A ? for you physics gurus
Alan Young
>If you have a 3/8" metal rod that is say 48" long and a same sized (3/8") square piece also 48" long. Is one more apt to "bend" or "sag" than the other simply by it's weight"
Alan
Knowledge You Can Build On — Est. 1998
A ? for you physics gurus
Alan Young
>If you have a 3/8" metal rod that is say 48" long and a same sized (3/8") square piece also 48" long. Is one more apt to "bend" or "sag" than the other simply by it's weight"
Alan
Re: A ? for you physics gurus
Jack Hartley
>The round rod would bend. The round rod has less cross-section area than the square section. Therefor less area to resist bending.
Re: A ? for you physics gurus
Bill Turpin in WNC mountains
>The square corner adds rigidity also.
Re: A ? for you physics gurus
DCinSTL
>Good question. There is more to this question than first meets the eye. The square bar has more mass, therefore more weight. It would be interesting to see the formulas used to determine the correct answer.
That's what I was thinking....but I asked
Alan Young
>one of the young bright "kids" down the hall this same question and he through a bunch of equations at me and said "Yup- a square is stiffer

...See?"
OMG I should have said "threw" not
Alan Young
>"through". I hope Ellis doesn't throw me out!
If they started to penalize for grammatical
JL
>and spelling errors - they would lose half of the posters.
Only Half????
Kneale Brownson..Awaiting return to CO snows
>
Re: A ? for you physics gurus
John McGaw
>It has been a lot of decades since my last physics class but my spur-of-the-moment calculations suggest that there are two extreme cases:
1) if the square rod is horizontal (flat surfaces up and down) it will sag more since it weighs 1.269 times more while the vertical section is 3/8", identical to the round
2) if you rotate the square rod so that the corners are up and down (rotated 45 degrees) you will gain on the vertical section which is now 1.41 times more than the round's and this should be enough to overcome the weight disadvantage meaning the round will sag more
Did that make any sense to anyone else?
We need an engineer
Ellis Walentine
>I'm used to thinking of deflection in terms of the section modulii of wooden beams. I imagine steel follows similar rules, and I'm not sure how to calculate the modulii of round or 45�-oriented square sections. I suspect there is a logical limit here, too. In other words, whatever the span is, both these sections may be plenty stiff enough to resist sagging and however much deflection due to distributed or point weight.
Like I said, we need an engineer. Dominic? Anybody?
Ellis
Re: A ? for you physics gurus
Alex Grigoriev
>In layman terms, the material in the middle contributes the least, and the material on the upper and lower edge contributes the most. This is why they make I-beams. Thus, the square rod will sag less.
Not 1.269 times more
Henry Higginbotham
>Not 1.269 times more, but 1.269 times as much (if the math were correct; I calculated 1.273). If something is one time more, it's twice as much. Two times more--three times as much. 50% more--1.5 times as much. Yes, I know you've heard it that way a bazillion times, but that doesn't make it right. This is math being abused by language (which several WCers tried characterize as irrelevant a day or two ago). It also leads to truly absurd phrases such as "three times less than . . . ." Do the math on THAT one.
I usta was one
Neal in San Jose
>Back in a previous life one of my job responsibilities was designing machines, and determining shaft deflection, etc. was a daily occurance. I've forgotten most of that, but think I remember enough to get us thru this example.
To figure the deflection of complex shapes, like an "I" beam, divide the shape into smaller, simple shapes and then figure their center of deflection from a reference axis. Clear as mud so far?
So, for our two shapes, lets assume they will both deflect around an axis on their horizontal center line. So for the round rod, we end up with two "D" shapes on their sides, one facing up and one facing down. Their individual centers of deflection would be the same as their center of mass or geometric center, whichever is easier for you to think about. It is clear, I hope, that the center of deflection would be pretty close to the straight side of the "D", so the distance times the area of the shape would be "strength" for our purpose, of this shape. Multiply by two for the top and bottom "D", and you get a number that represents this shape's "strength".
For the square, do the same divide in half routine, and you end up with two rectangles on their sides. The center of deflection is further from the axis, and the rectangle has more area, so the distance times the area, or "strength" is greater for this shape. The square would deflect less.
My head hurts! I'm afraid to go looking for my old text books; I might get eaten by a dust bunny.
Re: Not 1.269 times more
John McGaw
>Yes, an inexact choice of words. I did a few quick calculations and then posted a quick answer without applying my rhetorical skills which are nearly as rusty as my physics skills.
Re: A ? for you physics gurus
GalenV in Iowa
>I don't think you would see the difference as if a carbon steel material the square deflects 0.0012 inches and the round 0.0015 inches. (Square not standing on point or z axis)
Moment of inertia of square is I=t^4/12 or I= .375^4/12=0.00165 in^4
Moment of inertia of round is pi*d^4/64 or I= 3.147*.375^4/64= 0.00097 in^4
Deflection at mid span of l=48" rod is (5*w*l^4)/(384*E*I)
E=29,000,000 for carbon steel
Round w=0.03 lb/in and square w=0.04 lb/in
Or something very close to that.
not how I-beams work
Bill Tindall, E.TN
>I -beams are I-beams to prevent deflection orthogonal to the loading. . A beam of rectangular cross section will twist and bend when it is loaded. Once it twists along it axis the web is no longer ideally suited to counter the load. The height of the web contributes mainly to resisting deflection parallel to the load while the top and bottom flanges mainly contribute to resisting twisting.
Re: A ? for you physics gurus
Hal Smith
>The formula for deflection of a uniformly loaded simple beam is:
5 x w(load per unit of length) x L (length) to the fourth power, all the above divided by 384 x E (modulus of elasticity) x I (moment of inertia).
In the case at hand, w is simply the weight of your rod probably expressed in pounds per inch. Use 484#/cu.ft. for weight of steel to calculate w. Use thirty million for E. I for a rectangular section is b(width) x h(height) to the third power so for a square, I is simply the length of a side to the fourth power. I for the circular section is pi x diameter to the fourth power divided by 64.
Be sure to use consistent units for w & L. If w is in #/in., L is in inches. For your sections and L = 48in., I calculated deflections of .0551" and .0734" for the square and circular sections respectively. This calc done while on my second martini.
P.S. A long time ago, I was a structural engineer. Back in the days
when we had to figure out things like this for ourselves instead
of relying on computers.
Regards, Hal
Re: A ? for you physics gurus
Hal Smith
>P.S. As suggested by Bill Tindall, your square section at this length is not well suited to resist out of plane bending and will likel buckle and twist.
Re: A ? for you physics gurus
Pat Riley
>If we assume that both rods are suspended only at the ends, and they are equally loaded (due to gravity), the deflection at any point is:
y=[Wx(l-x)/24EIl]*[l^2+x(l-x)]
Maximum deflection at Center is:
y= [5/384]*[Wl^3/EI]
In both cases (the round and the square) the only difference in the calculations are the values for I - the Moment of Inertia and W - the load.
The load of the round rod is lower - 78.54% of the square rod.
Moment of Inertia for 3/8" square rod = .00165
Moment of Inertia for 3/8" rod rod = .00097
Eliminating all the other constants - length, material, etc. the driving calculation simplifies to a constant value (x) * W/I.
square rod = 1x/.00165 = 606x
round rod = .7854x/.00097 = 809x
The round rod will deflect 1.33 times farther than the square. If the cross sectional area were the same in both cases, the round rod would deflect 1.70 times more.
The easiest way to think of why this is, is to consider the internal forces on the rod. Think of the material at the center of the rod as it bends. The material at the top of the bend is in compression. The material at the bottom is in tension.
Resistance to bending is based on a materials strength in compression and tension and the amount of material at the extremities of the geometry. The more material at the limits of the geometry the better. A round rod has very little material at the limits of its geometry, so it will bend more.
Orientation is obviously critical when calculating stiffness. That is why I-beams are made the way they are. Removing the material in the center makes the beam much much lighter without sacrificing too much stiffness.
If we consider a rectangular rod of the same cross sectional area of the 3/8" square rod. One side is .25 inches, the other is .526 inches. With the .526" side up, the rod will deflect 4.46 times farther than with the .25" side up.
This is explanation is obviously not perfect, but it is close.
Re: A ? for you physics gurus
Pat Riley
>Well, Hal got 1.33 as a ratio also, so I'm feeling happy. I don't do this kind of math very often - I'm an electrical engineer.
Re: more than that
Brian in Brighton
>The flanges also come in 2 different widths because of tension strength when the beam is under load. That's why you have S and W flange beams. There is actually no such creature as an "I" beam. Typically a steel distributor will base your needs on the size. Most houses are built with W flanges because it's easy to get in smaller sizes. Most semi-trailers have a S flange configuration because of strength and weight issues.
In simplest terms on a small round or square rod. The larger cross section wins.
Brian in Brighton
Re: I usta was one
Don J
>For the square rod, you don't necessarily get two rectangles. It's orientation to gravity just as easily gives two triangles
Re: A ? for you physics gurus
GalenV in Iowa
>Hal:
You should have metioned my deflections were wrong!!
Right data just did not hit right buttons on calculcator.
Good thing I am retired!!!
Think my brain has gone to rest. Last job before my heart attack was for Omaha water treatment plant clarifiers. Guess the units are still standing. Probably make them a little nervous to read this now. Small structural steel beam supports out of 36" deep wide flange beams 80 some feet long.
Get about same numbers as you but I used 29,000,000 per AISC for A36 steel.
Maybe it is a good thing I used computer programs as I had checks of my inputs. Started out in 1968 with calcs by hand and thought it was grand when we got a calculator about the size of a bread box that could do square roots! Ended in 2006 with computer that would run cals and draw the assembly in 3D.
Galen