Expert: Arshad Khan - 5/24/2010

Question
QUESTION: Assalamu alaykum! It's good that muslim specialists in sciences are expanding... Could you help a brother.. im a fresh, newly licensed C.E. My task is to analyze/design a steel truss spanning 30m between R.C. columns with 8m on-center distances...so the truss spacing is also 8m.. the warehouse is supposedly 150m(or more) in length,30m in width, L-shaped(but the L-part is only small in ratio to its length).. My problems arise when:

1) Analyzing the truss, there is a laaarge gap between actual conditions and ideal..How am i to analyze a structure? Books analyze roof trusses as simply supported, but i think actual trusses have both ends hinged!(or even fixed)..

2) should i rely in simple analyses? To what extent is structural analysis carried out in practice? Do you know any good links?

3)The large uncertainty overwhelms me,and i dont have so much clue...Earthquake possibilities petrifies me!

many thanks...

ANSWER: Hi Nordz,

1. As far as trusses are concerned, simply supported and hinged at the supports is the same thing. You can assume hinged if the columns can support the gravity load and lateral loading by cantilever action. Sometimes, if columns are also in steel, you can design the trusses as fully fixed to the columns, in which case the whole structure behaves like a portal frame.

2. For trusses with hinged/SS supports, you can rely on simple analysis such as at this site:

http://www.civilengcalculations.com/index.html (check under Bridge Truss).

Analysis is vital to the design of all structures but the accuracy of the results depends on the accuracy of the input data (sizes of members, materials properties, loading and boundary/support conditions).

3. You can reduce uncertainty by clearly defining the input data as mentioned above. Earthquake loading on simply supported trusses can more or less be ignored as the columns/shear walls will resist all lateral loading, but you may need X-bracing within the plane of the roof,and along the periphery in the vertical plane, in order to stabilise the trusses from deflecting out of the vertical plane.



---------- FOLLOW-UP ----------

QUESTION: thank u so much..a very useful advice.. What just worries me is the scale of the structure and how much we idealize... I can't help but get a feeling and doubt of oversimplification.. Is there any actual sample analysis in practice found in the that u know?Many thankz again.

Answer
Hi Nordz,

You are correct that some idealisation is not accurate- such as assuming all joints in trusses as being pinned. In large span structures, where the struts and ties are made of heavy sections, there will be local bending stresses near the nodes even though the truss is idealised as pinned joints- so you need to design for these additional bending stresses, in addition to axial stresses.

You can find examples of analysis and design in many text books, such as the Steel Designers Manual.