Figure 22. The walletes prepared for grouting.

estimate its compressive strength using empirical formulae available in the literature. The calculated strength values were unacceptably scattered, thus making unsafe or over-conservative the selection of any of the calculated compressive strength values (Vintzileou 2002). Thus, the decision was made to construct wallettes and to determine basic mechanical properties by testing them before and after grouting (Vintzileou et al. 2006, Miltiadou et al. 2006). The geometry of the wallettes was chosen to simulate the upper and more vulnerable part of perimeter masonry. In order to avoid scale effects a scale 2:3 was selected. Six three-leaf stone masonry wallettes were constructed using materials of similar characteristics as the in-situ ones. Half of the wallettes were tested in compression and the other half in diagonal compression up to their maximum resistance. After unloading, the wallettes were grouted using one of the selected grouts of the Table 1 (Fig. 22). The injections were carried out using a specific methodology and the total grout consumption (calculated as the ratio of grout volume consumed for each wallette per total volume of wallette) was of the order of ~10%. Subsequently, they were tested

W2 and W5 injected with the ternary grout; W1, W3, W4, W6 injected with the hydraulic lime based grout.

again in compression or in diagonal compression to failure. The main results are summarized in Table 2.

Both grouts applied to the specimens were able to achieve homogenisation of masonry by filling cracks and voids of wallettes. Thus, the wallettes exhibited substantial improvement of their behaviour, in terms of compressive strength, tensile strength and reduction of the separation between the three leaves of masonry, without substantial increase in their stiffness, and proved to be efficient from the mechanical point of view. From the two alternative compositions, the natural hydraulic lime based grout was selected for the application to the Katholikon of Dafni Monastery, due to the substantial (compressive and tensile) strength enhancement of wallettes, the rather ductile behaviour under diagonal compression (compared to that of masonry grouted with the ternary grout), and the better durability properties (Miltiadou et al. 2007) that contribute to the protection of mosaics and frescoes.

9.3 Optimum grout composition

In order to further improve the hydraulic lime based grout, the addition of fine natural pozzolan (dmax < 45 |xm) in various proportions was investigated. The addition of a small percentage of pozzolan (10%) was decided, on the basis of additional data, deriving from porosity measurements, salt durability tests and from in situ pilot trials.

The mix proportions and the injectability characteristics of the optimum grout composition are presented in Table 3, whereas more detailed information is given in Miltiadou-Fezans et al. 2007, 2008. This optimum grout composition was used for the repair of the monument. Moreover it was used for the repair of a large scale model of a Byzantine groin vaulted structure, constructed and tested on a earthquake simulator, as briefly presented below.

Table 3. Optimum grout composition and injectability characteristics measured in the laboratory and in situ at the first pilot preparation.

Grout composition

NHL5 (StAstier) 90%

Pozzolan Petrotechniki 10%

Grout properties In lab

T36 (sec) - Sand column 19-22 1.25/2.50 mm (Wnom ~ .0.2 mm)

Bleeding <1%

0 min after mixing 21

60 min after mixing (agitated) 23

Apparent density (gr/cm3) In lab

0 min after mixing 1.5050

60 min after mixing (agitated) 1.4986

In situ

In situ 22 25

In situ 1.4978 1.4870

(2) superplasticizer based on polycarboxylic ether.

Figure 23. The groin vaulted structure on the seismic simulator.

Figure 24. The groin vaulted structure after grouting.

Figure 23. The groin vaulted structure on the seismic simulator.

9.4 Construction and testing of a large scale structure covered with a byzantine groin vault

As aforementioned in section 2, byzantine groin-vaults are used for covering the most of the parts of the Katholikon of Dafni Monastery, including the arms of the cross, where severe damages have been occurred. Thus, important effort has been undertaken for achieving a better knowledge of the dynamic behaviour of such structures. An experimental research was carried out by DTRR/HMC in collaboration with the Laboratory of Earthquake Engineering of NTUA. A model of a byzantine groin-vaulted structure bearing locally mural mosaics was constructed. The materials and the construction type used for its masonry walls were exactly the same with those used for the construction of the wallettes, whereas bricks and mortar were used for the construction of the arches and the groin vault (Fig. 23).

Figure 24. The groin vaulted structure after grouting.

It has to be noted that the groin vault was built without any formwork, following the traditional way of byzantine masons, as reported by Delinikolas et al. 2003. The total dimensions of the model are in plan 2.70 m x 2.60 m; the thickness and the height of the walls are 0.45 m and 2.60 m respectively, whereas the thickness of the vault in its centre is 0.20 m. The total height of the model is 2.85 m. These dimensions were selected taking into account the limitations imposed by the capacity of the seismic simulator.

The model was tested after nine months from its construction by imposing seismic loads gradually increasing, until rupture. Then local application of grouts and installation of ties to the arches was performed and after a suitable period of time the tests were repeated, until rupture. The model was then injected with the final grout composition (Table 3) to homogenize the whole structure, following the same methodology with that used for the injection of the wallettes and the injection of the monument itself (Miltiadou-Fezans et al. 2006, 2008). Tests were again repeated until rupture. In Figure 24 the model after the grouting application is shown. As expected, it was observed that the dynamic characteristics ofthe model were changed after the application of grouting to the whole structure, and the model could suffer stronger base motions. The results of this experimentation are still under elaboration and are going to be presented in a separate paper. An attempt will be made to correlate these results with those of the wallettes and the in situ seismic monitoring.

Was this article helpful?

0 0
Renewable Energy Eco Friendly

Renewable Energy Eco Friendly

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable.

Get My Free Ebook

Post a comment