Simply supported?

[CharlesJenkinson]

Hello all,

I'm a 'mechanical', looking into the subject of 'structural' - wondering if it's worth expanding my horizons. Perhaps that's the first question, Is it worth it? Secondly, I'm doing some initial reading with the purposes of trying to move into the area of designing Mez Floor structures, and as far as I can gather that involves understanding and applying BS 5950, as a starter / minimum. Are there any texts or pointers out there that can illucidate the differences between the 3 basic approaches to beam (and consequent joint) design, which are simply supported, rigidly supported, and 'something else' inbetween...? I've navigated my way to portal frame design in an online reference and I'm starting to understand the sorts of approaches, but IMO the BS5950 does little to make itself understandable about the '3 basic approaches'. If it helps to steer me in an appropriate direction (either away from or toward 'structural'), my background is 14 years in engineering with an MSc, mainly in analytical disciplines, FEA, rotating machines, fatigue and some dynamics. Thank you, Charles Jenkinson

 

[HamishMcTavish]

Hi Charles,

Try looking at a book like "design of structural elements" by Chanakya Arya (i only refer you to that because i have a copy... there will be other similar titles by other authors!)

remember that 5950 is for structural steelwork which may be a sledgehammer to crack a nut depending on what the brief is.

Also, there's a whole load of stuff on the BCSA website, construction guidance, connections, and steelwork - all downloadable as pdfs

cheers and good luck, HM



No more things should be presumed to exist than are absolutely necessary - William of Occam

 

[valleyboy]

If you are only concerned with the design of mezzanine structures, then these are usually designed as 'simple ' structures as defined in Clause 4.7.7 of the code.

Beams are simply supported, which simplifies the beam element design. Columns are usually designed in accordance with Cl. 4.7.7. The colum effective lengths will depend on whether your system is braced or un-braced (this will depend on whether your client can accomodate vertical bracing in the area beneath the mezzanine). If the columns are unbraced then their effective length will be greater than 1.0 - some guidance is given in one of the Apendices to BS 5950. I dont have a copy of the code to hand at present, so can't give you the exact reference.

Obviously, if you are looking to move into more mainstream steelwork design, then there are a range of publications which will help - pretty much anything published by the SCI - but I always find the steel designers manual a good reference.

 

[GerhardSA]

hi Charles
try these websites for some good basic principles and some background.

http://darkwing.uoregon.edu/~struct/courseware/461/461_lectures/461_lectures_index.html

http://www.bgstructuralengineering.com/

 

[CharlesJenkinson]

Thank you! - some great help there. Just narrowing it down a bit more... In terms of understanding how the 'simple' analysis is done. Take for example the bolted web cleat - the typical L-shaped loose bracket that would be used for bolted joint, joining purlins and main beams together on a mez floor. I'm not convinced it is a 'simply supported' joint because when the bolts are tightened, it would transmit some moment from the beam through to the column. I concur that it wouldn't transmit all the beam moment into the column due to its limited joint stiffness, but it would certainly prevent a beam deflecting as much as a simply supported beam that can free pivot where it joins the column.

 

[valleyboy]

If you have a look at the SCI green book - simple connections - a web cleat arrangement is defined as a simple connection - ie beam moment at mid span is wL2/8, and the column moments are nominal moments due to notional beam eccentricity as defined in Cl. 4.7.7.

 

[GerhardSA]

Don't know if this is what you're looking for but.

the way I see it is to always design for the worst case scenario. With pinned joints the isolated member forces will be greater. ex a simply supported beam has a bigger bending moment when it is not restrained at the ends.

the opposite is true for connections where the forces would be greater if it is fixed. And I usually design for a combination of the two, ex reduce the forces in the pinned beam somewhat and reduce the forces at the rigid connection. This is fairly easy and quick for small structures with lots of variables.

 

[valleyboy]

I must confess I don't follow the 'hybrid' method that GerhardSA does.

If you assume partial end fixity, you must design the columns for the resulting fixed end moment, which will usually increase the column size.

(In columns in simple construction, the dominant load effect is axial load, whilst in continuous construction the dominant effect is bending).

I take a view on the connection type, and design the members for the resulting forces accordingly.

 

[CharlesJenkinson]

I've checked out the bolted cleat joint in the Steel Designers manual. I can sort of accept that a bolted cleat joint has enough elasticity that it does not transfer significant moment into the column. Another aspect of a truely simply supported beam is that there will be no tension along it's axis (i.e. in the idealised diagram, one end is mounted on a triangle and the other end on a roller). Is this tension (making the column like a cantilever bolted to the floor) also neglected? and/or is the floor pin jointed for simply supported?