-
LOCATION:HOME >> NEWS >> Industry news
-
Industry news
An Open Purlin Twists, a Closed Tube Does Not
An open purlin twists; a closed tube does not
A C-purlin and a square tube can be formed from the same strip, weigh the same per metre, and carry the same vertical load as a beam. Move that load a few centimetres off centre and they stop being comparable. One twists, the other barely notices. The difference is not strength and it is not grade. It is whether the cross-section is closed, and where its shear centre sits relative to the screw holding the roof on. Most purlin trouble on site arrives as a fixing problem and gets argued about as a steel problem.
Where the load actually lands
The thin profiles in a frame come off coil that the Z275 listing puts at 0.12 to 2.0 mm, and the SGCC sheet names light steel keel among its cold-formed products. Roof load arrives through the sheet, through a screw in the top flange, and then into a section with only one axis of symmetry. For a C or Z the shear centre sits outside the section, behind the web, so a vertical force applied at the flange is eccentric to it. Take a 200 by 70 by 20 mm C in 2 mm strip with the screw line 35 mm off the shear centre. A modest 1.0 kN per metre of roof load then produces a distributed torque of 35 N about the section.
Two tables, one shear modulus
Twist is calculated with G, not E, and the catalogue supplies both. The 310S data sheet prints 200 GPa in tension and 78 GPa in torsion. The 201 sheet gives an elastic modulus of 197 GPa and a Poisson ratio of 0.27 to 0.30; putting those through E over twice one plus nu returns about 77 GPa. Two unrelated pages agree within a GPa or two, which is more accuracy than a six metre purlin calculation needs. Carbon structural steel sits in the same band. That 78 GPa is the number the twist sum uses.
The open-section constant has a cube in it
For a thin open section the St Venant torsion constant is the sum of each strip width times its thickness cubed, divided by three. The C above has roughly 380 mm of centreline, so at 2 mm the constant is 380 times eight over three, about 1,013 mm4. The thickness is cubed, which is the part worth remembering: a profile formed from 1.6 mm strip instead of 2 mm keeps 80 per cent of the gauge but only 51 per cent of the torsional constant. Rolling a lighter gauge costs far more in twist than it does in bending, and nothing on the order sheet says so.
The closed-section constant is a different animal
For a closed thin wall the Bredt expression is four times the square of the enclosed median area over the loop of wall length divided by thickness. Take an 80 by 80 mm square tube in 2.5 mm wall, a size well inside the 10 by 10 to 300 by 300 mm and 0.5 to 75 mm ranges on that page. The median area is 77.5 squared, or 6,006 mm2, and the loop is four times 77.5 over 2.5, or 124. The constant comes out near 1.16 million mm4. Its area is 775 mm2 against 760 mm2 for the C, so the two sections weigh within two per cent of each other and differ in torsional stiffness by a factor of about 1,150.
What that does to a six metre bay
Put the same distributed torque over a 6,000 mm span and use the St Venant rotation, m L squared over eight G J. For the C, 35 times 36 million over eight times 78,000 times 1,013 gives about 2 radians, or 114 degrees. For the tube the same numerator over 1.16 million gives 0.0017 radians, a tenth of a degree. No engineer would accept either figure as a real rotation: the open section also carries warping torsion, and its answer depends entirely on what it is tied to. That is the point. An open purlin does not resist twist by itself. It resists twist because the roof, the sag rods and the bracing hold it, and when one of those is missing the arithmetic becomes absurd.
What the frame uses to hold it
The steel structure material list is explicit about the parts: Q235B C and Z purlins, a knee brace, a tie tube of circular steel pipe, and vertical and horizontal support in angle steel, round bar or steel pipe. A tie tube works in tension, so it restrains one direction and has to be paired for the other. The profiled sheet above the purlin can act as a diaphragm only while its fasteners do not slip; a plain galvanized sheet with a lap detail that allows movement gives lateral restraint but not torsional restraint. The shear centre is outside the section on every bay, which is why restraint has to be continuous rather than occasional.
Why the tube is the bracing member
In a closed section the shear centre and the centroid coincide, so a load from any direction passes through it and generates no torque about the longitudinal axis. That is why circular pipe and square tube turn up as tie tubes, braces and posts rather than as purlins. The stainless pipe page lists round, square and rectangular items with squash, extended and water pressure tests; those are pressure-pipe checks, and a structural order should ask for wall thickness, straightness and dimensional tolerance separately. The thick-wall range and the 310/310S tube listing are the same geometry in different grades, and the geometry is what does the work.
Closing the section is a fabrication decision
Back-to-back C sections, or a plate welded across the open face, turn the open profile into a box and lift the torsional constant by three orders of magnitude. It is a legitimate fix, but it happens after cutting and shearing, and the weld burns the zinc at the very edge that was supposed to last twenty years. Repair coating then becomes a site item. The structural pipe note makes the same distinction from the other side: where a pipe carries compression or bending rather than internal pressure, the tolerance question changes, and an ordinary welded tube is often the sensible buy. Neither decision is about grade.
What the order has to say
Name the section type, not just its depth: C, Z, square tube or round tube, each with the wall thickness and the flange or lip dimensions that set the centreline length. State the fixing line and the assumed restraint, because an eccentric 35 mm is a design input rather than an accident. List the sag rods, tie tubes and their directions, and the fastener grade that lets the sheeting act as a diaphragm. Ask for straightness and wall thickness on structural tube instead of inheriting a pressure-pipe test schedule. A purlin order that gives only a depth and a coating mass has left the shear centre off the drawing entirely.
Next:No More!