Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

RE: Purlin- which one is better?

[Subject Prev][Subject Next][Thread Prev][Thread Next]

Here in Australia typically only z-sections can be lapped, they have unequal flanges to facilitate nesting. C-sections are typically installed double or triple span, or as many continuous spans as possible in the maximum length up to 12m (18m special order). No continuity is provided at the end of one member and the next. Most cold-formed shed fabricators oppose using the zed’s because they believe the laps are a waste of material and there are around 4 times the number of bolts required at each support to form the lap; extra materials and additional labour. Some do however use a fabricated sleeve when using the c-section’s, though it is seldom engineered and of any value. The problem is lack of detailing, if continuous span c-sections are used then either the building as to have an even number of bays, to install double span purlins throughout, or a triple span purlin needs to be installed some where. If you are unlucky, then a single span purlin ends up in the end bay, next to a wind ward edge it’s the most highly loaded, and therefore likely to be shown below capacity.


Lapped z-sections offer greater consistency, in that they can all be the same length, therefore no problems about locating the double span and triple span lengths of c-section. On larger buildings thicker gauge z-sections are used in the end bay, this is more economical than using the thicker gauge throughout. If C’s were used then a minimum of double span would be required with the thicker gauge in the end bays, the alternative is to half the spacing in the high wind regions of the walls and roof (eaves, ridge, end bays).


What I’ve not seen is detailed design of wall girts to allow for cutting in of PA doors and windows, creating spans of varying lengths and continuity, invalidating the double span design. Given that additional doors and windows maybe added in the future to a simple shed like structure, single span design seems the better option, though not economical.


Another problem is connection at supports. The Australian code, and manufacturers technical data are all based on purlins being attached to welded cleats, attachment is by two bolts through the web. The cold-formed shed manufacturers however, attach the c-sections girts/purlins flange to flange with the c-section portal frame. Both c-sections are punched by the c-section manufacturers when ordered, thus the shed fabricator has little work to do. If I understand correctly the girts/purlins attached in this way have higher torsional and lateral flexural instabilities than if they were fastened by the web to a cleat. Because the industry is established using small 75mm deep C’s, and none have yet experienced there design load, and they have generally been approved by the regulating authorities, it is not practical to declare the sections to be inadequate. However, the industry is moving into larger sheds, requiring deeper c-section girts/purlins and I am reluctant to specify the flange to flange connection. (Not that anyone ever specified such in the first place, that just happens to be the way they build them. And once you know, it’s reassuring to have calculations that match what gets built.)


My simplistic approach so far has been to specify:

1)       steel strapping to prevent rotation of the section at the supports OR

2)       Additional bridging members at the supports.


Alternatively I have considered ignoring the manufacturers design capacity tables, and designing the members direct to the cold-formed structures code as single span segments between supports, and determining the restraints required within the span. This however seems over conservative. (Designing the c-section is not a problem, I do that for the portal frame, but calculated capacities(to the code) are typically less than the tested capacities of manufacturers, or the finite strip buckling analysis they use to produce their tables)


Whilst the Australian code is based on the American cold-formed specification, it doesn’t cover flange to flange connections, but I believe flange to flange is the common practice in the USA. Is this true? Can someone advise whether the R-factors given in the American code for flange to flange are dependent on the flange thickness of the main frame? (eg, are they for heavy gauge plate of universal beams, I-beams, W- beams or light gauge to light gauge.)


I’m reluctant to get the American Code if it doesn’t fill in the gaps.




Steven CONRAD Harrison

B.Tech(Mfg&Mech), MIIE, gradTIEAust


METAMORPHS: Beyond Structures



-----Original Message-----
From: Sunil Thengale [mailto:Sunil_Thengale(--nospam--at)]
Sent: Thursday, 25 May 2006 15:40 PM
To: seaint-ad(--nospam--at); seaint(--nospam--at)
Subject: Purlin- which one is better?


Dear all,

Normally the design of cold formed purlin is carried out as continuous purlin. At the supports, the purlin is either overlapped or a separate sleeve purlin is used to cater for support moments, the length of which is dependent on point of contraflexure in both cases. Most of my clients specify that the purlin is to be sleeved rather than overlapped. My simple question is does it have any advantages? Is there any guideline that for a particular thickness or depth? Any design implications ?  Please share your views.


Sunil Thengale