beam deflection re. stress not force - useful?!

Aug 16, 2023 Last reply: 2 years ago 5 Replies

Hello everyone



While doing the project



"Rectangular Hollow Section beam calculator"



230724_ebbeam_rhs.html (-> 230720_ebbeam_rhs.php) which I mentioned here as thread "web-based beam calculator - RHS's only"

I needed a neat way to go from stress in a beam to its deflection. So came by this unexpected derivation.


  • have you seen it before?
  • is it known to be useful?
  • any thoughts???

Stating again - if you think in terms of stress not force, you get an expression for beam deflection which has uses...



Regards, Rich Smith


--------------------------------



Derivation - deflection of symmetrical section in relation to stress



This expression for elastic deflection of a beam would be very useful "on-site" because everything you need to know can be obtained with a tape-measure (and a length of string!) to calculate the stress on a beam...



Symbols - all as familiarly used



"I" = Second Moment of Area "Z" = Section Modulus "M" = a Moment; a bending and/or turning force "F" = a force (in Newtons; N) "L" = length of the beam from supports to load (length beyond supports is irrelevant) "H" = height of symmetrical section in direction it is being bent "sigma" (Greek symbol) = stress (in N/m^2) "." == "*" meaning multiply


(1) M=FL/4 ("simple" beam)



(2) the massively significant equation which combines material property and geometric property... M=sigma.Z



noting "Z" = I/half-height for a *symmetrical* section Z=2I/H thus M=2.sigma.I/H



(3) and deflection d=F.L^3/48EI ("simple" beam - supported each end; load in the middle)


Restating and taking the derivation a step further: M=FL/4 => F=4M/L M=2.sigma.I/H d=FL^3/48EI



F=4M/L substituted into d=FL^3/48EI d=(4M/L).L^3/48EI



Cancelling L^-1 and L^3 to L^2



4/48=12

d=ML^2/12EI



From "fundamental" M=sigma.Z for a symmetrical section (eg. I-beam; circular, rectangular or square hollow section) we derive M=2.sigma.I/H



Substituting M=2.sigma.I/H into d=ML^2/12EI



(2.sigma.I/H)L^2/12EI=2.sigma.I.L^2/12EIH



"I" cross-cancels completely.



2/12=1/6

So... d=sigma.L^2/6EH


If you had a beam and stretch a string between its ends, a tape measure / rule in the middle between string and beam would measure the deflection. The tape measure also gives you the length of the beam and the height of the beam (or "depth")



eg. a 200x100x12RHS used on its major axis would be measured as having a height of 0.2m (200mm)...



as d=sigma.L^2/6EH then sigma=6EHd/L^2



So there with L, d and H given by your tape (and d needing your string). Steel always has an elastic modulus E of around 210GPA (210e9Pa) - so you know E already.



You can either deduce stress "sigma" from a measured deflection sigma=6EHd/L^2



or calculate the deflection you would get at the yield stress and go measure how much of the beam capacity is currently being used. d=sigma_yield.L^2/6EH You can take the nominal yield stress specification of the steel as its yield stress - so for an "S355" steel (which would be typical of an RHS) that would be 355MPa = 355e6Pa



As I said, this can only be used for symmetrical sections like I-beams and RHS's and cylindrical tubes. The "H" has that implication that the section must be symmetrical (you could not use this for eg. angle-iron ("L-shape") which is not symmetrical).


Hello everyone

While doing the project

"Rectangular Hollow Section beam calculator"

230724_ebbeam_rhs.html (-> 230720_ebbeam_rhs.php) which I mentioned here as thread "web-based beam calculator - RHS's only"

I needed a neat way to go from stress in a beam to its deflection. So came by this unexpected derivation.

---------------------

I had a slightly different approach. The deflection that matters most is near or at the yield point, and can be figured or found with an on-line calculator, so I placed a marker of that height at the beam center. Instead of string I sighted down the beam from the end, standing out of the way in case something went wrong. The idea was to relate deflection with a load to without it, to know when to stop when straightening the slightly bent second hand steel.

After reaching the yield point I counted hydraulic pump strokes and increased the count until the beam (3" and 4" C channel) relaxed straight, since permanent deflection isn't predictably linear in the strain hardening range above yield.

formatting link
I could have measured deflection with string but the chained setup wasn't stable enough to risk getting that close. I hadn't yet determined where to drill the bolt holes that would have permitted more stable connections.

One of the four channel sections clearly was higher strength steel and required much more deflection to straighten. It also had slightly smaller dimensions and bent more below the yield point, allowing the trolley track to twist and deflect sideways instead of sagging at max proof loading, though it held my 2100 Lb log OK. The scrap dealer had no more to replace it.

In the calculators max deflection is given as a fraction of length, such as L/360 for floors that support plastered or sheet rock walls, to avoid cracking. They give it for the load centered or at two other points. It's for the live load, the dead load deflection occurs before finishing the walls. For unfinished areas I've seen L/180 suggested and it's what I use. IIRC the (unmeasured) deflection to straighten the high strength channel was roughly L/20. jsw

Very appreciating considered engaged informed response. How did I not guess that someone here would relate to some very like experience! :-) Will read again tomorrow and hopefully respond more. Regards, Rich S

Very appreciating considered engaged informed response. How did I not guess that someone here would relate to some very like experience! :-) Will read again tomorrow and hopefully respond more. Regards, Rich S

------------------

While most of what I've been asked to design and build was electronic there has been some necessary mechanical content, and more for my home projects, the bucket loader, sawmill and gantry hoists, so I've tried to learn enough engineering theory and practice that they don't break or weigh and cost 10x more than needed. Some electrical engineers I worked for knew little outside their specialty and couldn't solder.

I had almost no engineering perception - all my knowledge was fundamental science.

When I realised my scientific career dreams were simply not going to happen, working in construction for an hourly wage just to pay for a meagre existence while rehabilitating into society :-) my engineering knowledge lit-up. Going from almost zero to "these are somewhat advanced topics we are talking about here" (they weren't, but it takes a new part of you head with a new region of "landscape" perceiving and analysing structural things in what becomes a familiar way).

I remember as if it were yesterday just over 15 years ago waiting for the train to work to leave London Waterloo station and looking up at the roof-trusses and I started to "trace" their design and see the forces flowing.

Euler-Bernoulli beam and Finite Element Analysis modelling came when I came back from Turkey only 8 years ago. Having realised I could have so speeded-up my work as a welding engineer if I could independently do at-least-approximate analyses of structures and challenge things I knew were wrong but had not the means to estimate.

I "got" Finite Element Analysis modelling first - but realised it didn't suggest how to design long structural elements - beams and columns. I learned (Euler-Bernoulli) beam while working as the gateperson at a bricks-and-mortar residential construction site.

I had known for years into decades that I had no engineering perception and needed to do something about it, but had no idea how. Funnily enough (?), where as a welder and steel-erector ("ironworker") was where it happened...

Regards, Rich S

I remember as if it were yesterday just over 15 years ago waiting for the train to work to leave London Waterloo station and looking up at the roof-trusses and I started to "trace" their design and see the forces flowing.

-----------------------

I do that too, most recently to a circus-type tent.

The knowledge was gained by trial and costly error:

formatting link
We still aren't immune to error:
formatting link
FIU was supposedly the expert in that construction method. "Unlike most bridges in Florida, the design for this project was overseen by the university, not the Florida Department of Transportation (FDOT), in a program known as the Local Agency Program (LAP)."

"Florida International University is known for its expertise in accelerated bridge construction (ABC) and has attracted international scholars as PhD students."

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required