20170929 Eindhoven Car Park Collapse

Eindhoven car park collapse with BubbleDeck floor Because the outcome of the collapse at Eindhoven Airport has been in the news this week, it is good to inform you with facts on the conclusions. For that a news communication is cited in this memo, as well as some information from slides from the investigators. Technical cause of airport car park collapse reve revealed aled (source: www.newcivilengineer.com )

Dutch-based BAM has revealed the results of investigations into the technical cause of the partial collapse of a car park in the Netherlands. The P1 car park at Eindhoven Airport Airport was under construction constr uction when it collapsed on May 27 27 2017. 2017. Following the collapse t here here have hav e been two investigations. The T he first by the t he Net Net herlands organisation for applied a pplied scientific research TNO, on behalf of Eindhoven Airport. BAM assigned engineering consultant Hageman to investigate the technical cause of the collapse. The conclusions from the investigations concur. The car park used BubbleDeck floors, which are precast concrete floor slabs embedded with weight-saving plastic balls. After installation on site, a concrete topping was applied. A statement from BAM said: “The investigations show that the shear strength of the interface between the precast concrete floor slabs was insufficient to convey the traction. As a result, the structural capacity of the floor was compromised. The high ambient temperature on the day of the collapse was a contributing circumstance to the defect manifesting itself.” BAM also said structural engineers and building inspectors in general do not factor in enough the importance of shear strength for the adhesion of the floor.

It said: “The assumptions for the design and calculation of concrete structures are described in the construction standard NEN-EN 1992-1-1. In addition to the investigation it has been found that both structural engineers and building inspectors in the application of precast concrete floor slabs generally do not adequately recognise the importance of shear strength for the adhesion of the floor. An additional guidance of the NEN-EN 1992-1-1 would certainly promote the application of the standard.” BAM Bouw en Techniek director Jaap Hazeleger said: “Fortunately an event like this is highly exceptional. We are very grateful that no personal accidents have occurred. With the investigation results in hand, we will engage in further talks with Eindhoven Airport to determine the most appropriate follow-up steps.” BAM said it is sharing the results with the industry and now there will be discussion to determine the future of the building and who can be held responsible for the incident.

Engineering consultant Hageman investigated the collapse for BAM and the following technical details were published. The floors consisted of BubbleDeck floor elements – basically precast concrete filigran slabs of a few cm thick and approximately of 6 m length and 3 m width with balls on it (to reduce weight) and in-situ cast concrete. See next sketch. The red lines (“naad”) are the joints of the bubbledeck filligran plates. The floor is designed as floor spanning is two direction: 6 m in x-direction and 15 m in y-direction and point supported by the columns (no beams were used).

2

The collapse started at joint 3 at the top floor (see last page for some animations). The next sketch and photo show how the joint was made between the BubbleDeck elements in width direction. In order to couple the slabs, coupling reinforcement (“koppelwapening”) is used. These were 4 bars ∅16 per ball. The bars were placed directly on the precast concrete filligran elements.

Vertical failure crack through balls at the end of the coupling reinforcement

Horizontal failure crack through interface between filigran and insitu cast concrete

Vertical failure crack through joint between filigrans

3

The moment capacity of the floor at the joint is determined by: • Tensile strength of the coupling reinforcement Anchorage of the bundled coupling reinforcement that was directly placed on the filigran slab • Interface shear resistance between the filigran slab and the in-situ cast concrete. • Important is to note that self-compacting concrete was used to produce the precast filigran elements. Due to this sue of concrete the top side was too smooth - there was no friction present in the interface. The investigation conducted some experiments in the laboratory on the joint, and concluded that: the failure was caused by failure of the interface between the filigran slabs and the in-situ cast • concrete • failure was ductile the bending capacity present of 125 kNm/m was only 1/3 of the bending moment acting in joint 3 •

The floor was thus only able to carry its own weight. On 27 May 2017 it was a very hot day. The top side of the floor was heated up and as a result the floor curved upwards. But through the connection with the columns this bending upwards was hindered, such that it resulted in additional moments in the floor. Due to the low bending capacity the floor failed, and by progressive collapse the whole part of the bcar park came down. It was not due to the hot wheather, but if the hot day had not been there, failure would have happened during the use of the building when some live load was added. To conclude, key in the collapse was: self-compacting concrete for filigran, not roughened topside of filigran, BubbleDeck floor field point-supported by columns. The Dutch precast concrete industry in reaction mentioned that some precasters did not offer for this building in the first place, because their technical people were of the opinion that the car park floors were too slimly designed by the engineering consultant. The next page shows 4 screen prints of a video published at https://www.ed.nl/eindhoven/aansluiting-vanbetonplaten-en-temperatuur-oorzaak-instorting-parkeergarage-eindhoven-airport-mogelijk-meer-gebouwenkritiek~ae3eabcf/ 

4

5