Sustainability of Concrete Structures in Terms …
73
(reference) specimens. In this calculation, the flexural strength was replaced with
the tensile splitting strength—therefore, the frost resistance coefficient is not determined exactly by the standard [17] (which has already been noted in the text above);
however, the meaning and principle of the calculation is precisely the same.
If it is necessary to decide whether a concrete is or is not frost resistant (of course,
in relation to a certain number of F-T cycles), a minimal value must be defined
below which the frost resistance coefficient (or RDM) must not drop. The limit
value given by the Czech standard [17] is 75%, whereas the American standard [15]
prescribes to finish the test if the relative dynamic modulus falls below 60%. The
European regulations [16, 18] do not mention any criterion at all. For that reason,
75% is regarded as the limit value even for the NDT measurements in the assessment
of the experiment described. Tables 2, 3 and 4 show the average RDM values and
the frost resistance coefficients for each concrete separately and for both concretes
together (as for one statistical data set). Based on the tests, it can be concluded that
concretes A and B are probably frost resistant for 25 freezing and thawing cycles
within the meaning of the ˇ
CSN 73 1322 [17] standard. However, it must be noted that
this evaluation is not completely unequivocal. Concrete A, if assessed separately, is
not satisfactory for 50 F-T cycles according to any of the methods used and it is
satisfactory in terms of tensile splitting strength and UPV for 25 F-T cycles. Based
on the resonance method, it is not satisfactory even for 25 F-T cycles, which is caused
by considerably worse results of specimen A1 compared to the other test cylinders.
Concrete B, again if assessed separately, is satisfactory based on the average values
of all the test methods used for both 25 and 50 F-T cycles. However, the assessment
must take into consideration that after 25 F-T cycles one cylinder (B1) out of two
was not satisfactory in terms of tensile splitting strength and after 50 F-T cycles one
cylinder (B2) out of two was not satisfactory in terms of tensile splitting strength. Of
course it is a numerically completely unsatisfactory statistical set. However, the fact
that half of the specimens are not satisfactory must be taken into account. When both
concretes are assessed together (which is logical in terms of assessing the structure
as a whole but not suitable in terms of combining two statistical data sets—meaning
two concretes with different composition) the concrete is frost resistant based on
all the methods for 25 F-T cycles; based on the resonance method and the tensile
splitting strength it is not frost resistant for 50 F-T cycles.
Clearly, if specimens are obtained by core sampling and subsequent circular
cutting, the results can be different from those obtained by testing specimens from
a mould. This is true even for the same concrete being tested. The standard test specimens mature in prescribed conditions (mostly in a water bath), which cannot be said
about concrete in a structure. When extracting and adapting a test specimen from a
structure, micro cracks may occur, which might have a negative impact on its frost
resistance and, moreover, the specimens tested have an “open” surface. According to
the standard [26], if core samples are used to assess the concrete compressive strength,
it is sufficient for the result to reach 85% of the characteristic value of concrete
compressive strength. Concerning the assessment of frost resistance, it would be
appropriate to admit a similar tolerance as well. However, it should also be noted
that the drops in the flexural strength and the tensile splitting strength after the F-T
73
(reference) specimens. In this calculation, the flexural strength was replaced with
the tensile splitting strength—therefore, the frost resistance coefficient is not determined exactly by the standard [17] (which has already been noted in the text above);
however, the meaning and principle of the calculation is precisely the same.
If it is necessary to decide whether a concrete is or is not frost resistant (of course,
in relation to a certain number of F-T cycles), a minimal value must be defined
below which the frost resistance coefficient (or RDM) must not drop. The limit
value given by the Czech standard [17] is 75%, whereas the American standard [15]
prescribes to finish the test if the relative dynamic modulus falls below 60%. The
European regulations [16, 18] do not mention any criterion at all. For that reason,
75% is regarded as the limit value even for the NDT measurements in the assessment
of the experiment described. Tables 2, 3 and 4 show the average RDM values and
the frost resistance coefficients for each concrete separately and for both concretes
together (as for one statistical data set). Based on the tests, it can be concluded that
concretes A and B are probably frost resistant for 25 freezing and thawing cycles
within the meaning of the ˇ
CSN 73 1322 [17] standard. However, it must be noted that
this evaluation is not completely unequivocal. Concrete A, if assessed separately, is
not satisfactory for 50 F-T cycles according to any of the methods used and it is
satisfactory in terms of tensile splitting strength and UPV for 25 F-T cycles. Based
on the resonance method, it is not satisfactory even for 25 F-T cycles, which is caused
by considerably worse results of specimen A1 compared to the other test cylinders.
Concrete B, again if assessed separately, is satisfactory based on the average values
of all the test methods used for both 25 and 50 F-T cycles. However, the assessment
must take into consideration that after 25 F-T cycles one cylinder (B1) out of two
was not satisfactory in terms of tensile splitting strength and after 50 F-T cycles one
cylinder (B2) out of two was not satisfactory in terms of tensile splitting strength. Of
course it is a numerically completely unsatisfactory statistical set. However, the fact
that half of the specimens are not satisfactory must be taken into account. When both
concretes are assessed together (which is logical in terms of assessing the structure
as a whole but not suitable in terms of combining two statistical data sets—meaning
two concretes with different composition) the concrete is frost resistant based on
all the methods for 25 F-T cycles; based on the resonance method and the tensile
splitting strength it is not frost resistant for 50 F-T cycles.
Clearly, if specimens are obtained by core sampling and subsequent circular
cutting, the results can be different from those obtained by testing specimens from
a mould. This is true even for the same concrete being tested. The standard test specimens mature in prescribed conditions (mostly in a water bath), which cannot be said
about concrete in a structure. When extracting and adapting a test specimen from a
structure, micro cracks may occur, which might have a negative impact on its frost
resistance and, moreover, the specimens tested have an “open” surface. According to
the standard [26], if core samples are used to assess the concrete compressive strength,
it is sufficient for the result to reach 85% of the characteristic value of concrete
compressive strength. Concerning the assessment of frost resistance, it would be
appropriate to admit a similar tolerance as well. However, it should also be noted
that the drops in the flexural strength and the tensile splitting strength after the F-T