Beams Beams

In MOSES, an element is viewed as being the union of three sets of attributes: its vertices, its class properties, and its "additional attributes". Defining the first two of these has already been discussed. To complete the definition of a beam, one uses:

     BEAM, ELE_NAME(1), ~CLASS(1), -OPT(1), *NODE(1), ... *NODE(m), \
     ELE_NAME(2), ~CLASS(2), -OPT(2), NODE(n), ...

and in addition to the available options discussed above we have:


     -KFAC, KY, KZ






BEAM defines a string of beams. Here, the ~CLASS(i) defines the section attributes, and *NODE(1), *NODE(2), ...*NODE(n) is the list of connected nodes where, *NODE(1) connects to *NODE(2), *NODE(2) connects to *NODE(3), and so on. The locations of the command marked by -OPT(i) are positions where one may place as many element options as desired, and may be omitted if defaults are suitable for the beam in question. The beams from *NODE(1) to *NODE(m) have the properties defined by ~CLASS(1) and -OPT1, and the beams from *NODE(m) to *NODE(n) have properties defined by ~CLASS(2) and -OPT(2). In other words, a beam in the string has the properties defined by the last ~CLASS and/or -OPT data issued before the beam was defined. Here, ELE_NAME(i) is a name which can be assigned to the element. If it is omitted, MOSES will assign a name. The ~CLASS is a name for a set of element attributes which define the "sectional properties". An example of a beam definition accompanied by a sketch is shown in Figure 8.

The options specified above are used when checking codes. To alter the CM values for an element, one can use the -CMFAC option. Here, CMY is the factor for bending about the Y axis and CMZ is the factor for bending about the Z axis. The K factors and the "buckling lengths" are multiplied to obtain the effective length of the element about each axis. By default, the buckling lengths are taken to be the element length for beams and the square root of the area for generalized plates. In particular, to define the effective length multipliers for beams, one uses the option -KFAC. Here, KZ is used for bending about the Z axis, while KY is used for bending about the Y axis. If this option is not used, both factors will be set to 1.

Element lengths for beams include the effect of any offsets, invoked by either the -GO and -LO options on the BEAM card, or the -OFFSET option on the INMODEL command.

The -BLENG or -BLY and -BLZ options can be used to alter the buckling lengths about each axis. Here, the dimensions are feet or meters. The -BLENG, BELE_NAME construct offers a way to "bind" the buckling length of an element to the load state of another element. Here, BELE_NAME is the name of the "brace element". If this option is exercised, then the KL factors of the basic element will be those input if the brace element is in compression. If the brace element is in tension, then the factor for out of plane will be the same as for inplane. When this option is used, the "compression" lengths will be used in any report or computation outside of the Structural Post-Processing Menu. An alternative way of defining this type of dependence is provided in the MEDIT Menu with the XBRACE command.

The -CFB option defines compression flange brace spacing, CFSPAC, (inches or mm). If this option is omitted, then the value will be taken to be the element length. Again, this is the length after accounting for any offset. The -HAS_P-DELTA option tells MOSES that this beam has nonlinear effects taken into account. If YES/NO is YES then no interaction effects will be taken into account when the codes are checked. If YES/NO is NO, or no option is used, then standard interaction formulae will be used in the code checking. For a tube, they are generally computed based on the load path and the formulae specified. For other type sections, they are input.

As an example of defining beams, consider:

     ~STF  TUBE  30  .75  -FY  42 -LEN 3
     ~STF  TUBE  30  .675  -FY  42 -LEN 0.
     ~STF  TUBE  30  .75  -FY  42 -LEN 3
     BEAM ~STF -RELA MY -GO1 10 12 *AAA1010 *BBB1010

Here we have defined a single beam connecting the nodes *AAA1010 and *BBB1010, with properties defined by the class name ~STF. The beam consists of 3 segments, hence the three STF commands. The first and third segments are tubular sections with 30 inch OD and 0.75 inch wall thickness, while the middle section has the same OD, but only a 0.675 inch thickness. The first and third segments have a length of 3 ft. and the length of the middle segment is to be computed by the program. All segments have a yield stress of 42 ksi.

As a second example, consider:

     ~LONG, W18X40
     BEAM, ~LONG, *A, *B

This defines 1 beam as a wide flange beam (18 x 40). This is a prismatic beam, as only one ~LONG command is given.

The example below uses a string of nodes for a beam description, which can sometimes be quite useful:

     ~TUBE TUBE 42  1.625
     BEAM ~TUBE -GO1 10 10 10 *1 *2 *3  \
     -GO2 12 12 12 *4 *5

In the above example, there are five nodes used, and four individual beam elements created. Each of the beam elements has a name assigned to it by MOSES. The options apply in the order the beam is defined. For instance, *1 is the A end of the beam joining nodes *1 and *2, and the -GO1 is a global offset at *1. *2 is the A end of the beam joining nodes *2 and *3, with *3 as the B end of the beam. This sequence continues for as many nodes as there are to define the entire series of beams. For the last beam of this series, *5 is the B end of the beam joining node *4 and *5, and the -GO2 option is a global offset at *5. The backslash (\) used above is for a continuation of the same command on the next line.

In many instances, one needs to define beam load attributes other than those defined via the element description. Three mechanisms for this are provided: two commands and the -T_PRESSURE option on the &ENV command and a T_PRESSURE command. This option allows one to specify the temperature, internal pressure and the density of the contents of a beam. The first of the command provided is:


and the available options are:



     -A, XA

     -B, XB


This command is used to define additional intrinsic load attributes for a logical beam, and OBJECT is the name of the object to which they apply.

If OBJECT is two node names (they may include wild characters, but must begin with an *), the attributes will apply to all beams between those two nodes. Remember, there may be more than one beam defined between the same two nodes. Also, these two selectors can be used to define load attributes on a "logical" beam. For example, suppose that there is a beam between *1 and *2 and a beam between *2 and *3. If these two beams are colinear, then #ELAT *1 *3 will apply the load to both beams. If OBJECT is an attribute class name (begins with a ~), then the attributes will apply to all elements which belong to classes which match OBJECT. If OBJECT begins with neither an * nor a ~, then the attribute will be applied to all members whose names match OBJECT.

The data defines the attributes which will be added to the element. Here, WTPFT is a weight per foot (bforce/blength), DPFT is a buoyancy per foot (bforce/blength), BOD is the diameter (inches or mm) which will be used for buoyancy, DOD is the diameter (inches or mm) which will be used for viscous drag, AMOD is the diameter (inches or mm) which will be used for added mass, and WINOD is the diameter (inches or mm) which will be used for wind. With this command, one defines a line buoyancy with DPFT as well as a diameter which will be used for computing buoyancy. Both can be used at the same time. The added mass and viscous drag computed will be based on Morison's Equation. The -TOTAL option denotes the fact that the values input for WTPFT and DPFT are total quantities and should be divided by the length to obtain the distributed properties.

By default, loads produced from the properties specified with #ELAT are assigned to the default Extra Category. This can be changed with the -CATEGORY option. Also, the load attribute defined here will, by default, act over the entire length of the elements selected. If one wishes, he may alter this distribution with the options: -A, -B, and -LENGTH. Here, the attributes are defined over a segment of the original beam beginning XA (feet or meters) from the "A" end of the beam, and extending to XB (feet or meters) from the "B" end, and the length over which it is applied is LEN. The defaults are that both XA and XB are zero and LEN is the length of the beam. Thus, for a load over the entire beam, none of the option list is needed. Notice that the "B" end of an attribute can be defined by either -XB or -LENGTH. Both should not be used on the same command.

The second command for beam attributes defines an applied load which belongs to user defined load set,

     #LSET, OBJECT, VALA(1), .... VALA(6), -OPTIONS

and the available options are:

     -VB, VALB(1), ...  VALB(6)

     -A, XA
     -B, XB

Here, VALA(i), shown as FA, are the values of the attribute XA (feet or meters) from the "A" end of the beam, and VALB(i), shown as FB, are the values at XB (feet or meters) from the "B" end. If -VB is not specified, then the values at XB will be taken to be the same as those at XA. The units for VAL(i) are in bforce and bforce-blength. To define an attribute in beam coordinates, one simply adds the -LOCAL option. To input a concentrated attribute, one should specify LEN to be zero. Figure 9 illustrates how the options are used. The -TOTAL command denotes the fact that the values input for VALA(i) and VALB(i) are total quantities and should be divided by the length to obtain the distributed properties. The options -A, -B, and -LENGTH operate in the same manner as the corresponding option for #ELAT and were discussed above.