8 1020014 1.0030+71.0000000.0 1.0000000.0 0.0 .0 +000003 1-46023346 Documentation/QRG - MONPNT3 erro in QRG Remark 1 of MONPNT3 description in the Nastran QRG indicates that MONPNT3 is available for SOL 101, 144 and 146. This is incorrect, SOL 146 is not supported. 1-46027846 2007.0.0 SOL 700 - DYPARAM,LSDYNA,HOURGLASS is not effective. DYPARAM,LSDYNA,HOURGLASS to control hourglass is not effective in SOL 700. Avoidance: Use PARAM,DYHRGIHQ or DYHRGQH 1-46027901 DMP - DMIG, K2GG, M2GG - UWM 2070 (MTRXIN) DMIG input is not supported in the geometric domain decomposition methods for statics and normal modes. This may result in jobs failing with the messages: "*** USER WARNING MESSAGE 2070 (MTRXIN) REQUESTED DMIG MATRIX xxx IS UNDEFINED." "*** USER WARNING MESSAGE 3204 (SSG1) UNABLE TO FIND SELECTYED SET ( 311) IN TABLE (SLT ) IN SUBROUTINE (SSG1)." "***** RECEIVED SIGNAL 15, ABORTING ..." Avoidance: The avoidance is to use the matrix-based domain decomposition methods by using one of the following Executive Section entries: DOMAINSOLVER STAT (PARTOPT=DOF) for statics or DOMAINSOLVER MODES (PARTOPT=DOF) for normal modes 1-46077402 External Superelements - EXTSEOUT - SFM 3200 BDYINF, Multiple SPC sets If EXTSEOUT is requested with multiple SPCs, then only the 1st set of SPCs will be used to generate KAA/PA and the following fatal message will be issued: *** SYSTEM FATAL MESSAGE 3200 (BDYINF) LOGIC ERROR 29 ENCOUNTERED IN SUBROUTINE BDYINF USER ACTION: REPORT THIS PROBLEM TO MSC CLIENT SUPPORT 0FATAL ERROR Avoidance: Reduce each SPC set one at a time.: This error may occur for MPC sets. 1-46095441 2007.0.0 SOL600 - BEAM SECT Error If Type=ROD of the pbarl/pbeaml is specified, the Effective transverse shear area (As) in the tranlated marc.dat input is wrong. The marc default is As=A, the nastran default is 0.857*A(K=0.857 for MDLPRM,TWBRBML,0) or 0.9*A(K=0.9 for MDLPRM,TWBRBML,1). The effective transverse shear area of marc.dat is 0.5*A. The effective transverse shear area for marc is not the reference area factor for shear(nastran) of the pbar/pbeam entry. If SOL600 data is K=0.9, marc.dat should be As=K*A=0.9*A. However, the effective transverse shear area of marc.dat is As=0.85*A. 1-46245072 2007.0.0 Fluid Analysis - Wrong effective mass values with MFLUID and PARAM,VMOPT,2 Modal effective mass output will be wrong with PARAM,VMOPT,2 Avoidance: Insert the following DMAP alter: COMPILE MODEFSRS ALTER 'EFFMAS.*PHSA,LAMAS,'(,-1) IF ( VMOPT=2 AND NOCEE>0 ) THEN $ CALL MGEN1 GGE ,USET ,GM ,GOt ,goq , BEC ,CEE ,MEE ,MEC ,msaa / MEA ,MvAA / 0 /NOA /0 /false $ EQUIVX MvAA/MSAA/-1 ENDIF $ 1-46507156 2007.0.0 DBCMSG ERROR MESSAGE 22, SUBROUTINE ELMBLK - Missing quad4 element stress/strain in xdb The stress and strain output for quad4 elements may be missing from the xdb when CFAST elements are present. The message: DBCMSG ERROR MESSAGE 22 FROM SUBROUTINE ELMBLK is a clue that this condition may exist. 1-46561555 2007.0.0 Topology - CASI solver performance issue when TOPVAR bulk data is present SOL 101 SOLVIT module performance is adversely affected when the casi element-based iterative solver is selected and 1) the model contains a large number of 10-node TET elements to be processed in implicit mode (the SOLVIT default) 2) a TOPVAR bulk data entry is present that selects a PSOLID entry that is referenced by CTETRAs An avoidance is to remove the TOPVAR bulk data entries. 1-46581521 2007.0.0 Documentation/QRG - PARAM,AUTOMSET In QRG of MD 2006r1, the description of PARAM,AUTOMSET is "MD Nastran Implicit Nonlinear (SOL 600) only". The description for other supported solutions such as SOL 101, is missed. 1-46592824 2007.0.0 Aeroelastic Analysis - SOL 146 fails with UFM 5423 when several subcases are defined When several subcases are defined in SOL 146, Nastran fails with the message UFM 5423: *** USER FATAL MESSAGE 5423 (SADD5) ATTEMPT TO ADD INCOMPATIBLE MATRICES FILE# COLS ROWS FORM TYPE NZWD DENS TRAILER FOR SCRATCH IS 306 81 11 2 3 22 9877 TRAILER FOR PHF IS 110 162 11 2 4 0 0 0FATAL ERROR 1-46823292 2007.0.0 Postprocessing - Nastran system(143)=2, POST=0 - incomplete xdb If Nastran system(143) is specified along with PARAM,POST,0 then the resulting xdb file will be incomplete. Specifically CSTM, EPT and MPT tables are not written to the xdb. Avoidance: Insert the following DMAP alter: compile phase0 alter 'if ( (POST=0 or post','' if ( (POST=0 or post=-1) and not(disofp and (parfreq or segmode) and procid>1) and (not(sebulk) or ((disofp or procid<=1) and (pardcmp or acms))) ) then $ endalter 1-46863451 External Superelements - fluid, structure, dmigpch - ufm 7556 If the dmigpch option is used in the reduction run for the external superelement with fluid/structure coupling, then the job will fail in the assembly run with the following messages: *** USER FATAL MESSAGE 7556 (SEPMAB) *** USER FATAL MESSAGE 7556 (SEPMAB) SUPERELEMENT xx HAS BOTH INTERIOR STRUCTURAL AND FLUID GRID POINTS. USER ACTION: REASSIGN GRID POINTS AND/OR ELEMENTS SO THAT THE SUPERELEMENT'S INTERIOR AND EXTERIOR POINTS ARE ALL FLUID OR ITS INTERIOR POINTS ARE ALL STRUCTURAL. Avoidance: Use one of the other methods (e.g., dmigop2) 1-46890597 2007.0.0 External Superelements - CSUPER, DBLOCATE - SFM 3001 (MODACC) If data recovery is attempted on an external superelement using the old-style CSUPER type superelements and DBLOCATE with the user parameter EXTRCV, the solution will fail with the following message: *** SYSTEM FATAL MESSAGE 3001 (MODACC) THE INPUT DATA BLOCK NAMED OL1M AT POSITION 2 DOES NOT EXIST. USER INFORMATION: THIS ERROR IS CAUSED BY ONE OF THE FOLLOWING: 1. THE DATA BLOCK WAS NEVER CREATED. 2. THE DATA BLOCK WAS DELETED. 3. THE DATA BLOCK IS SPECIFIED ON A TYPE STATEMENT IN THE CURRENT SUBDMAP OR A HIGHER SUBDMAP BUT THE CURRENT QUALIFIER VALUES DO NOT MATCH THE QUALIFIER VALUES OF THE DATA BLOCK(S) ON THE DATABASE. USER ACTION: 1. IF YOU ARE EXECUTING AN MSC.SOFTWARE SUPPLIED SOLUTION SEQUENCE AND NOT USING THE ALTER EXECUTIVE CONTROL STATEMENT, THEN CHECK FOR BULK DATA AND/OR CASE CONTROL INPUT ERRORS. IF NO ERRORS CAN BE FOUND, THEN CONTACT MSC.SOFTWARE CLIENT SUPPORT. 2. IF YOU ARE EXECUTING A DMAP PROGRAM NOT SUPPLIED BY MSC.SOFTWARE, THEN FOR DEBUGGING PURPOSES INSERT STATEMENT DIAGON(20) BEFORE THE MODULE SHOWN ABOVE. PROGRAMMER INFORMATION: THE FIST NUMBER IS 102 AND SUBROUTINE IS MODAC1 0FATAL ERROR This is because the LAMA datablock which is DBLOCATEd from the system (AKA assembly or residual) run is not equivalenced to the LAMA2 datablock in SEMODES Avoidance: Insert the following DMAP COMPILE SEMODES NOLIST $ ALTER 'PERFORM PHASE III'(,-2) $ ELSE IF (EXTRCV>0) THEN $ EQUIVX LAMA/LAMA2/-1 $ ENDALTER $ 1-47123337 External Superelements - EXTSEOUT- Statics with TEMP(LOAD), OTM stresses incorrect in assembly In static analyis, for EXTSEOUT external superelements, the assembly run with TEMP(LOAD) will give incorrect results for Forces and Stresses on the external superelement. NOTE: The residual is correct and the displacements/spcforces on the external are correct. Avoidance: : Use PARAM,EXTDROUT and use the 3-step process to recover stresses in the upstream superelements. 1-47128492 2007.0.0 Matrix Methods - SMETHOD - SFM 4276 (unpack) ec 20 - casi solver When no loads are applied in a linear statics sol 101 run and the casi solver is requested (smethod = element) then the following error message appears in the f06: *** SYSTEM FATAL MESSAGE 4276 (UNPACK) ERROR CODE 20 PID= 0 and the job fails. A more meaningful message should be output. 1-47162343 2007.0.0 DMP - DOMAINSOLVER STAT(PARTOPT=DOF) with SUPORT and INREL - SFM 4276 in MRGCOL SYSTEM FATAL MESSAGE 4276 in MRGCOL may be issued with SUPORT and interial relief if used with DMP. Avoidance: compile sekrrs alter '216'(,-1) return $ compile sestatic alter 'call phase1c','' IF ( nouset>0 and NOT(NONLNR) and ( (dbrcvx='scratch' and not(pardcmp)) or (pardcmp and hlpmethx=1 and procid=1) ) ) CALL PHASE1C KAA ,USET ,SILS ,EQEXINS ,maa , PA , , ,cases , , drg , , , / KLL ,LLL ,Dar ,ULL ,mr , mlr ,mll ,mrr ,QR ,PL , LRSEQ , , , , , / SEKR /SEMR /SELR /NOLSET /NORSET / true /ERROR /FIXEDB /NOQSET /NOTSET / ' '/TRUE /false /false /noa / false $ alter 'or mdstat','' if ( not(gmconv or pardcmp or NONLNR) ) then $ alter 'call statrs','' IF ( NOT(NOPH2) AND (NOLLL>0 OR GETSYS(NP,216)<>0 or mdstat) ) CALL STATRS LLL ,KLL ,PL ,GPLS ,USET , SILS ,QR ,MATPOOL ,EQEXINS ,DRG , Dar ,GM ,GOT , ,LRSEQ , EDTs ,CASES ,kgg ,kaa ,kelm , kdict ,bgpdts ,cstm ,mpts ,dit , est ,ys ,pg1 / UL ,CRX ,PRECON ,epssex ,qrg / IRES /INREL /LUSETS /NSKIP /NLOADS / NORSET /NOA /PVALOLD /ADPTINDX/PEXIST / s,nullsol $ compile statrs alter 'fbs '(2),'' call mdstat kll,pli,uset,sils,eqexins/ ul $ putsys(0,307) $ mergeofp=no -- deactivate distributed data recovery endalter 1-47165003 2007.0.0 Installation - Support for Windows Vista The Nastran Windows Installshield based installation does not recognize the Windows Vista operating system and will not allow the installation to continue. Avoidance: The only avoidance is to install nastran on another machine and copy the installation to the Vista machine. Beyond the installation nastran has been certified on Windows. 1-47167162 2007.0.0 Miscellaneous - Linux RHE 4.0 - No Time Information - time: error waiting MSC.Nastran or MD Nastran V2006.0.1 or ealier on linux, linux8664, or linux64 with the RHEL4 or SuSE 10 (or later) distribution, may have the following message in the log file: time: error waiting for child process: No child processes The results of the analysis are correct. The timing information which is normally in the log file is not there though. Avoidance: Upgrade the "nastran" driver in $MSC_BASE/$MSC_VERSD/$MSC_ARCH to a newer version of Nastran. 1-47218191 2007.0.0 Acoustics - PANEL PARTICIPATION FACTORS are not correct if a panel does not coupled with fluid ACOUSTIC PANEL PARTICIPATION FACTORS are not correct if one of panels does not coupled with fluid as shown below. Example: 4 panels are defined and a panel named "bottom" does not coupled with fluid. Then PANEL RESPONSE of "bottom" should be zero, but it is not zero, and "top_right" does not appear. panel,bottom,2 panel,top_left,3 panel,top_cent,4 panel,top_righ,5 set1,2,16367,thru,16402 set1,3,16246,thru,16300 set1,4,16301,thru,16311 set1,5,16312,thru,16366 A C O U S T I C P A N E L P A R T I C I P A T I O N F A C T O R S GRID POINT = 37, TOTAL RESPONSE (R/I) = 3.05356E+00 / -5.16123E-03, (M/P) = 3.05357E+00 / 359.90 LOAD FREQUENCY = 1.00000E+00, (SUBCASE 1, DLOAD = 4) MAXIMUM PANEL RESP = 1.85680E+00 FOR PANEL = TOP_LEFT, SORTKEY = PANEL , SORT = ALPHANUMERIC , FILTER = 1.00000E-16 PANEL NAME PANEL RESPONSE PANEL RESPONSE PROJECTION REL. PANEL SCALED RESPONSE REAL IMAGINARY MAGNITUDE PHASE MAGNITUDE PHASE FRACTION MAGNITUDE BOTTOM 5.98452E-01 9.71956E-03 5.98531E-01 0.93 5.98435E-01 -358.97 1.95979E-01 3.22294E-01 TOP_CENT 5.98473E-01 9.71655E-03 5.98552E-01 0.93 5.98456E-01 -358.97 1.95986E-01 3.22305E-01 TOP_LEFT 1.85664E+00 -2.45973E-02 1.85680E+00 359.24 1.85668E+00 -0.66 6.08035E-01 9.99933E-01 1-47236346 2007.0.0 Acoustic Analysis - CACINF4 - External Acoustics may yield incorrect Panel Participation Factors External Acoustics may yield incorrect Panel Participation Factors 1-47324010 2007.0.0 Optimization/Design Sensitivity - EXTSEOUT, ANAL=DFREQ or MFREQ - UFM 9104 If an external superelement is attached in a SOL 200 run with ANALYSIS=MFREQ or DFREQ then the job may fail with: ^^^ USER FATAL MESSAGE 9104 (SELR) ^^^ THERE IS NO COUPLING BETWEEN INTERIOR POINTS OF THIS SUPERELEMENT AND ITS BOUNDARY POINTS. Avoidance: Insert the following DMAP alter: compile feafreq alter 'sep2dr slist,emap','' sep2dr slist,emap// s,n,seid/s,n,peid/s,n,sedwn/s,n,lpflg///// s,n,noup/s,n,scndry/s,n,extrn//'all'//-1/ s,n,partse/s,n,setype/s,n,reid $ alter 'if ( seid\>0','' if ( seid>0 and not(extrn<>-1 or setype='extrna') ) then $ 1-47688921 2007.0.0 Documentation/QRG - SOL 600 - PARAM,MARCSPCD does not work PARAM,MARCSPCD is documented in the QRG but is not supported. The documentation will be removed. 1-47711601 2007.0.0 Aeroelastic Flutter Analysis - Logic error in fa1pke When a singular mass matrix is used in flutter analysis, the job may fail with the message *** USER FATAL MESSAGE 3200 (FA1) LOGIC ERROR DETECTED BY SUBROUTINEFA1PKE 1 = LOCATE CODE OR VALUE. 1-47738693 2007.0.0 Nonlinear Transient Analysis - SE and CMS, NLTRD2 - SFM 1282 (SLCDRV) A nonlinear transient analysis with CMS fails in the NLTRD2 module with: *** SYSTEM FATAL MESSAGE 1282 (SLCDRV) AN ATTEMPT WAS MADE TO OPEN A NON EXISTENT ASSOCIATED FILE ULNTH ,FNUMERIC, TO READ. PROGRAMMER INFORMATION: CHECK THE NAME OF THE ASSOCIATED FILE TO OPEN. The same message appears when only CMS is used without the SE. 1-47879121 2007.0.0 Matrix Methods - CASI Solver, TET10, TEMP(LOAD) - Incorrect stresses CASI solver temperature loads When TET10 elements are present in the input file, the default behavior of the casi solver is to use the "implicit" formulation for the element. This means that the element stiffness matrix is formulated internally by the casi software. The nastran generated stiffness matrices are not used for the TET10s. This is usually acceptable, unless the applied loads are element-based, as in this case, where temperature loads are used. Furthermore, in this case, in order to achieve stress-free thermal expansion, it is essential that the element stiffness and loads be compatible. Implicit element formulation should not be used. This can be achieved by placing a PARAM,ITOPTX,4032 entry in the bulk data section. Avoidance: insert PARAM,ITOPTX,4032 in the bulk data section to prevent implicit TET10 usage. 1-48031006 2007.0.0 Nonlinear Analysis - Using NSM in sol 106 or 400 causes SFM 3001 Using NSM in sol 106 or 400 results in the message: *** SYSTEM FATAL MESSAGE 3001 (TA1) THE OUTPUT DATA BLOCK IN POSITION 9 DOES NOT EXIST. USER INFORMATION: THIS ERROR IS CAUSED BY ONE OF THE FOLLOWING: 1. THE DATA BLOCK IS UNSPECIFIED ON THE DMAP MODULE. 2. THE DATA BLOCK IS SPECIFIED ON THE DMAP MODULE AND ON THE SUBDMAP STATEMENT BUT NOT ON THE CORRESPONDING CALL STATEMENT. Avoidance: Use the following alter. compile phase1a alter ' ta1 ','' TA1 MPTS,ECTS,EPTS,BGPDTS,SILS,ETT,CSTMS,DIT,,/ EST,ESTNL,GEI,GPECT,ESTL,vgfd,ditid,nfdict,nsmest/ LUSETS/S,N,NOESTL/S,N,NOSIMP/2/S,N,NOGENL/SEID/ LGDISP0/NLAYERx/s,n,freqdeps/shldamp/s,n,mge/nsm $ 1-48240611 2007.0.0 Frequency Response - Wrong Answers (SOL 108 and 111) with K4 & TFL, and Multiple Subcases Wrong answers will result in SOL 108 or SOL 111 when structural damping (K4) is combined with transfer functions. Avoidance: use the following DMAP alter: compile sedfreq alter 'EQUIVX.*K4AA\/K4DD',''(,1) alter 'call gma'(,-1) EQUIVX K4AA/K4DD/NOUE $ IF ( NOUE>-1 ) UMERGE1 USETD,K4AA,,,/K4DD/'D'/'A' $ compile semfreq alter 'EQUIVX.*K4AA\/K4DD',''(,1) alter 'call gma'(,-1) EQUIVX K4AA/K4DD/NOUE $ IF ( NOUE>-1 ) UMERGE1 USETD,K4AA,,,/K4DD/'D'/'A' $ 1-48291201 2007.0.0 Aeroelastic Analysis/QRG - aedw, aepress, aeforce do not support MACH=0.0 The Quick Reference Guide specifies that the MACH input must be .ge. 0.0, but if the user sets this input to 0.0, is is marked as an illegal data. User Avoidance: The MACH must be set to a number .gt. 0.0 1-48291704 MDACMS - SDAMPING - Wrong Answers with K4 Structural Damping Modal Frequency response (SOL 111) with MDACMS may produce wrong answers with structral (K4) damping. Avoidance: The workaround is to include PARAM,MDK4OPT,0 1-48302021 Optimization/Design Sensitivity - SFM 6043 (XQAERR) - unable to evaluate expressions of oscar type 2 When a DRESP2 response with the EQID field = FUNC (such as SUM) is followed by another DRESP2 response with EQID=Integer the job may fail with the message: SYSTEM FATAL MESSAGE 6043 (XQAROS) UNABLE TO EVALUATE EXPRESSIONS OF OSCAR TYPE 2 Avoidance: The avoidance is to assign the largest RID to those DRESP2s with EQID=FUNC so that they will always be placed in the last. 1-48326271 2007.0.0 MDACMS - SOLs 111 and 112 with Enforced Motion fails with PARAM,ENFMETH,ABS MDACMS in SOLs 111 and 112 with Enforced Motion fails with PARAM,ENFMETH,ABS. The following message may be issued: SYSTEM FATAL MESSAGE 5423 (MPYAD1) ATTEMPT TO MULTIPLY INCOMPATIBLE MATRICES FILE# COLS ROWS FORM TYPE NZWD DENS TRAILER FOR MHS1 IS 101 26 30 2 2 60 4615 TRAILER FOR ACCE IS 102 202 1 2 2 2 10000 TRAILER FOR (NONE) IS 0 0 0 0 0 0 0 TRANSPOSE FLAG IS 0 Avoidance: Insert the following DMAP alter: compile getpyh alter 'call xADD5.*mhs','mpyad.*khs1,disp,pmbh'(,1) equivx yst/disp/-1 $ call xADD5 mhs,m2hs,,,/mhs1/cu/cu/cz/cz/cz $ call xADD5 Bhs,B2hs,,,/Bhs1/cu/cu/cz/cz/cz $ c1g=CMPLX(1.,G) $ CALL xADD5 Khs,K4hs,K2hs,kdhs,/Khs1/C1G/ci/cu/cu/cz $ else $ $ transient response dsar yst,,,/disp,velo,acce,,/-1 $ call xADD5 BHS,B2HS,,,/BHS1/cu/cu/cu/cz/cz $ call xADD5 KHS,K2HS,kdhs,,/KHS1/cu/cu/cu/cz/cz $ call xADD5 mhs,m2hs,,,/mhs1/cu/cu/cz/cz/cz $ endif $ app='freqresp' $ if ( enfmeth='abs' and nonzys>-1 ) then $ partn bhs1,nzys,/,,bhs2,/1 $ partn mhs1,nzys,/,,mhs2,/1 $ partn khs1,nzys,/,,khs2,/1 $ mpyad mhs2,acce, /pmh $ mpyad bhs2,velo,pmh/pmbh $ mpyad khs2,disp,pmbh/pyxt $ else $ mpyad mhs1,acce, /pmh $ mpyad bhs1,velo,pmh/pmbh $ mpyad khs1,disp,pmbh/pyxt $ endif $ 1-48865131 2007.0.0 Aeroelastic Analysis - SOL 145 may fail if MONPNT is in the deck SOL 145 does not support MONPNT. However, SOL 145 job may fail if MONPNT is simply in the deck. Avoidance: 1. remove or comment out MONPNT entry 2. use attached DMAP alters (add dmonpta/dmonpts to call of SUPER1) 1-48873671 2007.0.0 Aeroelastic Analysis - infinite loop in sol 144 If a Solution 144 (static aeroelasticity) run is submitted that does not have any TRIM or DIVERG case control commands in any of the subcases, an infinite loop will result in the PFAERO subdmap. User Avoidance: Make sure the required TRIM and DIVERG case control commands are included in case control 1-48879511 2007.0.0 Miscellaneous - ISHELL jobs terminate abnormally when run using a job scheduler If Nastran starts an ISHELL job that is suspended by a job scheduler (using a "kill -STOP"), the ISHELL processing routine (doit.c) does not properly recognize this situation and terminates processing with an error. 1-48996563 2007.0.0 SOL 600 - BCBODY - Friction not written correctly, Wrong Answers The marc CONTACT entry is not written correctly when the BCBODY entry contains a friction coefficient. 1-49381161 2007.0.0 SOL 700 - EOSGAM entry error EOSGAM entry gives the following FATAL ERROR : Illegal name for bulk data entry. Avoidance : Use the TODYNA- ENDDYNA entry to bypass this entry. 1-49495686 Rotordynamics - SOL 108 or 129, RSDAMP - Hybrid Damping READ Module, UFM 4683 A 3D rotordynamics model fails in the READ module with the following error *** USER FATAL MESSAGE 4683 (REIG) MASS MATRIX NEEDED FOR EIGENVALUE ANALYSIS when hybrid damping is requested in the case control with RSDAMP command. Quality 1-49098315 may be related since it is a similar setup but without a superelement. 1-49496581 2007.0.0 Aeroelastic Analysis - AECOMP with a continuation with data in field 9 cause an IFP error The AECOMP entry supports an open-ended list of set ids for the component. If the list of set id's ends with the 9th field of a continuation entry (i.e., there are exactly 14, 23,32, etc. ID's) the job will fail with an invalid data message. User Avoidance: The AECOMPL entry can be used to collect multiple AECOMPs so this provides a way to avoid the above liimitation. 1-49703571 2007.0.0 Aeroelastic Analysis - SPLINE6 - INSUFFICIENT CORE AVAILABLE FOR SUBROUTINE F6GD A deck that included spline6 entries that invoked over a thousand structural and aerodynamic grids produced the message: INSUFFICIENT CORE AVAILABLE FOR SUBROUTINE F6GD A review of the code indicates that the required amount of memory can be reduced drastically and the UFM avoided. The error occured for the spline6,but similar memory inefficiencis exist for all the spline type elements. User Avoidance: Run with the additional memory requested by the message or break the large splines into a number of smaller splines to get past the error. 1-50027590 2007.0.0 Optimization - UFM 6656 (MODTRK) THE TRACKED SUBSET PHGREF OF MODES FROM A PREVIOUS DESIGN CYCLE The message "USER FATAL MESSAGE 6656 (MODTRK) THE TRACKED SUBSET PHGREF OF MODES FROM A PREVIOUS DESIGN CYCLE" can be produced whenever the user specifies a MODTRAK command in a SOL 200 subcase that contains a METHOD command but is not an ANALYSIS=MODES subcase. E.g, using MODTRAK with ANALYSIS = MCEIG, MTRAN, MFREQ or FLUTTER can produce the error. Avoidance: Remove the MODTRAK case control command from subcases that do not also have an ANALYSIS =MODES command.