Evolution of the Creep Coefficient

The creep coefficient at a defined time t, pc(t), may be calculated as the ratio between the deflection produced by creep dp(t) and d(to), as indicated in Eq. (2).

Although this is not the classical formulation of creep, previous studies report the possibility of determining the creep coefficient in this way when the deformation is not directly measured.

Table 6 summarizes the creep coefficients at 15, 30, 90 and 150 days as well as the increments between these periods of time and Fig. 6 shows the whole evolution of the creep coefficient through time. The results of the beams presented in Table 6 correspond to the specimens without aluminium wrapping and did not suffer any load variations during the creep test. No results of GF2 are available from the day 47 onwards as a result of its failure.

The values of the creep coefficient were found to be in all cases higher in glass fibre reinforced beams than in steel fibre beams. At the day 15, GF3 presented the lowest creep coefficient of glass fibre beams, which was the same as the highest creep coefficient of steel fibres obtained in SF2. As reported in previous investigations [12], this effect may be attributed to the different elastic modulus of the two different concretes, since the higher capacity of deformation of glass fibres is a

Table 6 Creep coefficients in non-wrapped beams

Beam

Uc(15)

Uc(30)

Uc(90)

Uc(150)

Uc(15—30) (%)

Uc(30—90)

Uc(90—150)

SF1

0.25

0.35

0.61

0.75

44.5

71.3 %

23.6 %

SF2

0.26

0.36

0.47

0.54

39.2

31.5 %

15.0 %

SF3

0.21

0.30

0.37

0.42

41.5

23.9 %

15.1 %

GF1

0.71

1.09

1.88

2.27

52.9

72.4 %

20.8 %

GF2

0.45

0.73

-

-

61.8

-

-

GF3

0.26

0.40

0.70

0.88

50.4

75.4 %

26.0 %

result of the lower elastic modulus and tensile strength when compared to steel fibres (see Table 2).

In the case of SF, the greatest increments of the creep coefficient were produced between the days 15 and 30, whereas in GF the biggest increment occurred between the days 30 and 90. Nevertheless, in both types of beam, from the day 90 the creep coefficient reduces its value and the increments between 90 and 150 days experience a drop in comparison with those obtained for 30-90 days. Longer periods for the creep test would provide further information to make long term predictions and consider the possibility of sudden failure of fibres.

The variability in the results between beams is also noticeable. The variation of the creep coefficient in SF for days 15 and 30 was around 10 %, whereas the CV in GF at the same days was approximately 47 %. These CV increased at days 90 and 150, achieving an average value of 27 % in SF and 63 % in GF.

The creep coefficients of the wrapped beams are shown in Table 7 and their evolution during the test including the load changes in Fig. 7. Regarding the values gathered in Table 7, given that these beams experienced different load levels, the creep coefficients are shown with respect to the load level at which they were obtained. At days 15 and 30 two coefficients are presented since those were the days when the load levels were changed. Two additional coefficients (days 90 and 150) are also presented, these only for GF yet beams with SF collapsed between 2 and 4 days after the load was increased to 50 % of Pcr at the day 30.

The creep coefficient at 15 days for the same load level was in average two times roughly higher in GF than in SF. At 15 and 30 days with a load level of 35 % P*r, Uc of GF was found to be again higher than the coefficient for SF. The coefficient kept growing but at 150 days it decreased around 16 % in comparison with the value obtained at 90 days.

However, the biggest increments were detected at the load level changes rather than those produced as a result of the creep. An example of this situation is GF6, which accidentally failed as a result of an excessive load level. Regarding the rest of the beams, an increase of the load from 25 % of P*r to 35 % produced a boost of around 150 % in SF, whereas for GF that percentage was approximately 80 %. When in SF the load was changed at the day 30 to 50 % of Pcr the increments were of 315 % in SF6-SF7 and 206 % in SF8, producing the collapse of the beams between 2 and 4 days later.

Table 7 Creep coefficients in wrapped beams

Beam

Uc(15)

Uc(15)

Uc(30)

Uc(30)

Uc(90)

Uc(150)

(25 % P*„)

(35 % P*„)

(35 % P*„)

(50 % Pcr)

(35 % P*cr)

(35 % Pc;)

SF6

0.19

0.50

0.63

2.63

-

-

SF7

0.24

0.65

0.85

3.51

-

-

SF8

0.35

0.77

1.08

3.30

-

-

GF7

0.58

1.10

1.67

-

2.12

1.73

GF8

0.52

0.89

1.27

-

1.38

1.17

Evolution of creep coefficient in wrapped beams

Fig. 7 Evolution of creep coefficient in wrapped beams

 
Source
< Prev   CONTENTS   Source   Next >