CROSS Safety Report
Punching shear in concrete slabs at perimeter columns
A reporter raises concerns about the underestimation of bending moments in perimeter columns when assessing punching shear in adjacent concrete slabs. Punching shear failures can be brittle and catastrophic. The report proposes that designers need to include the appropriate ultimate moment in the design actions. The reporter's experience is that some engineers appear to lack an understanding of the requirements in AS3600 with regard to stiffness of perimeter columns that could lead to punching shear failures.
Key Learning Outcomes
For structural and civil design engineers:
- Clauses and information in design codes and standards must be interpreted to comply with principles of structural design, and not to provide results preferred by designers
- Values of bending moment used for assessment of punching shear in slabs must be determined using worst-case scenarios of stiffnesses of connecting columns
- Failure to provide adequate reinforcement to resist punching shear may result in brittle failure and sudden catastrophic collapse of concrete slabs
Find out more about the Full Report
Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others. If you would like to know more, please visit the reporting to CROSS-AUS page.
As a forensic structural engineer with many years of design experience, the reporter has observed that shortfalls in punching shear capacity appear to be increasingly commonplace in Australian apartment buildings and especially at perimeter columns of buildings. This is often in areas where slab depths are reduced due to balcony set-downs, and bending moments are relatively high as columns are loaded from one side only. The reporter has discovered errors in design which have led to these problems. In many cases, this has been a failure to design for the correct value of bending moment at the slab and column interface.
Calculation of section stiffness
When performing linear-elastic analysis and considering how stiffness should be apportioned to columns and slabs AS3600:2018.2.4.1 gives clear direction. It states that:
'The stiffness assumptions chosen shall be consistent with loading conditions, and shall generate critical worst-case actions under all failure modes to be considered. Where multiple degrees of stiffness are possible, the stiffness assumption chosen for the failure mode under consideration shall induce worst-case actions in the element being designed'.
For example, if a designer considers that the columns may be cracked under ultimate loads, then they need to establish by calculation that this will be the case. In those calculations, the relevant flexural tensile strength is the upper Characteristic Tensile Strength defined in AS3600, which is 1.8 times the default Characteristic Tensile Strength.
The maximum design actions for punching shear mode at the slab/column connection would be those that occur using the properties of the gross section of the column, assuming the column does not crack. If the column is found to crack under ultimate load, then the load condition immediately before the column cracks needs to be examined as it may represent the critical load condition for punching shear.
Redistribution of bending moments
In the reporter's experience, it is sometimes the case that engineers will argue that they have redistributed the bending moments away from the external columns. However, the reporter notes that although a degree of redistribution of bending moments is permissible for the purposes of flexural reinforcement, punching shear is a brittle (non-ductile) failure mode and no moment redistribution is permissible as part of its design process.
punching shear is a brittle (non-ductile) failure mode and no moment redistribution is permissible as part of its design process
Also, whilst AS3600:2018 permits moment redistribution away from (or towards) internal columns for the purposes of flexural design, it is the Reporter’s view that it does not permit redistribution of moments at external columns (see clause 18.104.22.168 of AS3600).
The reporter was recently handed a critique of a forensic report they had written that argued that their assessment of punching shear was overly conservative. The author of this critique considered that the column stiffnesses that the reporter had assessed as uncracked could be reduced by 60% through the use of Table 6.2.4 of AS3600. There were no calculations to support this view. Table 6.2.4 in AS3600 is located in a section entitled “Stiffness of lateral force resisting elements” which is included in the code for the assessment of “inter storey drift, periods of vibration and distribution of internal actions". This section of the code is to be used for determining column stiffness in circumstances where reduced column stiffness acts to exacerbate problems such as P-Delta effects or vibration. It is not intended for use in assessing the distribution of moments under gravity loads.
Lateral load distribution
The reduced value of stiffness given in Table 6.2.4 of AS3600 is in any case only permitted to be used for assessment of lateral load distribution once the designer has shown by calculations that the sections will indeed crack under the relevant lateral loads (see clause 22.214.171.124 of AS3600). In such situations, cracking in columns makes the situation worse so that lower-bound cracking stresses apply. In the case in question, the lateral loads could be carried by substantial shear cores which would not crack under lateral loads, and the reporter concluded that the critique portrayed a disturbing lack of understanding of AS3600.
To provide a situation where lateral loads reduced column stiffness for the purposes of punching shear assessment, the reporter observes that a series of design events would be required whereby:
- the lateral load causing the columns to crack occurred prior to any gravity load causing punching shear; and
- the location of the cracks in the columns induced by the lateral load happened to be on the same side of the column as those that would be induced by gravity loads.
The reporter considers it should have been obvious to the engineer that such a proposition defies any logical reasoning. It is useful to remember that AS3600 provides a series of criteria that must be adhered to, and it pre-supposes a level of logical reasoning and design expertise. Serious design errors of the type described above can result if designers use any code or standard without having an understanding of building design based on first principles.
Submit a report
Your report will make a difference. It will help to create positive change and improve safety.
Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others.
No feedback has yet been published for this page.
Expert Panel Comments
An Expert Panel comment on the reports we receive. They use their experience to help you understand what can be learned from the reports. If you would like to know more, please visit the CROSS-AUS Expert Panel page.
The fact that this report is actually needed is a sad reflection on the current state of structural engineering both in Australasia and internationally as it is not confined to Australasia.
The shear provisions in AS3600 have changed significantly in recent years. However, whilst the rules for shear in beams were extensively modified in AS3600:2018, the rules for punching shear did not change except for the addition of a requirement for bottom face reinforcement at columns to attempt to make the punching condition more ductile. This rule is already being seen to be misinterpreted by some engineers, producing designs at increased risk of failure. Punching shear is listed for revision in the next review of AS3600 where it may be given further consideration with respect to behaviour under certain fire conditions.
This rule is already being seen to be misinterpreted by some engineers, producing designs at increased risk of failure
Attention is drawn to the recently published commentary to AS3600:2018, which provides good advice for designers of slabs and beams.
Unfortunately, in practice, modelling of structures does not necessarily cover all load cases that designers need to consider. Careful design needs to be performed in conjunction with appropriate checking and review, particularly with shallow band beams or flat plates.
All design codes need to be "interpreted" by designers to apply the rules for each design situation. This interpretation requires the designer to have a good understanding of structural design principles and an ability to comprehend the rules. Both appear to be missing in this particular case. Design standards must be interpreted to suit principles of structural design, and not simply to suit results more convenient for designers.
We are aware of situations where punching shear has been designed for zero moment because it was the only way the slab could be made to work, and where major columns have been designed for zero moment because the beam connected to the column was designed assuming no fixity (even at internal columns).
As noted by the reporter, AS3600:2018 requires that, when carrying out a linear-elastic analysis: 'the stiffness assumptions chosen shall be consistent with loading conditions, and shall generate critical worst-case actions under all failure modes to be considered. Where multiple degrees of stiffness are possible, the stiffness assumption chosen for the failure mode under consideration shall induce worst-case actions in the element being designed. In the case of an assessment of bending moment in a slab, this could well involve design for uncracked column conditions'.
Previous CROSS reports
There have been several previous reports that refer to punching shear in flat slabs, including:
1050 Concerns about punching shear in a flat slab
950 Inadequate punching shear reinforcement in flat slabs
906 Missing punching shear reinforcement in concrete slabs
886 Unconservative design of flat slab due to software modelling issues