** WARNING ** A SOFT MATERIAL WITH (1.0 / 1.750E+01) TIMES THE STIFFNESS OFCONCRETE ENTERED. PLEASE CHECK.
Please explain to me in plain English what StaadPro is trying to tell me.
STAAD checks to see if the E (Modulus of Elasticity) assigned to members and elements is comparable to the values of steel, aluminum, concrete or timber. If it falls below or above the range of these materials, warning messages similar to the one you encountered are displayed. This is done to notify the user in case he/she is not aware of this fact, or if he/she may have specified the value in an incorrect unit system.
If you believe that your E is specified correctly, you may ignore the message. Else, correct the number.
Unlike the 3 materials mentioned in your question, timber comes in several varieties, with each variety having its own unique set of material properties. Douglas Fir alone comes in several varieties, as explained below.
The American Wood Council and the American Forest & Paper Association publish a document called the "Supplement NDS for Wood Construction", 1997 edition. It provides design values for structural sawn lumber and glued laminated timber. There is also a category called Visually Graded Decking.
Under each category, Douglas Fir comes in various species or combination of species. Under each species, there are various commercial grades. Each of those grades have a unique value of E, ranging from 1000 ksi to 1900 ksi. If the category, species, and commercial grade is known, the E value can be read from the tables in this document.
The American Wood Council and the American Forest & Paper Association also publish a document called the "ASD Manual for Engineered Wood Construction". In the 1999 edition of this document, Table 8A, page 15 contains the specific gravity of Douglas Fir as ranging from 0.46 to 0.5.
For plane frames with no beta angle, what is needed is IZ, not IX. IX is the torsion constant. IZ is the moment of inertia about the Z axis. Members of a plane frame with a beta angle of zero will bend about the Z axis, which explains the need for IZ. They are not prone to twisting, and that is why IX is not needed.
Table 1.1 from the Technical Reference manual, which shows the properties required for various types of structures, is reproduced below.
By default, STAAD assumes the connection between any 2 members to be fully capable of transmitting all 3 forces and all 3 moments from one member to the other. This is usually achieved in practice by moment resistant connections, such as between a concrete beam and a concrete column which are monolithically cast.
If you want the connection to be of the type which does not permit one or more forces/moments to be transmitted, use member releases. A shear connection is such an example. The degrees of freedom FX through MZ that you release are based on the local axis of the member at whose end the release is specified.
See section 5.22.1 and the figures in Section 1.19 of the STAAD.Pro Technical Reference manual for additional information.
When creating a model consisting of beams and columns, generally, the START or END face of the member is assumed to be located at the nodal point. In other words, the distance from the respective node to the start or end face of the member is treated as zero. Thus, for example, if member 47 is defined as being connected between nodes 12 and 13, then, the start face of the member is located at node 12, and the end face at node 13.
This assumption may not always reflect the true physical condition on the structure. For example, when a beam meets a column, the common node between the beam and column is usually defined as being at the shear center (centerline for symmetrically shaped) of the column.
But, physically, the start face of the beam is not at that node, but at half the column depth away from the node. One may choose to ignore this "shift" if the column depth is negligible in comparison to the span of the beam. However, if one wishes to take advantage of the high stiffness that the half-depth region of the column offers, he/she may consider this using the member offset command.
The member offset is a way of declaring that the region, whose length is defined by the offset, is a rigid zone. Hence, if the offset values in X, Y and Z coordinates are a, b and c, the length of that region is d=sqrt(a*a + b*b + c*c). The face of the member is then assumed to be "d" away from the node.
The member end forces that STAAD reports are at the face of the member, not at the node, when an offset is specified. If the offset is applied at the base of a column, then the member end force may not be equal in magnitude to the corresponding support reaction terms. If one is interested in checking static equilibrium based on the free body diagram at that support, the member end forces must be transferred from the member face to the support node taking into consideration the rigid link defined by the offset.
In fact, that is exactly what STAAD is designed to do already. There is no need to keep re-specifying the MEMBER TENSION command, unless you want to specify a different list of such members. So, specify it once for the first analysis, and you don't have to specify it again. Same goes for the MEMBER TRUSS command.
a - Is the graphical interface a reliable representation of my input?
b - If yes, can you think of some other possible sources of this particular error?
If you look at the coordinates of the columns which appear to be oriented in the wrong way, chances are that you will find the Z coordinate of the 2 ends to be different by a very minute value, such as 0.001. For example, one end may have a Z value of 5.999 while the other end may be at 6.000. If so, you could do the following to correct it. Select the Geometry-Beam page along the left side of the screen, and it will display the node coordinates in the tables on the right hand side. In those tables, make the necessary correction so both ends of the column have the same Z coordinate.
The potential cause of this difference in coordinates is the following. The program has something called a Base Unit system. You can find this by starting the program, and before opening any file, go to the File menu, select Configure, and see if it says "English" or "Metric". If the model you are going to create is in Metres and KNs, you ought to have the base units in Metric. If the model you are going to create is in Feet and Kips, you ought to have the base units in English. Mixing unit systems causes the program to perform internal unit conversions which can result in loss of digits because the built-in conversion factors have only upto 8 digits of accuracy.
In fututure versions of STAAD, there will be a feature which will enable you to select the "offending" column and make the Z coordinate of its 2 ends to be equal so it becomes truly vertical.
The numbers reported in the STAAD output for Iz and Iy are the moments of inertia about the principal axes of the single angle. The values in the AISC publication that you are comparing them with are most probably the values about the geometric axes. That is very likely the cause of the mis-match.
Before we can explain why, we first need to understand a few facts about loads in STAAD. There are two types of load cases in STAAD : Primary load cases, and Combination load cases.
Primary load cases
A primary load case is one where the load data is directly specified by the user in the form of member loads, joint loads, temperature loads, element pressure loads, etc. It is characterized by the fact that the data generally follow a title which has the syntax
LOAD n
where "n" is the load case number. For example,
LOAD 3
MEMBER LOAD
2 UNI GY -3.4
JOINT LOAD
10 FX 12.5
LOAD 4
ELEMENT LOAD
23 PR GY -1.2
LOAD 5
TEMPERATURE LOAD
15 17 TEMP 40.0 -25.0
Combination load case
Here, the user does not directly specify the load data, but instead asks the program to add up the results of the component cases - which are defined prior to the combination case - after factoring them by the user specified factors. It is characterized by the title which has the syntax
LOAD COMBINATION n
where "n" is the case number of the combination load case.
LOAD COMBINATION 40
3 1.2 4 1.6 5 1.3
What is a REPEAT LOAD type, and Which category does is belong to?
A Repeat Load type is a Primary load case. That is because, when the program runs into this command, it physically creates the load data for this case by assembling together the load information from all the component load cases (after factoring them by the respective load factors) which the user wants to "REPEAT". Thus, when you specify
LOAD 10
REPEAT LOAD
4 1.4 5 1.7
STAAD creates a physical load case called 10 whose contents will include all of the data of load case 4 factored by 1.4, and all of the data of load case 5 factored by 1.7. If we use the same data used in the definition of the primary load case above, STAAD internally converts the REPEAT LOAD case 10 to the following :
23 PR GY -1.68
15 17 TEMP 68.0 -42.5
What is the difference between a REPEAT LOAD case and LOAD COMBINATION?
The difference lies in the way STAAD goes about calculating the results - joint displacements, member forces and support reactions. For a load combination case, STAAD simply ALGEBRAICALLY COMBINES THE RESULTS of the component cases after factoring them. In the example shown above, it
gathers the results of load case 3, factors them by 1.2,
gathers the results of load case 4, factors them by 1.6,
gathers the results of load case 5, factors them by 1.3,
and adds them all together. In other words, in order to obtain the results of load 10, it has no need to know what exactly is it that constitues load cases 3, 4 and 5. It just needs to know what the results of those cases are. Thus, the structure is NOT actually analysed for a combination load case. With a REPEAT LOAD case however, the procedure followed is that which occurs for any other primary load case. A load vector {P} is first created, and later, that load vector gets pre-multiplied by the inverted stiffness matrix.
[Kinv] {P}
to obtain the joint displacements. Those displacements are then used to calculate the member forces and support reactions. Thus, the structure IS analysed for that load case {P}.
Why should the difference in the way STAAD treats a REPEAT LOAD case vs. a COMBINATION LOAD case matter?
Normally, if you are doing a linear static analysis - which is what a PERFORM ANALYSIS command does - it should make no difference whether you specify REPEAT or COMBINATION. However, if you are doing a PDELTA analysis, or a NONLINEAR analysis, or cases involving MEMBER TENSION and MEMBER COMPRESSION, etc., it matters. That is because, in those situations, the results of those individual cases acting simultaneously IS NOT the same as the summation of the results of those individual cases acting alone. In other words,
(Results of Load A) + (Results of Load B) is not equal to (Results of Load (A+B))
Take the case of a PDelta analysis. The P-Delta effect comes about from the interaction of the vertical load and the horizontal load. If they do not act simultaneously, there is no P-Delta effect. And the only way to make them act simultaneously is to get the program to compute the displacement with both loads being present in a single load case. A REPEAT LOAD case achieves that. A COMBINATION load case does not.
There are two ways. One can directly enter the reduced values for the sections using the PRISMATIC option as shown next
…
UNIT IN KIP
MEMB PROP
11 TO 21 PR YD 21.0 ZD 16.0 IZ 4321 IY 2509
The YD and ZD represent the overall dimensions which are used by the software to calculate the properties that are not input directly by the user. The IZ and IY represent the cracked section properties. Using the GUI, one can do the same from within the General > Property page. Click on Define within the Property – Whole Structure window and define the properties using the General option as shown next
Alternately one can define the cracked section property from within the modeling mode, by going to the General > Spec page. From within the Specification - Whole Structure dialog box on the right, click on Beam button and there is the Property Reduction Factors tab in there. Using that one can assign reduction factors to properties like cross sectional area and moment of inertia as shown next
The master slave specification is a method to model certain linkages between a control point (called a master joint) and a set of nodes (called slave joints) that are connected in the physical structure to the master joint through those linkages.
Those linkages cause the displacement of certain degrees of freedom at the slave nodes to be patterned after the displacement of those d.o.fs of the master node.
One of the limitations of the master-slave feature is that those degrees of freedom can be defined in the global directions only. Constraining a displacement or rotation along non-global direction is not available.
If you want to apply the constraints along directions which are inclined to the global directions, one way would be to define a dummy beam element connecting the two nodes along the inclination and assign a material having high E and zero density to that dummy member.
Product TechNotes and FAQs
Structural Product TechNotes And FAQs
Bentley Technical Support KnowledgeBase
Bentley LEARN Server
Bentley's Technical Support Group requests that you please confine any comments you have on this Wiki entry to this "Comments or Corrections?" section. THANK YOU!!!