Hi all.. I a trying to learn MOSES the hard way, from the reference manuals, online tutorials and looking at the files from vessel library. I am creating the model of a semisubmersible of arbitrary dimensions, for hydrodynamic analysis. I have reached this stage after few trial and error. I have been writing commands in the command line. I have also given those steps below. I have tried a lot but, still not been able to figure out how to create DAT file and CIF file. Is there any command to save the current status of the model to a DAT file? Also, I have used the &SURFACE command till now. Should I define panels separately from here or create a mesh? I know these are lot of questions but, I am really stuck here. Any help or hint would be appreciated. Thanks.
>&device -g_default screen >&surface >block pont -loc 0 28 0 0 0 0 >plane -50 -40 -30 -20 -10 0 10 20 30 40 50 -rect 0 8 14 >end block >&pict iso pont >block pont2 -loc 0 -28 0 0 0 0 >plane -50 -40 -30 -20 -10 0 10 20 30 40 50 -rect 0 8 14 >end block >union pont pont2 pont3 >delete pont pont2 >&pict iso pont3
>block col_f1 -loc 40 28 8 0 270 0 >plane 0 8 -rect 0 14 14 >end block >block col_f2 -loc 40 -28 8 0 270 0 >plane 0 8 -rect 0 14 14 >end block >union col_f1 col_f2 col_f >delete col_f1 col_f2
>block col_a1 -loc -40 28 8 0 90 0 >plane -8 0 -rect 0 14 14 >end block >&pict iso >block col_a2 -loc -40 -28 8 0 90 0 >plane -8 0 -rect 0 14 14 >end block>union col_a1 col_a2 col_a >delete col_a1 col_a2 >block brace1 -loc 33 -21 8 0 0 90 >plane 0 42 -circular 0 0 1.5 0 45 5 >end block >&pict iso >block brace2 -loc -33 -21 8 0 0 90 >plane 0 42 -circular 0 0 1.5 0 45 5 >end block
>list_blocks +++ B L O C K S +++ =================== BRACE1 BRACE2 COL_A COL_F PONT3
Hi!
I think your way is the best way to model semis in Moses.
I usually have two different CIF files. One for making the mesh and one for analysis.
My makemesh.cif says:
&DIMEN -DIMEN METERS M-TONS
&DEVICE -OECHO YES
INMODEL
&EOFILE
Then there is a makemesh.dat which includes the &SURFACE command.
Remember you can use REFLECT_BLOCK to reduce the amount of modelling comands.
You can use &CHANNEL MODEL -FILE yourmesh.dat in top of your makemesh.dat to output your mesh.
Then combine to a single block and use
RENAME_BLOCK yourblock -EQUIVALENT 0.1
to remove tiny bits.
Then
EMIT yourblock -PART
EMIT yourblock -USE_NAME YES
And finally
&CHANNEL MODEL -FILE dummy.dat
to close your file
This should give you a dat file with points and panels that you can use for your analysis. The panels are usually too big for diffraction analysis, but you refine MOSES makes smaller panels automatically with
&PARAMETER -M_DISTANCE 2.5
that you put on top of you cif file you use for analysis
Use
&DEVICE -AUXIN yourmesh.dat
and then
To read in your mesh for analysis.
Hi Lars.. After the "&CHANNEL MODEL -FILE dummy.dat" I am not getting the dat file. Instead I have got a 'mod00001.txt' file in the ans directory whose contents seems like the definition of points and panels (given below). What might be the problem? Can I just rename the file to .dat and use?
$$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$@ @$@ &SURFACE Results @$@ @$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$$$**************************************** Set Dimensions$&dimen -save -dimen Meters M-Tons$$**************************************** Defaults$&LOCAL xfac = 1 yfac = 1 zfac = 1&default -save&default -fyield 248.04 -alpha 3.6111E-6 -spgravit 7.8492 -emodulus 1.9981E5 \ -poi_ratio 0.3 -kfac 1 1 -cmfac 0.85 0.85 -flood no -use @$$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$@ @$@ Define Points @$@ @$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$$$**************************************** Set factors$&set xfac = 1&set yfac = 1&set zfac = 1$$**************************************** Set part to ONE$&describe part ONE$$**************************************** Define Coordinates$**S00001 0.000*%xfac 0.000*%yfac 7.000*%zfac**S00002 0.000*%xfac 6.000*%yfac 7.000*%zfac**S00003 0.000*%xfac 6.000*%yfac**S00004 0.000*%xfac 0.000*%yfac -0.000*%zfac**S00005 0.000*%xfac -6.000*%yfac**S00006 0.000*%xfac -6.000*%yfac 7.000*%zfac**S00007 10.000*%xfac 0.000*%yfac 7.000*%zfac**S00008 10.000*%xfac 6.000*%yfac 7.000*%zfac**S00009 10.000*%xfac 6.000*%yfac**S00010 10.000*%xfac 0.000*%yfac -0.000*%zfac**S00011 10.000*%xfac -6.000*%yfac**S00012 10.000*%xfac -6.000*%yfac 7.000*%zfac**S00013 20.000*%xfac 0.000*%yfac 7.000*%zfac**S00014 20.000*%xfac 6.000*%yfac 7.000*%zfac**S00015 20.000*%xfac 6.000*%yfac**S00016 20.000*%xfac 0.000*%yfac -0.000*%zfac**S00017 20.000*%xfac -6.000*%yfac**S00018 20.000*%xfac -6.000*%yfac 7.000*%zfac**S00019 30.000*%xfac 0.000*%yfac 7.000*%zfac**S00020 30.000*%xfac 6.000*%yfac 7.000*%zfac**S00021 30.000*%xfac 6.000*%yfac**S00022 30.000*%xfac 0.000*%yfac -0.000*%zfac**S00023 30.000*%xfac -6.000*%yfac**S00024 30.000*%xfac -6.000*%yfac 7.000*%zfac**S00025 40.000*%xfac 0.000*%yfac 7.000*%zfac**S00026 40.000*%xfac 6.000*%yfac 7.000*%zfac**S00027 40.000*%xfac 6.000*%yfac**S00028 40.000*%xfac 0.000*%yfac -0.000*%zfac**S00029 40.000*%xfac -6.000*%yfac**S00030 40.000*%xfac -6.000*%yfac 7.000*%zfac$$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$@ @$@ Define ONE @$@ @$@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@$$$**************************************** Set Piece to ONE$&describe piece ONE$$**************************************** Define Panels$PANEL SU#00001 **S00006 **S00005 **S00004 **S00003 **S00002 **S00001PANEL SU#00002 **S00001 **S00002 **S00008 **S00007PANEL SU#00003 **S00002 **S00003 **S00009 **S00008PANEL SU#00004 **S00003 **S00004 **S00010 **S00009PANEL SU#00005 **S00004 **S00005 **S00011 **S00010PANEL SU#00006 **S00005 **S00006 **S00012 **S00011PANEL SU#00007 **S00006 **S00001 **S00007 **S00012PANEL SU#00008 **S00007 **S00008 **S00014 **S00013PANEL SU#00009 **S00008 **S00009 **S00015 **S00014PANEL SU#00010 **S00009 **S00010 **S00016 **S00015PANEL SU#00011 **S00010 **S00011 **S00017 **S00016PANEL SU#00012 **S00011 **S00012 **S00018 **S00017PANEL SU#00013 **S00012 **S00007 **S00013 **S00018PANEL SU#00014 **S00013 **S00014 **S00020 **S00019PANEL SU#00015 **S00014 **S00015 **S00021 **S00020PANEL SU#00016 **S00015 **S00016 **S00022 **S00021PANEL SU#00017 **S00016 **S00017 **S00023 **S00022PANEL SU#00018 **S00017 **S00018 **S00024 **S00023PANEL SU#00019 **S00018 **S00013 **S00019 **S00024PANEL SU#00020 **S00019 **S00020 **S00026 **S00025PANEL SU#00021 **S00020 **S00021 **S00027 **S00026PANEL SU#00022 **S00021 **S00022 **S00028 **S00027PANEL SU#00023 **S00022 **S00023 **S00029 **S00028PANEL SU#00024 **S00023 **S00024 **S00030 **S00029PANEL SU#00025 **S00024 **S00019 **S00025 **S00030PANEL SU#00026 **S00025 **S00026 **S00027 **S00028 **S00029 **S00030
HelloVishnu,
The portion of the file that you are showing does look like a model data file. This is the dat file.
I do not see a command "&describe body one" in the portions you have shared. Please check the file for the command "&describe body one". If it is not there you can add it before any point definitions. You can add it before the "&describe part one" command. Or you can go back to the surface menu and add it via one of the "emit" commands.
Please give this a try.
Georgina Maldonado
You need "&channel model -file yourmesh.dat" before your "&surface end"-block and "&channel model -file dummy.dat" is just to close the first file. Not sure why it doesn't work.
But you can rename your .txt file to .dat as well.
Always a good idea to inspect it from different angles with "&pict -cull" to look for internal panels that need to be removed (the union commands are not always 100% robust). Just click those panels to find their names, and comment them out in your txt/dat file.
If I join two blocks with a common cross-section using union, will there be any problem. I tried it and the blocks seem joined, but gives a warning about 'Contour' which I don't understand.
Warnings or not - You should always check the panel model visually to check that it is closed, that there are no internal elements, and that the panels have the correct orientation. If it looks good, I wouldn't worry about the warnings from the joining of the blocks.
Hi Vishnu,
I agree with Lars' response.
The message about "Contour" is telling you the program is having a difficult time establishing what is the inside and outside of the volume. This mostly happens when two panels occupy the same plane. Many times moving a block just 1 cm will fix the difficulties.
If you review the model as Lars' instructed and are happy with it, then do not worry about the warnings.
Thank you for keeping us updated on your use of MOSES. :-)
I have managed to create a hydrodynamics database for my model using 3d diffraction but, I feel that I am not yet clear about the basic idea of bodies, parts and pieces. I may have just gotten lucky after few trials. What's the difference between emitting my final block as -BODY, -PIECE, -PART? Diffraction theory worked only when I included -PIECE, not when -PART. One more question, how to estimate the SPE_MULTIPLIER for drag linearization?
The body, part and piece are explained in the manual. The piece is a closed volume made up of panels (i.e the hull), and you need a piece to do diffraction analysis. The piece belongs to the body you are analysing (i.e. the semi). You can have other pieces as well, like internal tanks. I wouldn’t worry about the parts unless you are doing structural analysis.
The selection of -spe_mult will affect the loading on Morison elements which is useful if you want the bending moment in your braces for instance. If you are just looking at the motions, you just have to make sure that your combination of drag factor and -spe_mult gives appropriate damping. The -cs_curr 1 1 1 together with the default -spe_mult give too much damping. I normally use the default -spe_mult and -cs_curr 0.15 0.15 0.15. You may very well use -cs_curr 1 1 1, but then you should reduce your -spe_mult.
I agree with Lars.
Thank you Lars.
Georgina