Need a book? Engineering books recommendations...

# Re: Structural Engineering Rules of Thumb

• To: <seaint(--nospam--at)seaint.org>
• Subject: Re: Structural Engineering Rules of Thumb
• From: "Daryl Richardson" <h.d.richardson(--nospam--at)shaw.ca>
• Date: Fri, 5 Feb 2010 11:58:16 -0700

Vish,

This is, indeed, a useful "rule of thumb"; I use it a lot. It can, in fact, be improved upon.  Consider the following items.

1.)  Consider Defl. = m*L^2/(K*E*I)

2.)  For uniform load, K = 9.8, ~ 10 as you have pointed out.

3.)  For concentrated load at centre, K = 12.

4.) For equal but opposite end moments (M = constant throughout the beam length). K = 8.

Given the above, one can "inspect" the bending moment and ( without calculation) arbitrarily adjust the K value.  Perhaps you can use K = 9, or K = 11, to get even better results from the "rule of thumb" you have proposed.

Regards,

H. Daryl Richardson
----- Original Message -----
Sent: Friday, February 05, 2010 4:30 AM
Subject: RE: Structural Engineering Rules of Thumb

 Thanks for this initiative in the seaint list Let me also make a humble contribution.   Here is one rule of Thumb I have found useful during may years as a designer of steel structures.   If you need to do a quick check of the maximum deflection of a steel beam subjected to some udl and some  point loads, you can use the following easy formula   defln = M x L^2 / (10 x E x I)   M = max BM L = span E = Elastic modulus I = Moment of Inertia. Take care to use a consistent set of units. You will be surprised to note how close you are to the exact value if you do a rigourous calculation. This is for beams with a mix of udl and a few point loads scattered here and there along the span. Don't use this for a standard beam with a standard udl only or beams with just one point load. You can use it with beams with three or more point loads and some udl of partial udl.     In nearly all cases the defln value using this simple formula will be sufficiently accurate for  a code compliance check compared to the rigorous calculation for each point load and the udl and the sum of these deflections.   Built up girders are  very common in India, where we have very limited rolled sections available and the deepest is only about 24" For cooking up a suitable size for a built up girder I use the following thumb rule.   Let half the cross sectional area be taken up by the web Each flange can use up 25 percent of the required area.   You can decide an optimum depth for a built up girder using the simple formula optimum depth = 1.2 x Sqrt ( S / t)   S = required sectional modulus for the given moment t = web thickness. Use these as a good value to start the design with and fine tune it if needed.   A third thumb rule I used is for choosing gusset thickness in trusses  with no fancy calculations   Use 8 mm thick gussets for forces upto 20 tons   10 mm  thick gusset for forces from 20 to 45 tons   12 mm  thick gussets for forces from 45 to 75 tons   14 mm thick gussets for forces from 75 to 115 tons   16 mm thick gussets for forces from 115 to 165 tons   18 mm thick gussets for forces from 165 to 225 tons   20 mm thick gussets for forces from 225 to 300 tons.   The forces are the maximum forces in the support diagonals and usually the same thickness is adopted uniformly but in large span trusses, we often reduce the gusset thickness as we approach the mid span.   These tips are from an old now dog eared copy of a Russian Text on Steel structures called "Design of Metal structures" by Mukhanov   I purchased this book at the start of my professional career  in 1974 for the then princely sum of Rs 7.25 (approximately  15 cents at today's prices but perhaps about 35 to 50 cents at 1974 prices. I can't recall the exchange rate in 1974)   Regards Vish