STAAD Direct Analysis Method FLEX and AXIAL Parameters

Which members are intended to be included in the parameters FLEX and AXIAL for a braced bay when using Direct Analysis Method? Does this include just the braced columns, beams and bracing or the entire frame (including gravity only beams/columns)? I'm a little confused by the definition provided by STAAD. The definition STAAD provides is as follows:

FLEX, identification of members whose flexural stiffness is considered to contribute to the lateral stability of the structure, along with the initial value of τb that should be used. Members listed with FLEX will have their EI factored by 0.80 times τb while performing the global solution. The final member forces and code check will be with 100% of the flexural stiffness.

AXIAL, identification of members whose axial stiffness is considered to contribute to the lateral stability of the structure. These members will have their EA factored by 0.80 while performing the global solution. The final member forces and code check will be with 100% of the axial stiffness.

 

I am currently trying to model Example 2.1 in AISC Guide 28 in STAAD and analyze it using direct analysis method and compare the results to those provided in the example. I think it may be worth noting that the example usese effective length method, where I am using direct analysis method. Currently, my results for the moment in the beams and axial loads in the columns match that of the example (or are within 0.50%) except for the axial loads in the columns for combinations 5, 6 and 7. The results for these combinations have been up to 30% off. These are the only combinations that include the wind load; which is why I think the error might be coming from the columns in the braced bay. Any help is appreciated, thanks!

 

Lauren

Parents
  • I think to prevent artificial modification of structural stiffness leading to unintended distribution of forces, you should apply the FLEX and the AXIAL to all members in your structure and NOT just the ones contributing towards the lateral stability. The section AISC C2.3(1) of the AISC 360-10 specification, has a user note related to this.

    The DIRECT ANALYSIS and EFFECTIVE LENGTH methods are significantly different and so I would not expect the results to match.



  • Thank you, that seems reasonable to me. I tried running another analysis and the results did come a little closer but are still about 30% off. I agree that the two methods would have some variations between them but I didn't think it would be as significant as they currently are.

    Another area that I'm a little unsure of in this model is how the bracings are modeled. In the example they are specified as 1" and 1.5" tension only rods. To model this I deleted the bracings that would be in compression from the model and modeled the ones that would be in tenion as prismatic circular members. I noticed that in the deflected shape of the structure, the bracing has significant deformation. Is this expected or is there a better way to model these shapes?

    Thank you again.
Reply
  • Thank you, that seems reasonable to me. I tried running another analysis and the results did come a little closer but are still about 30% off. I agree that the two methods would have some variations between them but I didn't think it would be as significant as they currently are.

    Another area that I'm a little unsure of in this model is how the bracings are modeled. In the example they are specified as 1" and 1.5" tension only rods. To model this I deleted the bracings that would be in compression from the model and modeled the ones that would be in tenion as prismatic circular members. I noticed that in the deflected shape of the structure, the bracing has significant deformation. Is this expected or is there a better way to model these shapes?

    Thank you again.
Children
  • It is OK to define the braces as prismatic circular sections. However you may assign the TRUSS specification to these rods as otherwise these would take up moments/shears. If you still have doubts after making these changes, please upload the .std file and I will take a look.



  • I tried to give the truss specification to the rods; however, STAAD gave me an error message almost immediately when I ran the analysis. I think  this is because I am using the direct analysis method. I have uploaded the file using Bentley Secure File Upload.

  • While taking a look at the model, I noted a few points

    Point 1

    The error was stemming from the fact that MY MZ releases were applied to these members in addition to the TRUSS specification. So you need to change the MEMBER RELEASE as shown next

    MEMBER RELEASE

    *4 5 9 10 14 15 19 20 24 25 28 30 START MY MZ

    *4 5 9 10 14 15 19 20 24 25 28 30 END MY MZ

    4 5 9 10 14 15 19 20 24 25 START MY MZ

    4 5 9 10 14 15 19 20 24 25 END MY MZ

    Point 2

    In your DIRECT ANALYSID definition, the FYLD values specified as 7200/ 5184 are not consistent with the last unit command before that, which is Inch Kip

    DEFINE DIRECT ANALYSIS

    FLEX 1 ALL

    AXIAL ALL

    FYLD 7200 LIST 1 2 4 TO 7 9 TO 12 14 TO 17 19 TO 22 24 TO 27

    FYLD 5184 LIST 28 30

    So these should be changed to  

    FYLD 50 LIST 1 2 4 TO 7 9 TO 12 14 TO 17 19 TO 22 24 TO 27

    FYLD 36 LIST 28 30

    Point 3

    You have specified the notional load 10 and have used that in various REPEAT LOAD cases. However the load case 10 only consists of notional load generated based the D case ( load case 1 ) only. When you are combining D+L or D+L+LR with appropriate factors, the notional load would have to be based on the vertical loads for each case. So instead pof having the load case 10, you specify the notional load as part of each load case itself. This is illustrated below for two cases 102 and 104. Similar changes has to be made for the other cases too  

    *LOAD 102 COMB2 = (D + L) + N

    *REPEAT LOAD

    *1 1.0 3 1.0 10 1.0

    LOAD 102 COMB2 = (D + L) + N

    REPEAT LOAD

    1 1.0 3 1.0

    NOTIONAL LOAD

    1 X 0.002 2 X 0.002

    Similarly for load case 104 it would be

    *LOAD 104 COMB4 = (D + 0.75L + 0.75LR) + N

    *REPEAT LOAD

    *1 1.0 2 0.75 3 0.75 10 1.0

    LOAD 104 COMB4 = (D + 0.75L + 0.75LR) + N

    REPEAT LOAD

    1 1.0 2 0.75 3 0.75

    NOTIONAL LOAD

    1 X 0.002 2 X 0.0015 3 X 0.0015

    Point 4

    If the design is to be carried out based on the AISC 360-10 and following the ASD approach, the code assignment should be changed from

    PARAMETER 1

    CODE AISC

    to

    PARAMETER 1

    CODEAISC UNFIED 2010

    METHOD ASD

    Point 5

    Since the code check cannot be carried out for the bracing rods, instead of using CHECK CODE ALL you should use

    CHECK CODE MEMB 1 2 4 TO 7 9 TO 12 14 TO 17 19 TO 22 24 TO 27

    Point 6

    The parameters LY and LZ for braces are not required as these cannot be designed anyway.

    LY 0.0833333 MEMB 28 30

    LZ 0.0833333 MEMB 28 30

    A modified file is attached for your reference.

    DOM ASD 6-28-16_modified.std