Proceedings of the International rilem conference Materials, Systems and Structures in Civil Engineering 2016

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

Only three participants have submitted some results about cracking in the gusset: Team 50, 

Team 25 and Team 40. Each team uses a different methodology. Although, comparing results 

is difficult. Only Team 50 and Team 40 have given some quantitative results about number of 

cracks, spacing between cracks or openings. 

All teams give results after 4 days of concrete pouring. At 4 days after the pouring, the 

spacing between two cracks is about 0.8-2m (cf. team 40 and 50), while in situ observations 

show a spacing of 1.2 m. Therefore, the width of the cracks at 4 days is about 0.07-0.1mm, 

which is similar to what is observed in situ. Given to team 50, the maximum opening of the 

cracks increases significantly from 0.1 at 4 days to 3.3 mm at 10 months. 

 

2.2 Theme 2: prediction of the behavior of the containment wall 

The predictions of the behavior of the containment wall were expected at different steps: 

before prestressing, after prestressing and at 5.2 bar abs. during the pressurization test. The 

strains and stresses in 10 points defined in the gusset, the cylindrical wall and the dome, and 

cracking state evaluation (inner face cracks, outer face cracks and through cracks) were 

expected. 

The experimental data are below 

[Figure 8]

 (the results are limited to the cylindrical part in 

this paper). 

 

Figure 8: Cylindrical part strains from pouring to first pressurization test 

 

---

87

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

The experimental results showed in this part [Table 3] correspond to the visual inspection 

made after the first pressurization test. During this inspection, only cracks which had an 

opening superior to 0.1mm were measured and mapped. The difference of cracking state 

between these results and those showed in Theme 1 part can be explained by the fact that 

during early age inspection, all visible cracks (without any criterion on the opening)  were 

measured and by the fact that prestressing can have closed some of the early age cracks. 

 

Table 3: Visual inspection results on cracking; Total length and max opening measured. 

 

Dome: Outer face  Dome: Inner face  Gusset: Outer face  Gusset: Inner face 

Number 

38 197 28  0 

Total length (m) 

13.73 138.07 24.63 

/ 

Max length (m) 

1.16 1.66 3.14 

/ 

Average length (m) 

0.36 0.70 0.88 

/ 

Max opening (mm) 

0.10 0.10 0.10 

/ 

NB: The cracks identified in the dome inner face are located on precast slab forms and don’t 

reflect the dome mechanical behavior. 

 

Nine teams answered Theme 2. Their results are given and compared to experimental results 

in this part. Teams 70, 50, 37 and 24 began their calculation since the raft concreting. Teams 

49, 21, 15 and 14 took the end of containment erection date as starting date. Team 23 took the 

raft concreting date as starting date, but didn’t provide values for strains and stresses at the 

end of concreting date. 

The experimental results represented in this part are strains corrected from temperature 

effects. The teams are divided into two groups: 

 

Group 1, which includes teams which began their calculation since the raft concreting 

(Team 70, 50, 37, 24, 23) 

 

Group 2, which includes teams which began their calculation since the end erection 

(Team 49, 21, 15, 14) 

These 2 groups are separated by a vertical red line on the following graphs. 

2.2.1 Strain evolution results 

The results are presented for the cylindrical wall strain evolution, in tangential directions. The 

results are quite the same in vertical direction. 

Teams’ results are scattered 

[Figure 9]

. 

In both directions, all teams, except Team 15 for tangential strains, show a more or less high 

shortening for all captors as experimental results. 

Team 15 results for tangential strains show elongation for captors H5 ET and H6 IT and a 

very low shortening for H1 ET. 

---

88

International RILEM Conference on Materials, Systems and Structures in Civil Engineering 

Conference segment on Service Life of Cement-Based Materials and Structures 

22-24 August 2016, Technical University of Denmark, Lyngby, Denmark 

 

Except for Team 14, the right evolution is predicted between the end of the erection and the 

end of prestressing, and between the end of prestressing and the pressure test. But the 

amplitudes are not always well reproduced. 

 

Figure 9: H5 Extrados Vertical strains evolution 

2.2.2 Prestressing effects in the cylindrical wall 

On the following graph 

[Figure 10]

, experimental results are represented by a red horizontal 

line. The experimental results represented in this part are strains corrected from temperature 

effects. The strains represented in this part correspond to strains between the end of 

construction and the end of prestressing. As the prestressing begins short after the end of 

construction, it can be assumed that they correspond to the prestressing effects, but also 

include some shrinkage and creep effects. 

Near the inner surface (figure not presented here), the vertical (not presented here) and the 

tangential measured strains are greater than near the outer surface. This is also obtained in the 

simulation for Teams 24, 21 and 14. For Teams 50, 37, 23 and 15, the vertical strains are the 

same near the outer surface as near the inner surface. In general, in both directions, 

prestressing effects are underestimated by the teams, except Team 49 which overestimates 

them. Team 15 results for tangential prestressing effects show elongation for captors H5 ET 

and H6 IT and a very low shortening for H1 ET. However in many cases, Teams 70 and 14 

are closer than the others. 

 

---