Square or round tube?

[Billzilla]

We have to replace the trailing arms on the rear suspension on the racing car, as one of them bent very badly under compression last weekend,
My question is that for a set size (say, 30mm) what would be stronger in terms of resisting being bent, a 30mm diameter round tube or a 30mm to a side square tube?
And how much more would the square tube weigh?
Same wall thickness for both tubes.

[ed]

the square tube will be stonger but 27% heavier

note: term 'stronger' is axis specific

[ed]

bill, you should clarify actually - was the original failure 'bent' or 'compressed'

[roadsailing]

im too tired to calculate it, but I would say Ed is on the money.

some reading:

http://en.wikipedia.org/wiki/Second_..._cross_section

and here is some in german.

http://de.wikipedia.org/wiki/Fl%C3%A...A4gheitsmoment

technically, if you use a rod end at each end it should be in pure compression, but if there are any irregularities in the tube at all it will kink or buckle.

I would tend to use round as it will be symmetrically loaded, and for a given mass it should be stronger.

[thechuckster]

the pics suggested it was bending

[clubagreenie]

Wouldn'ta rectangular section be preferrable? With the longer side vertical, Would be in keeping with OEM designs I've seen, though mostly front radius arms have been sq/rect and rear trailing arms round, and upper rear arms sq/rect.

The only alloy rear arms I've seen were made from 2 x rect sections welded to make a sq section.

[ed]

without getting too techy or mathematic about it - the shape that has the greatest quantity of material, where you need it, wins

in pure bending, you want material perpendicular to the load, with the location of the greatest quantity of material (as in an I beam) in tension/compression as far away from the rod/tube's centreline as possible - so the box section wins. if you wanted to be even smarter, youd pick the primary load axis and get a rectangular section

in longitudinal tension/compression, just the pure increase in material quanity that you achieve with a box over round - wins. as mentioned just be careful of dents/imperfections

torsional loads - intuitively (for me anyway) a tough one in this comparison

[roadsailing]

pics? pics would help, but my money is still on using round tube, unless it needs to fit through a square gap 31mm wide

[ed]

round is a very material and weight efficient shape. it is also good for non axis specific loads

if youve got a specific problem though, you have the opportunity to pick the right shape

how many motorbikes run round swingarm main stays?
how many roofing main members are round?
how many square profile arrows do you see?

we still dont know bills specific installation, or the details of the failure, so were all pissing in the wind a bit here

[roadsailing]

i agree ed, but it is more fun than just sitting around.

as for torsion, circular is intuitively the best for this reason:

no other shape will get more material further away from the neutral axis, to bastardise your comparison, how many tailshafts have ypu seen that have not been round?

[ed]

ah, yes, but first imagine the shaft didnt 'spin' but was just twisted...

now make it bigger, 27% heavier, and square - would the round still be the strongest if tested to failure. thats this specific comparision

if we were stuck weight for weight - id pick the circle in an instant - but were not

[clubagreenie]

Trailing arms being fixed at the chassis fwd of the diff/swingarm/axel tube?

Bending in compression in that location, what did it hit? In 5 years of slamming a rangie off road never bent a trailing arm and they're round, about an inch and about 6mm wall steel. Not overly heavly though comsidering the length. backed up by a massive a arm setup on top of the diff but no other lateral location.

[ed]

'bending in compression' ???

[ed]

bending:

.....|............................................ ...........|.....
....V............................................. .........V.....
=================================
...............................^.................. ...............

compression:

--> ============================ <--

[clubagreenie]

Bills words in first line

I've sen compressive deformation but only head on into a wall.