AS3600-2018 CHANGES: DESIGN OF BEAMS FOR SHEAR, IS THERE A MISTAKE IN THE CODE?

More and more newer projects are adopting the latest revision to the Australian Concrete Design code AS3600. Our previous article covered a small tweak to AS3600 regarding how to treat set downs in a 200mm thick flat plate slab (located HERE). This article will focus on a more significant update to AS3600 which has been a long time coming… This is the substantial revamp of the methodology for designing beams for shear. Lets take a closer look…

This article has a nifty little tool which determines the bending and shear capacity of a concrete beam using the following approaches for a side-by-side comparison…

  • AS3600-2018 – (General Approach)
  • AS3600-2018 – (Simplified Approach)
  • AS3600-2009

To jump straight to that relevant section which also covers a potential short-coming with the current code update, click on THIS jump link. For a more thorough understanding of the similarities and differences between the two versions of AS3600, read from start to finish…

Notable Changes to Design of Beams in Shear to AS3600:2018

Straight off the bat, there appears to be a handful of significant updates to the philosophy behind the new approach to designing beams for shear in accordance with the latest version of AS3600…

Consideration for Torsion and its Effects on Shear

The effects of torsion play a much larger role in the shear strength analysis of beams in AS3600-2018.

Indeed the new shear section in clause 8.2 is now intertwined with torsion under the one chapter. AS3600-2009 contained a separate section for shear and torsion design of beams (sections 8.2 and 8.3 respectively), the separate torsion chapter now does not exist within AS3600-2018.

This potentially changes the design approach Structural Engineers now need to take for shear design of beams in a relatively drastic way.

Lets take a look at the new clause 8.2.1.2 which shows up very early in the beam shear chapter…

There is a mistake in this clause, which was subsequently rectified in Amendment 2 (released in 2021). At the time of producing this article, there is no consolidated standard available incorporating the current amendments (1 and 2 are currently released). So I have made this correction manually in the original clause.

It is worth noting that although previous code revisions made allowance for this requirement before, it has largely been overlooked by design engineers locally as it was mostly lost within the torsional section. Not so much a new requirement, just a re-located one…

I believe that historically, this minimum torsion requirement has been overlooked and not enforced adequately in the past by some Structural Engineers. Lets take a closer look at what its referring to…

This clause aims to address the issue of Structural Engineers seemingly ignoring torsion in beam design for generations. In Australia, a very popular design software called RAPT is used. It is a 2D frame analysis tool for the design of RC and PT slabs and beams. For the large part, engineers have been using this software to model three dimensional slabs in a two dimensional environment.

Even the humble band-beam system is a quasi two-way slab with stiffened column strips in one direction. The approach of designing a three dimensional system in a two dimensional environment means that incidental torsion experienced by beams are generally not picked up in the analysis. This is generally OK within reason (provided incidental torsion within beams is not excessive).

The widely used approach in the RAPT analysis software does at the end of the day assume that all incidental torsion within the beams are “strapped out” by the perpendicular adjacent slabs so there is a load path being provided. Therefore no cause for alarm, the slabs designed over the last few decades will not collapse following the release of AS3600-2018.

This clause is however closing a small gap in the industry approach to beam design for shear and torsion. However what does this mean in real terms? Lets unpack this specific clause a little further…

The clause does mention that the current approach is still “OK” i.e. Structural Engineers can continue happily ignoring the effects of torsional stiffness and strength of beams in a two-way system. However this comes at a cost…

The clause goes on to say that this approach can be taken provide that the minimum torsional reinforcement is provide within the beam (the erroneous reference to clause 8.2.5.5 now correctly updated 8.2.5.4). And also that torsional reinforcement needs to be detailed accordingly (that is the reference to clause 8.3.3).

This is a game changer for shear ligatures in the beloved band beam, as this clause now tells us:

  • If your analysis does not consider the torsional stiffness or strength of the beam, you must provide at least minimum torsional reinforcement within that beam…
  • Torsional reinforcement needs to be detailed such that it forms a contiguous outer loop around the perimeter of the beam through either an enclosed ligature or a series of bars with adequate anchorage through beams, cogs or bends.

This would effectively mean that a conventional band beam reinforcement arrangement like this…

Typical band beam reinforcement arrangement to AS3600-2009, note the open ligatures
Typical band beam reinforcement arrangement to AS3600-2009, note the open ligatures

Now becomes this…

Typical band beam reinforcement arrangement to AS3600-2018, note the closed ligatures
Typical band beam reinforcement arrangement to AS3600-2018, note the closed ligatures

It can be argued that an open “U” ligature coupled with a top negative moment reinforcing bar over the beam will constitute the detailing for closed ligature thus satisfying “minimum torsional reinforcement” in the eyes of AS3600.

However, as a standard approach which has been adopted historically, the majority length of a typical band beam has the “U” ligature spaced at centres of 500mm and sometimes 1000mm apart (outside of critical shear zones). This would fall outside of the minimum torsional requirements which at most would need the bars spaced at 300mm apart.

However you interpret the detailing side of things for beams in torsion, it appears that a slight up-tick in reinforcement tonnage may be required compared to previous projects designed pre AS3600-2018.

It will be really interesting to see how this is interpreted and how it plays out in the industry as more and more newer projects head towards construction under the new concrete code.

It may lead more Structural Engineers to undertake a full three dimensional FEA analysis of slabs to allow the torsional reinforcement to be justified asway in most instances. Are the days of the simple 2D RAPT approach doomed for a banded slab system?

Will this make software packages such as RAM Concept and SAFE more popular as these incidental torsional effects can be readily checked? (incidentally, for a quick article on vibration analysis of slabs using RAM Concept, take a look at THIS article).

Updates to the Phi Factor For Shear

AS3600-2018 has also tweaked the Phi factor for shear. It has relaxed the safety factor from 0.70 to 0.75 for beams that are provided with shear reinforcement larger than the minimum shear reinforcement requirements. It still requires a factor of 0.7 for shear with no assistance from steel and for torsion…

Phi Factor updates in AS3600-2018 Table 2.2.2

This probably won’t have a significant impact to the approach engineers take on their design. Nor will it have a significant impact on cost.

The relaxation of the safety factor is more than outweighed by more conservatism in the actual beam shear equations themselves (depending on the approach taken, which we will get into later on in this article).

Shear Limited by Web Crushing

The ceiling to the limit of the amount of shear a beam can take has been lifted slightly. Although its through the introduction of a more complex equation to determine the web crushing limit for shear in beams. Here is a direct comparison between the two (AS3600-2018 on the left, AS3600-2009 on the right)…

Comparison of maximum shear for beams controlled by web crushing. Clause 8.2.3.3 AS3600-2018 shown on the left, Clause 8.2.6 AS3600-2009 shown on the right.

This will mostly affect the beams with significant shear. Think lintel beams (or header beams, or coupling beams) within a core or heavily loaded transfer beams. For the most part, this wont significantly alter the design outcomes for conventional beams within an office, car park or apartment building. The relaxation appears to be in the order of 6-10% based on some trial runs I have performed.

There is also an additional clause added in AS3600-2018 covering a web crushing limit for combined shear and torsion. This was non-existent in 2009. It is very rare for beams to be heavily loaded in torsion under regular conditions on a self-supporting floor system. It will be interesting to see how this one plays out in the field.

The next notable comparison is the most significant, which deserves its own section in this article. It is the update to shear design capacity from concrete contribution in beams…

Concrete Contribution to Shear Strength, Comparing AS3600-2018 to AS3600-2009

Lets take a look at perhaps the biggest revamp to chapter 8.2 of AS3600. This is a direct comparison for the two equations to determine concrete contribution to shear capacity for AS3600-2018 (left) and AS3600-2009 (right)…

Side-by-side comparison of the changes between AS3600-2018 (left) and AS3600-2009 (right) for for concrete contribution to shear strength of beams
Side-by-side comparison of the changes between AS3600-2018 (left) and AS3600-2009 (right) for for concrete contribution to shear strength of beams

At a glance the updated equation looks simpler and more elegant. However looks may be deceiving. You will notice that the beta factors have been taken away as well as the tension steel allowance and the fcv factor.

It appears to be mostly replaced by this ominous kv factor. For instances where the shear reinforcement provided is more than the minimum required shear reinforcement kv looks like this…

AS3600-2018 equation to solve the nwe kv factor
AS3600-2018 equation to solve the nwe kv factor

This equation is simple enough. However another factor (epsilon x) is indicated. This is the longitudinal strain of the concrete an is a bit of a doozy…

AS3600-2018 new equation to solve "Epsilon x" longitudinal strain of concrete due to shear.
AS3600-2018 new equation to solve “Epsilon x” longitudinal strain of concrete due to shear.

The code does give allowance for a “simplified approach” to determining kv and therefore does way with the lengthy “epsilon x” formula. However this applies to non post-tensioned sections. Considering that the majority of beams in Australia are post-tensioned this may not have many applications. However if your confident that your post-tensioning effects are providing an increase to the beams shear capacity (which 90% of the time it will) the simplified approach should still be safe to use…

Here is the full simplified approach clause for determining the new kv factor here…

AS3600-2018 Simplified approach to solving kv factor to determine shear strength of beam
AS3600-2018 Simplified approach to solving kv factor to determine shear strength of beams.

These equations are a lot easier to digest and provide a great alternative for use in the real world when quick answers are required. There is just one problem however…

Most codes give alternative approaches to analysis such as this. More often than not there is a “simplified” or “easier” method, then there is a more “in-depth” or “harder” method. Generally when these two options are given in a code you sacrifice speed for conservatism. That is the simplified approaches generally give more conservative answers whereas the in-depth approach may take longer to complete but give more refined and usually less conservative results.

This simplified approach appears to give much higher beam shear strength capacity than the “non-simplified” approach. Lets take a look at a comparison for the results of shear strength of beams comparing…

  • AS3600-2009
  • AS3600-2018 (General Approach)
  • AS3600-2018 (Simplified Approach)

Lets tackle this now with a specific example…

Worked Example Beam Design Comparison between AS3600-2018 and AS3600-2009

Lets take a look at a representative beam with proportions and detailing which would be regarded as an “average” Australian beam.

I have chosen the humble band beam for this application, here are the details (to keep things simple, we will assume no post-tensioning effects)…

ElementValue
Design Shear Force V*1000 kN
Design Bending Moment M*500 kNm
Design Axial Force N*0 kN
Concrete Grade40 MPa
Beam Dimensions600d x 2400w
Shear Ligatures6x Legs N12-250
Reinforcement Grade500 MPa
Concrete Cover 30mm
Longitudinal Tension Reinforcement6N24
Youngs Modulus of Reinforcing200,000MPa
Max. Aggregate Diameter20mm

Churning through the code equations gives the following results…

Design ApproachResultant Shear Capacity
AS3600-2018 (General Approach)1146 kN
AS3600-2018 (Simplified Approach)1528 kN
AS3600-20091348 kN

The big take away from this little study is that the general approach within AS3600-2018 gives almost 25% lower capacity than the simplified approach. The simplified approach is then around 12% higher than the previous AS3600-2009 approach.

This leaves the obvious question , which one to use, the general or simplified approach within AS3600-2018? It will be interesting to see if subsequent amendments to AS3600 address this as I do feel that something doesn’t look quite right with the current arrangement.

To have a play around yourself, here is a section design calculator tool which gives results for bending capacity as well as the three approaches to shear design capacity of beams outlined above (user inputs are cells highlighted in yellow)…

To see the simplified approach applied to an isolated pad footing design example, take a look at THIS article. There i go through the philosophies of pad footing design, the failure mechanisms and a step-by-step worked example. I even provide a nifty design spreadsheet you can use which designs pad footings in accordance with the latest version of AS3600.

Similarities for Design of Beams in Shear between AS3600-2018 and AS3600-2009

This brings us conveniently to the similarity (note singular) between AS3600-2018 and AS3600-2009 for design of beams in shear…

The determination of the contribution for shear reinforcement to the beams shear capacity remains largely un-changed. This is good news considering the significant update to the concrete contribution. here is another side-by-side comparison between the two equations…

Near identical formulas for calculating the contribution of shear reinforcement to the shear strength of beams.  AS3600-2018 (left) and AS3600-2009 (right)
Near identical formulas for calculating the contribution of shear reinforcement to the shear strength of beams. AS3600-2018 (left) and AS3600-2009 (right)

Published by

Quentin Suckling is a full time practicing Structural Engineer based in Melbourne Australia. He has been practicing in the local market at tier 1 engineering consulting firms over the last 16 years.

4 thoughts on “AS3600-2018 CHANGES: DESIGN OF BEAMS FOR SHEAR, IS THERE A MISTAKE IN THE CODE?

  1. The requirement for minimum torsion reinforcement when ignoring torsion stiffness isn’t new. It was in the 1994 code if not earlier. I believe the open U stirrup arrangement was being justified on the basis that the slab top reo closed the stirrup.

    What’s more concerning is how the torsion design rules were changed to equivalent shear, then back shortly after through an amendment. It casts doubt on how rigorous the code committee is. I know they were virtually dysfunctional around the time of the 2009 code revision.

    1. Hi Jeremy, thanks for your comment.

      I agree with you regarding the minimum torsion requirement not being a new thing in AS3600 (rather a standing requirement which has changed location due to the shear and torsion sections now becoming one). I do believe however that now having the shear and torsion sections amalgamated as one, it will bring it to the attention of more engineers. The reason why I believe this is I have seen a lot of typical band beams historically detailed with the “U” open ligature spaced at 500 to 1000mm spacings and largely regarded as “tie steel” (outside of critical shear zones).

      Spacing of 300 or more would leave it outside of the code requirement for detailing of torsional reinforcement and not satisfying what the code is trying to say. I also agree that coupling the “U” open ligature with a top bar over the beam would satisfy the torsion detailing in the eyes of AS3600 (provided the top bar and ligature are at least no more than 300mm apart). I have tweaked the article to reflect the comments above, I appreciate your contribution!

      The changing of design approach from revision to revision of the code is quite confusing and does reduce confidence in the overall document, I agree. In addition, some of the errors which came out in revision one of AS3600-2018 were also quite a concern. Some have been amended with subsequent revisions, some still remain. This all contributes to Structural Engineers scratching their heads and loosing faith in the Australian Concrete design code (unfortunately).

Leave a Comment

%d bloggers like this: