General approach

  • Since the analysis procedure is basically numerical, and since it is based on the finite-element method, it will always be required to endorse a finer mesh (one with more smaller elements) in order to yield more accurate results.

  • In the following, we will adopt a procedure that maintains that general understanding.

Vertical Loads

  • A story-by-Story analysis under the effect of dead and live loads shall be performed using SAP2000, taking into consideration the following:

    • A finer mesh shall be used.
    • Columns shall be includes in the analysis (not as knife-edge supports).
    • Walls loads shall be included as line loads.

  • Design of slabs and beams (under vertical loads) shall be done using the results of the analysis
  • Column and wall normal forces due to dead and live loads shall be summed up for all stories.
  • Columns and wall moments due to dead and live loads shall be recorded at this stage.

Lateral Load Analysis (The Etabs File)

  • Build the Etabs model, taking into consideration the following.
    • For columns, use frame elements
    • For walls, use a horizontal mesh of 0.5-0.8ms and a vertical mesh of 1.2-1.8ms (an element 0.6X1.2 seems appropriate, also may be 0.6X1.75). Don’t forget composite columns (if any). Group sum each wall and each core separately (pier forces) are required stories.
    • For slabs include slab limits and openings ignore small openings since they add more shell elements. Add beam elements at their locations as frame elements. Discretize slab with the finest mesh possible (less than or equal 2.5 x 2.5m, if) possible. Try to have the mesh as regular as possible All points within each horizontal level shall be included in a separate diaphragm.
    • Use reduced stiffness 0.35 for beams and 0.7 for columns and walls, and 0.25 for slabs.

  • For the dead loads, add dead load as a distributed load. It includes flooring and wall weight (A value of 0.3 – 0.5 seems usually reasonable). Include the self weight in the dead load. (Including the walls as a line load is an option?!)
  • For the live load, add live loads as a distributed load.

  • For the earthquake loads. Use UBC’97 loads in both directions with the following options:
    • Include eccentricity 0.05 in each direction.
    • Include Ct
      • First option: Ct = 0.02ft (user-defind)
      • This includes Ta as ct hn ¾ (Eq. 30-8)
      • Second option: Ct = 0.02ft (Program Calculated)

Calculate T using equation from code and compares it with Ta according to Eq. (30-8). You have to include dynamic analysis option for this option to work. Otherwise, program sometimes chooses T wrongly as We recommend using this second option.

Third option: Ta is user-defined as ct hn ¾.

There should be a match between the Eq. (30-8) first and third options.

    • R = 5.5 (or as per code)
    • Soil type = Sc (or per code)
    • Z = 0.15 (or as per code)
    • These loads are ultimate. To get working loads divide by 1.4

For the wind loads using BS 6399-95, use the following options!

  • Exposure extents from rigid diaphragm
  • Wind direction Oo (x-direction) and 90 (y-direction)
  • Front Coeff CP = 0.8
  • Rear Coeff CP = 0.3
  • Exposure height: Top STORY & BOT. Story (exclude basements)
  • Include parapets (if any)
  • Effective speed ve (m/sec) = 45 (or as per code)
  • Size effect factor, Ca = 1
  • Dynamic augmentation factor, Cr = 0.25
  • Exposure width (table)         Calculated from diaphragm extents. This wind loads option is recommended to be used.

For the wind loads using ASCE7-98, use the following options:

  • Exposure extents from rigid diaphragm
  • Direction angle Oo (x-direction) and 90o (y-direction)
  • Windward Cp = 0.8
  • Leeward Cp = 0.5
  • Exposure height: Top STORY & Bot. STORY (exclude basements)
  • Include parapets (if any)
  • Basic wind speed v= 100 mph
  • Exposure category = C
  • Importance factory, Iw=1
  • Kzt = 1
  • Kd = 0.85
  • Gust factor G = 0.85
  • Exposure width (tables)          calculated form diaphragm extents

For the winds loads using UBC-97, use the following options:

  • Exposure extents from rigid diaphragm
  • Direction angle Oo (x-direction) and 90o (y- direction)
  • Wind ward Cq = 0.8
  • Lee ward Cq = 0.5
  • Exposure height: Top STORY & Bot. STORY (exclude basements)
  • Include parapets (if any)
  • Basic wind speed v= 100 mph
  • Exposure type = C
  • Importance factory, Iw=1
  • Exposure width (tables)          calculated form diaphragm extents.

For the response spectrum, use the following options:

  • UBC’97 response spectrum
    • Damping 5%
    • Scaling factor to equate base shear (may be you first need to run the program to get initial base shear values for response spectrum then calculate the scale factor)
    • Use Ritz vector.
    • Choose number of modes = 15
    • Use SRSS option for combination of modes.

Also, review these options before you run your file:

  • Mass per unit volume = 0.25 t/m3 (concrete)
  • Weight per unit volume = 2.5t/m3 (concrete)
  • Mass per unit volume = 0.785 t/m2 (Structural Steel)
  • Weight per unit volume = 7.85 t/m3 (Structural Steel)
  • Ec = 140000(fcu)1/2   t/m2  for a given fcu (concrete)
  • Es =  21000  t/m2                                 (Steel)

For the mass source data, use:

  • Mass defined from loads
  • Mass source data
  • Mass lateral lump mass from mass only at stories load No   Yes
  • Mass source loads

Load   multiplier

Dead       1

For the P-Delta effect, choose:

  • Number of iterations  =  2
  • P-Delta load combination = 1.2D + 0.5L

  • It is not recommended to perform a P-Delta analysis, unless a run is done without P-Delta to know that everything is ok, since P-Delta run takes more time.

No load combinations are recommended to be included. No design using Etabs is also recommended.

Consider reversibility of lateral loads
Results of all loads are working loads, except for earthquake load, which are ultimate.

Analysis Results & Design

Drift: drift values should not exceed 0.005h for lateral load cases

Slabs: slab design is basically dependent of item ‘2-vertical loads’. However, effect of lateral loads should be checked from step ‘3-lateral load analysis’, especially in case where framing in based on flat slab behavior, rather than connecting beams

Beams: Beam design is based on superposition of vl loads from ‘2-vertical loads’ and lateral loads from ‘3-lateral load analysis’.

Columns and walls

  • Vertical Loads (working loads)

  • Normal forces: from step ‘2-vertical loads’. Results of normal forces due to D.L and L.L from step ‘3-lateral load analysis’ should be overviewed only, since meshing is usually course (wide) and doesn’t produce accurate results and may be wall loads were not entered as line loads. Note that normal forces in columns and walls intersecting retaining walls drop in value in basements, due to interference of retaining walls.

  • Moments: use recorded values in step ‘2-vertical loads’ or from minimum sec (as per codes). Results of moments from step ‘3- lateral load analysis’ should be overviewed only, since meshing is usually course and doesn’t produce accurate results, in addition to other reasons mentioned above.

  • Lateral Loads (working loads)

  • Normal force and moments from ‘3-lateral load analysis’ for max. cases of wind, EQ, and response spectrum shall be used for design.

  • Note that EQ forces should be divided by 1.4 in order to be service loads.

  • Use PCA column with pervious service straining actions for column and wall design.

Leave a Reply