structural engineering rules of thumb

huangyhg

structural engineering rules of thumb
there have been some great posts skirting around this subject, however i really think we need a tips & tricks faq.  with that goal in mind, i thought it would be good if we all posted the routine rules of thumb we use in our daily work, in order to perhaps learn some useful shorthand and tips or tricks.
structural engineering rules of thumb:
1.  beams need to have shear effects considered specifically (increased deflection, possibly reinforcement requirements in concrete, etc) when span/depth <=10
2.  beam design is normally deflection governed when span/depth >=25.
3.  when checking drawings, looking at moment/plastic section modulus is a good guage of a section's sizing, even if using limit states design.  knowing the extreme fibre stress is a good "feel" for the beam size.
4.  always consider a minimum accidental eccentricity of 100mm in your construction.  increase this to 150mm in residential work.
5.  design shelf angles for the load at the very tip for strength (uls), centre of bearing for serviceability (sls).  this ensures that any rotation of the beam at the support does not lead to overstress in the fixing; particularly for stiffened angles.
6.  when in doubt, add confinement to concrete.  curtailment of reinforcing should occur at a distance of 130% development length past the point where strength is last required, or ld+d from support, whichever is greater.
7.  to minimize the risk of timber floors (and all high frequency floors; applies to cold formed steel as well), check that the deflection is no greater than 1 to 2mm under a 1kn point load at centre.  do not consider t-beam stiffening effect for this check unless the plywood is glued and screwed; slip and fastener loosening may not permit adequate composite action otherwise.
8.  for steel and concrete beams, check the estimated natural frequency, equal to 18/sqrt(total deflection in mm), result in hertz (hz).  use anticipated actual loads in this check (thus typically 0.25kpa to 0.35 kpa) rather than full sls loads.  a result of 15hz or higher should be double checked with the point load check, a result between 8hz and 15 hz is likely okay, with likihood of difficulty increasing as the result decreases, and anything between 5hz and 8hz should be subject to a full accelerative methodology vibration check (such as the atc guideline or cisc guide 11).  picking the loading is very important, and entirely subjective; a good guide is to consider 30% of your floor load as the likely "routine" load.   that way you are basing the load used on the code's anticipated exposure loads for the floor type.  remember that vibration problems normally happen under light loading.
9.  for good ground checks in the field, get a metal or timber block made up which should impose the ultimate bearing pressure required of the soil.  stand on this for a count of five anywhere you have some doubt over good ground.  an indentation of anything more than a mark (so say greater than 1mm) is considered a failure.  example:  i weight 100kg, my typical "good ground" value is 300 kpa ultimate (rupture) bearing pressure, thus my block is roughly 57mm by 57mm.
i'll keep posting as i think of more.
cheers all,
ys
  
b.eng (carleton)
working in new zealand, thinking of my snow covered home...
check out our whitepaper library.
good start.
points 1 and 2 are material specific?
3 - agreed.  in uk we have a limit state code (similar to lrfd) the section properties are given in a separate document (blue book) which can readily be used for simple cross checks.  checking designs could really do with a whole topic of its own...
point 9 - i like this.  i've never done it.  it's only likely that i'd be responsible for this sort of thing on small projects but i might just make up some blocks to test it out.  guess i'll need to weigh myself.  maybe later...
a good rule of thumb our office uses is stay away from the metric system...hahah.
if any of you have been or are planning on going to the nascc, you may have heard of the course "rules of thumb for steel design" given by socrates ioannides of structural affiliates international.  well worth the hour and a half. plus he has a handy pocket quide that he hands out at the end.  

concrete slabs:  20 to 25 to 1 span to depth ratio
concrete beams:  2 to 1 depth to width ratio
  
mike mccann
mmc engineering
as = m/4d
inches of course
does anyone have a rule of thumb for the aspect ratio of a diaphragm in which the chord forces can be neglected??
what's an inch? is it one of those archaric units no longer used in the advanced world?
25.4mm per inch.
i'm going to upset people but the us and other nations still using imperial units should convert to metric.
it makes much more sense, it is all based on water.
youngstructural
excellent post and i would support an faq section on rules of thumb.  here's a contribution.
base width of a stable retaining wall is typically 55% of retained height. (considering zero passive pressure at the front of wall, a level backfill and zero surcharge.)
the presentation by socrates ioannides is available here as "e19 rules of thumb for steel design" as a sound file with accompanying slides.  well worth listening to for anyone interested in rules of thumbs.  some of the relationships this guy has come up with are almost spooky.