Question about periodic symmetry analysis with muti-kinds element? 2

[Lirock]

Hi,everyone:
I did a periodic symmetry modal analysis.The structure contains both solid and shell parts.According to the help file,I used 'CYCLIC' command,which make me have to mesh the structure with 'VMESH' and 'AMESH' commands.
I got solid mesh in the solid parts with 'VMESH' using SOLID95,but after that,I issued 'AMESH', the programm gave me nothing,no mesh in the shell parts,no crush out,just nothing.
Is there anyone met this problem?Or someone give me some advice?

Thank you!

Rock Li

 

[meshparts]

Hello,

Try this: select only the shell areas (not yet meshed) and then issue AMESH,ALL. Also check the status of the command /nerr or issue /nerr,defa. However the errors, if any, should be found in file.err.

Regards,
Alex

 

[cbrn]

Hi,

yes, you should at least see some error if nothing happens... So it means that there was nothing to apply the command upon: I'm with Mihaiupb in saying that you should ensure that the areas are currently selected.
In my opinion it doesn't matter if the areas belonging to the volume are also selected: in this case, a message box should pop up saying something like "xxx areas are already meshed. Do you want to re-mesh them? <yes; no>"; simply say "no" and you should be OK.
Note: you can not AMESH with SOLIDxxx, of course (I suppose you skipped to tell us which kind of elem you use for the "areas").

Regards

 

[Lirock]

Thank you,guys.I will try your advice.
Another thing,I noted that for a periodic modal analysis, the output appears as 'nodial diameter xxxx' in old version,such as ANSYS6.0 etc.,but in ANSYS9.0, it appears as 'harmonic index xxxx'. Is it a development or two kinds formats?I know it has no effect to the results, but I want to make it clear.
Thank you and regards!

Rock Li

 

[cbrn]

Hi,
these are the same things. "nodal diameter #" refers to the number of nodes in the waveform Fourier-series decomposition of the cyclic load, whether "harmonic index #" refers to the order of the truncation of the Fourier-series.
The new notation reflects more clearly the command you use in order to set non-axisymmetric loads on an axisymmetric model ("MODE,<mode>,<harmonic index>").

Regards

 

[Lirock]

Hi,
I tried the advice,chosing the area and excute 'AMESH',but nothing changed. I made a excersice and paste the commands below.It's just end at the command 'AMESH'. You may will see the problem I met.
/PREP7
wpro,,,90.000000
WPCSYS,-1,0
wpro,,90.000000,
CSYS,4
RECTNG,2,4,-0.5,0.5,
RECTNG,2,4,-0.5,2,
CSYS,0
K,,
K,,,,1
FLST,2,1,5,ORDE,1
FITEM,2,1
FLST,8,2,3
FITEM,8,9
FITEM,8,10
VROTAT,P51X, , , , , ,P51X, ,15,1,
FLST,2,1,5,ORDE,1
FITEM,2,1
FLST,8,2,3
FITEM,8,9
FITEM,8,10
VROTAT,P51X, , , , , ,P51X, ,-15,1,
FLST,2,12,5,ORDE,2
FITEM,2,1
FITEM,2,-12
APTN,P51X
NUMMRG,ALL, , , ,LOW
FLST,2,2,6,ORDE,2
FITEM,2,1
FITEM,2,-2
VPTN,P51X
NUMMRG,ALL, , , ,LOW
ET,1,SOLID92
ET,2,SHELL93
mp_define
ANTYPE,2
MSAVE,0
MODOPT,LANB,4
EQSLV,SPAR
MXPAND,4, , ,0
LUMPM,0
PSTRES,0
MODOPT,LANB,4,0,0, ,OFF
FLST,2,2,5,ORDE,2
FITEM,2,6
FITEM,2,11
/GO
DA,P51X,UX,
CYCLIC
ESIZE,1
TYPE,1
VMESH,ALL
ASEL,S, , , 13
ALLSEL,BELOW,AREA
TYPE,2
AMESH,ALL

Regards!

Rock Li

 

[meshparts]

It works for me! I don't have the macro mp_def but it works! Areas and elements are meshed!

Alex

 

[cbrn]

Hi,
try forcing a replot...

Regards

 

[Lirock]

Hi,guys:
It still doesn't work for me.I don't know the reason and I am going to find some help in ANSYS company.
Thank you for your help!

Rock Li