Fire Resistance of Aluminum-Glass Partitions with a Parallel Structure of Intumescent Layers

Aim: The aim of the article is to verify the fire resistance of aluminum-glass construction partitions with a parallel structure. The paper presents the results of tests in a fire chamber of a selected partition, a leading national manufacturer of aluminum-glass systems. The results of the fire tests were used to validate a simple parallel model of the reliability of non-renewable systems. Introduction: Fire-retardant properties of partitions made of glass and aluminum profiles determine their ability to stop the spread of fire by clos ing the fire in separate zones. Fire resistance of such partitions is measured according to various criteria, in particular the requirements concern: stability R (glass does not break), tightness E , radiation limitation W and insulation I . The tightness requirement E means that the partition effectively protects the fire compartment against flames, smoke and hot gases. The insulating postulate I means that the average temperature of glass and the profiles on the surface of the partition on the protected side does not exceed the contractual value during the nominal duration of the fire. The measure of fire resistance of a partition is time t , expressed in minutes, in which the structure of an aluminum-glass partition meets one or more of the criteria listed. Methodology: The experimental database consists of the results of routine fire resistance tests of system building partitions obtained in the certification process of selected facade systems. Interpretation of the obtained results of laboratory tests was based on simple models of reliability of non-renewable systems. Conclusions: The graphs of average temperature increase on the outer surface of the tested glass are the same in each case in terms of quality and quantity. Up to about 70% of the nominal fire resistance, the temperature increase is linear, followed by a non-linear phase according to a concave curve. The course of the temperature-rise curves on the outer surface of aluminum profiles is qualitatively different. The graphs are non-linear, convex from the beginning of heating, with the inflection point reached after about 20 minutes of the test. The obtained results may indicate that the multi-chamber structure of aluminum profiles does not ensure a parallel reliability structure, because the object has a quasi-parallel structure. The presented results of the temperature-rise curves and their reliability interpretation require confirmation in further laboratory tests of aluminum-glass partitions with a different structure of panes and profiles.

The temperature in the furnace during the heating of the partition was measured with 16 plate thermometers, and the results of these measurements are shown in Figure 3 which indicate good compliance of the heated air temperature in the furnace with the standard curve according to the formula (1). In Figure 3   Source: Technical report [1]. Źródło: Raport techniczny [1].

Time / Czas
Type of damage / Rodzaj uszkodzenia The internal panes of fields 2, 3, 4 and 6 break as indicated in Table 1

80′ 17′′
All panes white with brown spots / Na wszystkich szybach białych pojawiły się brązowe plamy  The results of the measurements of the temperature rise on the outer surfaces of the aluminum profiles were similar for all installed thermocouples. In Figure 5   During the heating of the partition, horizontal displacements were recorded in five selected points on the profiles (see Figure 1, points A, C and E) and shafts (see Figure 1, points D and E). The measured displacements are summarized in Table 3    View of the partition from the outside after the heating test is completed (121' of test) shown in Figure 6.

Summary
The standard requirements [3][4] for fire resistance mean that the following conditions are met in the heating test of the tested partition: -the glazing will not break under its own weight,  The laboratory test results shown in Figures 4 and 5 prove that the tested partition meets the standard criteria of the nominal fire resistance EI 120 minutes in terms of temperature rise. The technical report [1] shows that also the other fire resistance criteria of this partition were met. Fire-resistant laminated glass according to Figure 2, composed of several repetitive gel layers, is a simple, non-renewable, parallel-structured reliability system. At a temperature of 120°C, the gel layers expand to form a hard, opaque layer that provides temporary protection against fire. A similar model of reliability seems to describe five-chamber aluminum profiles filled with thermal insulation (see Figure 2). Analysis of the tem- Wykres ten jest charakterystyczny dla szyb ogniochronnych, o równoległej strukturze warstw pęczniejących.

Conclusions
Due to the standardized nature of laboratory tests, reviews and contributions devoted to the subject of fire resistance of aluminum-glass partitions are available in the national literature, for example [5][6][7][8]. The publications that provide the results of fire resistance tests for other system partitions are particularly important. This is the nature of works [9][10] in which there are diagrams of temperature increases on the outer surfaces of the glass panes and profiles. The results on the aluminum profiles in these studies are consistent with those presented in Figure 5.
It should be added, however, that the aforementioned tests [9][10] concerned partitions with a nominal fire resistance of EI 30, i.e.