Fire test of ventilated and unventilated wooden façades

(1)

Lars Boström, Charlotta Skarin, Mathieu Duny,

Robert McNamee

S

P

T

ec

hni

ca

l R

es

ear

ch I

ns

tit

ut

e of

S

w

ed

en

(2)

Fire test of ventilated and unventilated

wooden façades

Fire test of ventilated and unventilated

wooden façadesFire test of ventilated

and unventilated wooden façades

Lars Boström, Charlotta Skarin, Mathieu Duny, Robert

McNamee

(3)

Abstract

Fire test of ventilated and unventilated wooden façades

Three large scale façade tests in accordance with SP Fire 105 as well as an ad hoc SBI test have been carried out. The façade tests included an inert façade made of lightweight concrete, one façade with a plywood cladding and finally a façade with plywood cladding with a fully ventilated cavity behind the cladding. The SBI test was made with plywood without ventilation cavity. The aim of the tests was to perform well controlled tests with numerous of measurements including heat release rate, heat from the combustion chamber, temperature on the façade surface, heat flux, plume temperatures and temperatures in the ventilation cavity. The results from the tests will be used for

validation of simulation techniques as well as input for further development of the façade test methodology.

Conclusions from the study were that the surface temperature for charring of the plywood cladding in this configuration was in the region of 300 °C and that the energy release originated from the façade during the test was almost twice as high when there was a 20 mm wide cavity behind the plywood cladding.

Key words: fire test, facade, wood, ventilation cavity

SP Sveriges Tekniska Forskningsinstitut

SP Technical Research Institute of Sweden SP Report 2016:16

ISBN 978-91-88349-20-0 ISSN 0284-5172

(4)

Preface

The façade is an important part of a building. It forms the envelope of the building, and has several functions. When combustible materials are used, or there are other risks in case of fire, i.e. large parts of the façade may fall down, it is necessary to evaluate the behavior of the façade when exposed to fire.

The present project has mainly been an experimental study, involving large test

specimens. We would like to thank the eminent group of technicians who have taken care of all practical details within a very pressed time schedule, no one named, no one

forgotten. We would also like to thank Mr. Pär Johansson for valuable discussions on what, how and where to do all the different measurements and what the results might represent.

The project is part of a collaboration between Centre Scientifique et Technique de Bâtiment (CSTB) in Paris, France, and SP Fire Research in Borås, Sweden, who both have financially supported the study.

(6)

Summary

Three well documented façade fire tests in accordance with SP Fire 105 have been carried out. The tests consisted of one inert façade, one façade with a plywood cladding and finally one façade with a plywood cladding with a fully ventilated cavity behind the cladding. In all three tests numerous of measurements were made such as heat release rate, heat from the combustion chamber, surface temperatures, plume temperatures, heat flux, and temperatures in the ventilation cavity. In addition to the full scale façade tests, a SBI test was made in order to evaluate how flame spread on the surface can be

determined by temperature measurements.

There are some different aims with the study. Firstly, the experiments should be well documented and measurements should be done so the results can be used for validation of different simulation techniques, such as CFD (Computational Fluid Dynamics)

modelling. Another aim was to assemble good data for the ongoing development of the next generation façade test methodology. Currently there are many different façade test methods, often national methods, which makes it difficult for industry. There is thus a demand for a harmonized test method. In the present study techniques for measurement of flame spread on a surface as well as how to measure or determine the heat exposure to the test specimen has been investigated.

The present report is mainly an assembly of test results and a description of the tests carried out. The analysis of the results is limited, and more extended analysis will be published elsewhere. Furthermore, the tests made are limited to one type of combustible material, i.e. plywood, and thus more data will be needed for validation of the technique to measure fire spread on the surface and in cavities. Conclusions from the study were that the surface temperature for charring of the plywood cladding in this configuration was in the region of 300 °C and that the energy release originated from the façade during the test was almost twice as high when there was a 20 mm wide cavity behind the plywood cladding.

(7)

1 Introduction

1.1 Background

Façades are important from several perspectives for a building. It forms the envelope of the building and shall thus be the protection towards the outdoor climate, it is an

important part controlling the energy demand of the building and it is also the face of the building and has thus architectural requirements. The façade has thus a number of functions for the building.

During the last decades the use of combustible materials in buildings has increased, partly as a consequence of higher demands on energy efficient buildings. In order to determine the fire risks for different types of façade systems a good and reliable test methodology is needed. There are presently a large number of different test methods used around the world [1] which all have certain advantages but also disadvantages. In Sweden the test method SP Fire 105 [2] is currently used, but it is currently under a major revision. The Swedish test method was originally developed in the middle of the 1980-ties as a part of a large study including burning of 14 façades with a fully developed room fire as a heat source [3]. The fire exposure was optimised in the SP Fire 105 test to correspond to the large test. In the large scale tests the fire load density was 110 MJ/m2_{of total surface area}

of the enclosure. The enclosure dimensions were 4 x 2.2 x 2.6 m3_{with the opening factor}

0.06 m1/2_{. During the tests approximately 50% of the fuel was burning outside the}

chamber. This factor is dependent on the geometry of the enclosure and on fuel

parameters such as composition and geometry. In the SP Fire 105 test the fire source is 60 litres of heptane burning in trays with attached flame suppressors.

The present study has the aim to give input data for the revision of SP Fire 105 as well as contribute with valuable data for validation of CFD modelling (Computational Fluid Dynamics). Another part of the study is to measure the difference in flame spread of combustible façade claddings with and without a ventilation cavity.

One important aspect is how the flame spread on the surface, and in the cavity shall be determined. Presently the flame spread is determined by visual observations after the fire test in SP Fire 105, i.e. by visual observation of what type of damage has occurred, and how high up on the façade. This is a blunt tool, which in some cases is dependent on the persons doing the observation especially when much soot, tar and char is mixed in the area where the flame front is. In other methods measurements or estimations are done in different ways. In the British method BS 8414-1 [4] temperature is measured a distance of 50 mm from the surface at different heights. In the international standard ISO 13785-2 [5] a surface temperature as well as a temperature 10 mm from the surface are measured, in addition to heat flux measurements at different heights.

1.2 Limitations

Only one test has been performed on each type of test setup. Even if the tests have been performed under controlled conditions, the fire behaviour in the combustion chamber and how it is evolving and protrude out of the combustion chamber might slightly change from test to test since it is a free burning test, i.e. it is not possible to control how the

(8)

discussion on trends observed. Only a limited analysis of the results is included in the present report. These data will be used in other projects aiming for more thorough analysis.

(9)

2 Experimental setup – SP Fire 105

2.1 Test method

The experiments has been based on the Swedish test method SP Fire 105 [2]. In the present tests the façade cladding has been mounted on a lightweight concrete backing with the maximum density of 600 kg/m3. Openings for two fictitious windows have been used in the tests. The total area of the façade was 4 x 6 m (width x height). A principle drawing of the façade rig is presented in Figure 1.

Figure 1. Test rig used in SP Fire 105 [2].

Section A-A 6710 600 1200 1500 1200 1500 590 710 3000 500 500 4000 2700 2700 100 150 1510 1245 1245 6000 600 600 A A Fire Room Storey 1 Storey 2 Storey 3 Fire Source mm Eave Fictitious window Lightweight concrete wall Heat flux meter Steel frame

(10)

Figure 2. Principle drawing of the fire source.

There is an air intake directly behind the fuel source with the dimensions 3140 x 300 mm, centrally placed 50 mm from the back wall of the fire room, see figure 3.

Figure 3. Drawing of the fire room.

The test rig was placed in the large fire hall at SP Fire Research in Borås, Sweden. The fuel trays as well as the heptane and water used were conditioned to the room temperature before any tests were performed. When filling the fuel into the trays first the right amount of heptane was filled. Thereafter water was added until the liquid surface reached the

100

2000

Tray of steel sheet 2000 x 500 x 100 mm Water

Heptane

Fire Source - Heptane

Water to raise the surface of the heptane fuel to the lower edge of the flame suppressor at the start of the fire test.

Flame suppressor

mm

20

Flame suppressor

of perforated steel sheet, hole Ø=25 mm, center distance 40 mm.

(11)

creates a wind. Some measurements have been done showing that the wind is in the order of 0.1-0.5 m/s, and in the direction parallel to the face of the façade rig, going from right to left side of the façade.

2.2 Test specimens

Three different tests were carried out. The first test was made without any cladding on the lightweight concrete, i.e. an inert façade. In the second test a cladding of 24 mm thick plywood was mounted directly onto the lightweight concrete, i.e. no ventilation cavity behind the plywood. In the third test plywood was mounted with a ventilation cavity of approximately 20 mm between the lightweight concrete and the plywood.

The plywood was bought at a local store for building materials, and it was a standard product for the Swedish market, with the dimensions 1200 x 2400 mm2_{. Samples were}

taken from the plywood for measurement of moisture content and density. These measurements were made approximately at the time of the tests. The plywood had a moisture content of 6.5 % and the density at that moisture content was determined to 489 kg/m3_.

The plywood was mounted on the whole front surface of the test rig, except at the fictitious windows.

In order to get an air cavity behind the plywood for Test 3, strips of mineral wool, with density of 180 kg/m3_{, was used. These strips had a thickness of 20 mm (giving an air gap}

of 20 mm) and a width of approximately 30 mm. The mineral wool strips where mounted along all vertical edges of the plywood boards, i.e. they were not acting as fire stops in the vertical direction, although they were acting as fire stops in horizontal direction in the cavity.

2.3 Instrumentation and measurements

The tests were performed with a lot of extra instrumentation in addition to the required instrumentation in accordance with SP Fire 105. Figure 4 shows the placement of different measuring devices on the surface of the test specimens.

Devices C1-C19 are 0.5 mm wire diameter thermocouples, type K. The measuring point of the thermocouples were made with Quick-Tips, i.e. a small metal ring, diameter 3 mm and length 2 mm, that clamps the two wires together. The weight of a this size Quick-Tip is 0.1 g. In all tests the tip of the thermocouple (the Quick-Tip) was pressed into the surface of the exposed surface of the façade, i.e. the measurement represents

approximately the surface temperature of the façade. In Test 3, plywood with a cavity, thermocouples were also mounted in the plywood surface on the cavity side at the same positions as on the front surface.

Devices PT1-PT6 are conventional 100 x 100 mm plate thermometers, i.e. the same devices as used in fire resistance furnaces, see for example EN 1363-1 [6]. PT1-PT3 were placed on a distance 500 mm from the combustion chamber and pointing towards the fire,

(12)

Device HF1 is a Schmidt Boelter heat flux meter. The device was mounted flush with the surface of the fictitious window which is specified in SP Fire 105 [2].

Figure 4. Placement of measuring devices on the test specimens. PT1-PT3 are placed 500

mm outwards from the combustion chamber and directed towards the fire.

In addition to these measurements temperature was also measured at a distance 100 and 200 mm from the façade with 0.5 mm type K thermocouples with a welded tip. These temperature measurements were performed at the same positions as thermocouples C2, C3, C6, C7, C13, C14, C17 and C18. The same positions were also used in Test 3 in the ventilation cavity where the thermocouples were mounted in the centre between the lightweight concrete and the plywood, that is 10 mm from the plywood surface. Furthermore, as specified in SP Fire 105, two 0.5 mm type K thermocouples with a

(13)

(14)

3 Experimental setup – SBI

A complimentary test was performed with the Single Burning Item (SBI) method, EN 13823 [8]. The SBI test is not a suitable method for testing façades due to several reasons, but it can be used for evaluation of the temperature measurements used for determination of fire spread on the façade surface. In the test two panels are used, one large wing with the dimensions 1.0 x 1.5 m2_{(width x height), and a small wing with the}

dimensions 0.49 x 1.5 m2_{. In the bottom corner between the two panels a sand burner is}

used, which is burning with a constant effect of 30 kW.

In the present study a total of 40 thermocouples were mounted on the surface of the test specimen, 20 thermocouples on the large wing, and 20 thermocouples on the small wing, see figure 5. The thermocouples were positioned at different heights, as shown in figure 5, and at different distance from the corner (100, 200, 300 and 400 mm).

Figure 5. Position of temperature measurements and type of hot tip of the thermocouples.

The thermocouples were all of the same type, 0.5 mm diameter wire and of type K. The hot tip of the thermocouples, i.e. the tip where the two wires are connected, where

(15)

Tip, i.e. a cylinder with diameter 3 mm and length 2 mm and a weight of 0.1 g, is clamped around the two wires. The Quick-Tip is then pressed into the surface of the specimen so it will be flush with the surface. This is a robust method which enables the measuring point to be pressed into the surface. Although, the thermal inertia is higher due to the mass of the Quick-Tip, compared with welded wires, and thus the response time is slightly longer for this type of thermocouple.

At the height 500 mm and 900 mm the thermocouple was welded to a copper disc with diameter 10 mm, i.e. the same type used for measurement of surface temperature in fire resistance tests in accordance with EN 1363-1 [6], but without using the insulation pad. The thermocouple and the copper disc were attached to the surface by clamping and tape. With this type of thermocouple the measured temperature is in principle averaged over a larger area, i.e. the area of the highly conductive copper disc. Although, the copper disc protects the underlying material for radiation from the fire, but it may be possible to detect eventual fire spread on the surface around the copper disc.

At the height 600 mm and 1000 mm a welded tip of the thermocouples were used. The wire was attached to the test specimen by clamping and tape. This type of thermocouple tip has the lowest thermal inertia and thus it gives the fastest response time in this study. Although, it is more difficult to mount it flush to the surface for estimations of surface temperature.

At the height 800 mm the thermocouples were welded to a horizontal steel wire, i.e. all thermocouples were fixed to the same steel wire. The steel wire was clamped to the test specimen. The idea with this type of mounting was mainly to have a method for easy mounting of thermocouples.

(16)

4 Test results

4.1 General observations

Three tests were performed indoors in the large test hall at SP. In Test 1 the backing material of the façade rig consisted of lightweight concrete and was used without any further cladding. In Test 2 the lightweight concrete was covered with the plywood cladding. In Test 3 the same type of plywood cladding was used, with a 20 mm air gap, cavity, between the lightweight concrete and the plywood.

The lightweight concrete on the façade rig, i.e. the inert façade, was in equilibrium with approximately 50 % relative humidity, and had a density of 550 kg/m3_.

The moisture content and the density of the plywood was measured at the time of testing. The moisture content was 6.5 % and the density at that moisture content was determined to 489 kg/m3_.

Table 1 gives some general information regarding the three tests made.

Table 1. General information on the tests

Test 1 Test 2 Test 3

Test date 2015-11-26

(morning) 2015-11-26 (afternoon) 2015-11-27 (morning)

Initial air temperature 18.8 °C 17.3 °C 18.0 °C

Initial relative humidity 38 % 40 % 49 %

Temperature of heptane and water 19.2 °C 16.2 °C 15.9 °C In tables 2-4 the visual observations made during the test are presented. It shall be noted that the visual observations represent the observations made during the tests, and that the actual issue observed may have occurred earlier than the noted time.

Table 2. Visual observations during Test 1 – inert façade.

Time

min:s Observations

00:00 The fire source of heptane is ignited. The test starts. 03:30 Smoke emerges out from the fire room.

04:00 Dark smoke emerges out from the fire room.

04:10 Flames and dark smoke emerge out from the fire room.

04:35 Some flames reach up to the lower edge of the lower window opening. 05:30 Some flames reach up to the upper edge of the lower window opening. 07:00 Flames reach half the height of the façade.

08:20 Flames reach up to the lower edge of the upper window opening. 10:30 Flames reach half the height of the façade.

12:50 Some flames reach up to the lower edge of the upper window opening. 14:00 Some flames reach just above the lower edge of the upper window opening. 15:30 Some flames reach up to the lower edge of the upper window opening. 16:50 The flames cease.

17:00 Flames reach just out from the fire room. Dark smoke comes out from the fire room. 18:00 The test terminates. The fire source has become extinct. The test specimen is

(17)

Time

00:00 The fire source of heptane is ignited. The test starts. 03:00 Smoke emerges out from the fire room.

04:30 Flames reach up to the lower edge of the lower window opening. 05:00 Flames reach above the upper edge of the lower window opening. 05:25 Flames reach the eave.

05:50 It burns along the upper edge of the lower window opening. The plywood panel between the window openings glows.

07:00 It burns in the lower left corner of the upper window opening. 09:00 Flames reach up to the lower edge of the upper window opening. 10:00 Flames cease and reach half the height of the façade.

11:45 Flames along the right side of the façade reach up to the upper edge of the upper window opening.

12:15 Flames reach the eave. It burns in the lower right corner of the upper window opening.

13:00 It burns along the left side of the upper window opening.

13:45 It burns in the plywood panel on the right side between the window openings. 14:30 Flames along the left side of the façade reach up to the lower edge of the upper

window opening.

15:30 Flames reach half the height of the façade. Small glowing pieces from the plywood fall down in front of the façade.

16:00 Flames reach up to the upper edge of the lower window opening. 16:15 Flames reach up to the lower edge of the lower window opening. 16:30 Flames reach just out from the fire room.

17:15 The test terminates. The fire source has become extinct. The test specimen is extinguished with water.

(18)

Table 4. Visual observations during Test 3 – plywood with cavity.

Time

00:00 The fire source of heptane is ignited. The test starts.

03:00-04:00 Smoke emerges out from the fire room.

05:00 Flames reach up to the lower edge of the lower window opening. 05:45 Flames reach up to the upper edge of the upper window opening. 06:00 It burns along the upper edge of the lower window opening

06:30 It burns along the lower edge of the upper window opening along the left side. 07:30 Flames reach the eave.

09:40 It burns along the upper edge of the lower window opening and along the lower edge of the upper window opening.

10:30 Flames reach almost to the eave.

11:00 It burns along the left side of the upper window opening. 11:20 It burns in the upper left corner of the upper window opening. 11:40 It burns along the upper edge of the upper window opening. 12:10 The lower left panel is not attached to the wall on the right side. 12:40 Flames reach the eave.

13:00 Part from the lower left panel falls down in front of the façade. 13:40 Some small burning pieces on the floor in front of the façade. 14:00 Flames reach above the eave.

14:20 It glows in the air gap by the eave. 16:00 Flames above the eave.

17:00 Flames cease. Dark smoke.

17:45 The test terminates. The fire source has become extinct. The test specimen is extinguished with water.

Photographs taken during the tests are presented in figure 6. It can be noted that visual observations on the flame spread can be difficult due the extensive smoke production during the tests.

(19)

Test 1. Time: 09:00 min:s Test 2. Time: 09:00 min:s Test 3. Time: 07:00 min:s

Test 1. Time: 12:00 min:s Test 2. Time: 11:00 min:s Test 3. Time: 12:00 min:s

(20)

4.2 Fire source

The temperature from the fire source was measured by three plate thermometers placed 0.5 m from the opening of the combustion chamber (measured from the inner side of the lightweight concrete). The plate thermometers were pointing towards the fire, and would thus measure the heat exposure mainly from the fire source, see figures 4 and 7.

Figure 7. Position of plate thermometers in front of the combustion chamber.

The measured temperatures in front of the combustion chamber from each test are presented in figures 8 – 10. It is clearly shown in the growth phase in the figures that the flames from the combustion chamber is more intense on the left side compared with the right side in all tests, which also was noted visually during the tests. During the growth phase the buoyancy force in the plume is apparently not strong enough to create a vertical flow speed high enough to counteract the side winds in the laboratory. It can also be noted that it is generally hotter in the centre, but there are shorter periods during the tests where it is hotter on the left side.

1300 710 590 100 Fire source Air intake PT (Plate thermometer)

(21)

Figure 8. Temperature measured with plate thermometers 0.5 m from the combustion

chamber in Test 1, inert façade. PT1 on left side, PT2 at the centre, and PT3 on the right side.

chamber in Test 2, plywood façade without ventilation gap. PT1 on left side, PT2 at the centre, and PT3 on the right side.

0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 1 - inert facade

PT1 PT2 PT3 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

PT1 PT2 PT3

(22)

chamber in Test 3, plywood façade with a ventilation gap. PT1 on left side, PT2 at the centre, and PT3 on the right side.

A mean temperature from the three plate thermometers in each test is shown in figure 11. The start of the third test was slower compared to the two first tests. Therefore a small time compensation of 30 seconds have been subtracted for the third test. The results show that the initial phase of the fire is very similar in all tests, and the duration of the tests is the same. Although there is a difference in the interval from 5 minutes up to 15 minutes, where the façade with plywood cladding had both the highest and lowest temperatures. The plywood façade with cavity gap showed the lowest temperatures in front of the combustion chamber. 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

PT1 PT2 PT3

(23)

Figure 11. Mean temperature measured with plate thermometers 0.5 m from the

combustion chamber.

4.3 Heat release rate

The heat release rate was measured by collecting the exhaust gases in a hood above the façade test rig [9].

The measured heat release rate is presented in figure 12. It can be noted that the initial phase of the test, up to 5 minutes, is similar in all tests, i.e. no difference between a combustible surface and an inert façade. From 5 minutes up to 10 minutes, the behaviour of the two tests with plywood is similar, i.e. the ventilation cavity does not influence on the heat release rate. The heat release rate increases with approximately 500 kW for the combustible façades compared with the inert one.

After 10 minutes into the tests, the heat release rate start to increase again, and here a difference shows up between the unventilated plywood façade and the one with a

ventilation cavity. After 15 minutes the peak is reached in Test 3, and the heat release rate is approximately 500 kW higher compared to the peak value obtained with plywood façade without ventilation, and more than 1000 kW compared to the inert façade.

Mean temperature from each test

Test 1 - inert facade

Test 2 - plywood without ventilation Test 3 - plywood with ventilation

(24)

Figure 12. Heat release rate measured during the fire tests.

The total energy produced during the tests has been calculated as the area below the heat release curves. In Test 1 the total released energy was determined to 1550 MJ, in Test 2 it was 1870 MJ, and in Test 3 it was 2140 MJ. Thus the additional energy from the plywood in Test 2 was 320 MJ, and in Test 3 the additional energy from the plywood was 590 MJ. This means that the additional energy from the plywood is almost doubled by the

ventilation cavity.

4.4 Thermal exposure on the façade above the

combustion chamber

The thermal exposure was measured with a plate thermometer placed 750 mm above the upper edge of the combustion chamber, plate thermometer PT4 in figure 4. The plate thermometer was mounted on the façade surface pointing outwards. The measurement thus represent the heat exposure to the façade surface.

The results from the three tests are presented in figure 13. It is clear that on the height 750 mm above the combustion chamber there is almost no difference in heat exposure

towards the façade, i.e. if there is a contribution from the burning plywood to the thermal exposure 750 mm above the combustion chamber, it is minimal compared with the thermal exposure from the combustion chamber. Theoretically it is also possible that there is no contribution from combustion from the surface material originated in this zone as the diffusion flames from the combustion chamber has a very low oxygen content, i.e. release of pyrolysis gasses happens from the materials in this zone but the actual burning is higher up in the fire plume where more oxygen is blended in.

0 500 1000 1500 2000 2500 3000 3500 4000 0 5 10 15 20 Hea t r el ea se ra te ( kW ) Time (minutes)

Test 1 - Inert facade

Test 2 - plywood without ventilation Test 3 - plywood with ventilation

(25)

Figure 13. Thermal exposure to the façade 750 mm above the combustion chamber as

measured with a plate thermometers directed outwards.

4.5 Thermal exposure and heat flux one storey above

the combustion chamber

In the centre of the fictitious window one storey above the combustion chamber the thermal exposure towards the window was measured with a plate thermometer, i.e. 2.1 m above the upper edge of the opening to the combustion chamber, plate thermometer PT5 shown in figure 4. The measured temperatures are shown in figure 14.

As expected the inert façade show the lowest temperature. Comparing the two plywood façades the temperature development is similar during the first 8 minutes, at that time the temperature is approximately 650 °C. Thereafter the temperature in the window cavity during the test of the specimen without air cavity behind the plywood cladding starts to decrease slowly, whereas the temperature continues to increase for the specimen with a ventilation gap behind the plywood cladding and reaches a maximum of 750 °C after 13 minutes.

It is clear that a combustible façade cladding increases the heat exposure to the window one storey above the fire source, and that a ventilation gap behind the combustible cladding increases the heat exposure even further.

0 100 200 300 400 500 600 700 800 900 1000 0 5 10 15 20 Te mp er atu re (° C) Time (minutes)

Test 2 - plywood without cavity Test 3 - plywood with cavity

(26)

Figure 14. Thermal exposure one storey above the fire source measured with plate

thermometer directed outwards.

In addition to the temperature, the heat flux was measured with a Schmidt Boelter total heat flux meter at the centre of the fictitious window (gauge HF1 shown in figure 4). The measured heat fluxes are shown in figures 15 and 16. In figure 16, the mean heat flux for each minute is shown.

Test 2 - plywood without cavity test 3 - plywood with cavity

0 10 20 30 40 50 60 70 80 90 0 5 10 15 20 He at fl ux (k W /m 2) Time (minutes)

(27)

Figure 16. Heat flux (mean valued over one minute) measured one storey above the fire

source.

4.6 Thermal exposure two storeys above the

combustion chamber

In the centre of the fictitious window two storeys above the combustion chamber the temperature exposure towards the window was measured with a plate thermometer, i.e. 4,8 m above the upper edge of the opening to the combustion chamber, plate thermometer PT6 shown in figure 4. The measured temperatures are shown in figure 17.

At this height there is a clear difference between the tests. The ventilation cavity used in Test 3 has a significant effect on the heat exposure to the fictitious window two stories above the fire room. After 15 minutes the temperature is over 600 °C in Test 3. When comparing Test 1 and Test 2, i.e. the inert façade and the plywood cladding without cavity, the difference between these tests is relatively small. The façade with plywood cladding have a faster temperature rise between 5 and 7 minutes, after which the temperature stabilizes between 200-250 °C. The inert façade has a slower temperature rise, but it reaches almost as high in temperature.

0 10 20 30 40 50 60 70 80 0 5 10 15 20 He at fl ux (k W /m 2) Time (minutes)

(28)

Figure 17. Thermal exposure two storeys above the fire source measured with plate

thermometer directed outwards.

4.7 Surface temperature

The surface temperature was measured by 0.5 mm wire diameter thermocouples type K with a Quick-Tip as described above. The Quick-Tip of the thermocouple was forced into the surface, measuring the temperature of the surface during the test. A total of 19

thermocouples, C1-C19, was used on the fire exposed surface of the façade. In the third test, with a cavity behind the plywood cladding, an additional 19 thermocouples, C21-C39, were positioned on the cavity side of the plywood, opposite C1-C19. Thermocouple C21 is positioned opposite C1, thermocouple C22 opposed to C2, and so on. The

positions of thermocouple C1-C19 are shown in figure 4.

4.7.1 Temperature 1500 mm above the combustion chamber

The measured temperatures on the surface at a level 1500 mm above the edge of the opening of the combustion chamber are presented in figures 18 – 20, one figure for each test. The results clearly show that the fire was more severe on the left side.

0 100 200 300 400 500 600 700 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

(29)

Figure 18. Temperature on the surface during Test 1, 1500 mm above the combustion

chamber.

chamber. 0 100 200 300 400 500 600 700 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 1 - inert facade

C1 C2 C3 C4 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

C1

C2 C3 C4

(30)

chamber.

Figure 21. Temperature on the surface of the plywood in the cavity 1500 mm above the

combustion chamber.

The temperature was measured on the surface of the plywood in the cavity during Test 3. The results from these measurements are shown in figure 21 above. Unlike the

0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

C1 C2 C3 C4 0 50 100 150 200 250 300 350 400 450 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Surface temperature in the cavity

C21 C22 C23 C24

(31)

each test, as well as the mean surface temperature on the plywood surface in the cavity. As the figure shows there is a substantial difference on the temperature level between the combustible cladding and the inert façade.

Figure 22. Mean surface temperature at a level 1500 mm above the upper edge of the

combustion chamber.

4.7.2 Temperature 2700 mm above the combustion chamber

Mean surface temperature in each test

Mean test 1

Mean test 2 Mean test 3 Mean test 3 - cavity

(32)

chamber.

chamber. 0 50 100 150 200 250 300 350 400 450 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 1 - inert facade

C5 C6 C7 C8 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

C5

C6 C7 C8

(33)

chamber.

The temperature was measured on the surface of the plywood in the cavity during Test 3. The results from these measurements are shown in figure 26. Unlike the temperatures measured on the outward surface, which decreases at the end of the tests, the temperature increases rapidly. 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

C5 C6 C7 C8 0 20 40 60 80 100 120 140 160 180 0 5 10 15 20 Te mp era tu re (° C)

Test 3 - plywood with cavity

Surface temperature of plywood in cavity

C25

C26 C27 C28

(34)

In figure 27 the mean temperature of all thermocouples at this level presented for each test, as well as the mean surface temperature on the plywood surface in the cavity. As the figure shows there is a substantial difference on the temperature level between the combustible cladding and the inert façade.

combustion chamber.

4.7.3 Temperature 3450 mm above the combustion chamber

Mean surface temperature in each test

Mean test 1

(35)

chamber.

chamber. 0 50 100 150 200 250 300 350 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 1 - inert facade

C9 C10 C11 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

C9

C10 C11

(36)

chamber.

The temperature was measured on the surface of the plywood in the cavity during Test 3. The results from these measurements are shown in figure 31. At this level the temperature in the surface of the plywood in the cavity do not show the same increasing temperatures at the end of the test.

Test 3 - plywood with cavity

C9 C10 C11 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Surface temperature of plywood in cavity

C29

C30 C31

(37)

test, as well as the mean surface temperature on the plywood surface in the cavity. As the figure shows there is a substantial difference on the temperature level between the combustible cladding and the inert façade. It can also be noted that the mean surface temperature in Test 3 is approximately the same on the outer surface as on the surface in the cavity at the end of the test.

combustion chamber.

4.7.4 Temperature 4200 mm above the combustion chamber

The measured temperatures on the surface at a level 4200 mm above the edge of the opening of the combustion chamber are presented in figures 33 – 35, one figure for each test. The results clearly show that the fire was more severe on the left side in Test 2 and Test 3. In Test 1, however, the temperature readings show a more centrally exposure at this level. 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Mean surface temperature in each test

Mean test 1

(38)

chamber.

Test 1 - inert facade

C12 C13 C14 C15 0 100 200 300 400 500 600 700 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

C12

C13 C14 C15

(39)

chamber.

The temperature was measured on the surface of the plywood in the cavity during Test 3. The results from these measurements are shown in figure 36. At this level the temperature in the surface of the plywood in the cavity is continuously increasing, and accelerating at the end of the test.

Test 3 - plywood with cavity

C12 C13 C14 C15 0 20 40 60 80 100 120 0 5 10 15 20 Te mp era tu re (° C)

Test 3 - plywood with cavity

Surface temperature of plywood in cavity

C32

C33 C34 C35

(40)

In figure 37 the mean temperature of all thermocouples at this level is presented for each test, as well as the mean surface temperature on the plywood surface in the cavity. As the figure shows there is a substantial difference on the temperature level between the combustible cladding and the inert façade.

combustion chamber.

4.7.5 Temperature 5400 mm above the combustion chamber

The measured temperatures on the surface at a level 5400 mm above the edge of the opening of the combustion chamber are presented in figures 38 – 40, one figure for each test. The results clearly show that the fire was more severe on the left side of the

specimen in all three tests.

Mean surface temperature in each test

Mean test 1

(41)

chamber.

Test 1 - inert facade

Test 2 - plywood without cavity

C16

C17 C18 C19

(42)

chamber.

The temperature was measured on the surface of the plywood in the cavity during Test 3. The results from these measurements are shown in figure 41. At this level the temperature in the surface of the plywood in the cavity is continuously increasing, but decreasing at the end of the test.

0 50 100 150 200 250 300 350 400 450 500 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Surface temperature of plywood in cavity

C36

C37 C38 C39

(43)

test, as well as the mean surface temperature on the plywood surface in the cavity. As the figure shows there is a substantial difference on the temperature level between the combustible cladding and the inert façade.

combustion chamber.

4.8 Plume temperature at different heights

The temperature was also measured 100 mm and 200 mm from the surface, i.e. in the plume, with 0.5 mm thermocouples wires type K with welded tip. The thermocouples were fixed on long screws that were fixed to the façade surface. The gas temperature in the cavity in Test 3 was also measured with the same type of thermocouples, and the measurements were made in the centre of the cavity, i.e. 10 mm from the surface. Details about the positions are explained in the text following figure 4.

4.8.1 Plume temperature 1500 mm above the combustion

chamber

4.8.1.1 Measurements 100 mm from the surface

The measured temperature 100 mm from the surface at a level 1500 mm above the edge of the opening of the combustion chamber is presented in figures 43 – 45, one figure for each test. The results clearly show that the fire was more severe on the left side of the specimen in all three tests.

0 50 100 150 200 250 300 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Mean surface temperature in each test

Mean test 1

(44)

Figure 43. Temperature in the plume 100 mm from the surface during Test 1, 1500 mm

above the combustion chamber.

Test 1 - inert facade

A1 A2 0 100 200 300 400 500 600 700 800 900 1000 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

A1

(45)

In figure 46 is the mean temperature of the two thermocouples 100 mm from the surface shown for each test. The results are similar to the temperatures measured on the surface at the same level, figure 22, i.e. the lowest temperatures in Test 1, and highest in Test 2.

Test 3 - plywood with cavity

A1 A2 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Mean temperature 100 mm from surface

Mean test 1

Mean test 2 Mean test 3

(46)

4.8.1.2 Measurements 200 mm from the surface

The temperature was also measured 200 mm from the surface, but only in Test 1, i.e. the inert façade. The measured temperature is shown in figure 47. A mean temperature (mean of two measurements) is presented in figure 48 for the measurement of temperature 100 mm as well as 200 mm from the surface.

Figure 47. Temperature in the air 200 mm from the surface during Test 1, 1500 mm

Test 1 - inert facade

B1 B2 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C)

Test 1 - inert facade

Mean 100 mm from surface

(47)

4.8.1.3 Measurements in the cavity

The temperature was also measured in the cavity in Test 3, centrally between the

lightweight concrete backing and the plywood. The test results at a level 1500 mm above the fire source are shown in figure 49.

Figure 49. Temperature in the cavity 1500 mm above the fire source during Test 3.

4.8.2 Air temperature 2700 mm above the combustion

chamber

4.8.2.1 Measurements 100 mm from the surface

The measured temperature 100 mm from the surface at a level 2700 mm above the edge of the opening of the combustion chamber is presented in figures 50 – 52, one figure for each test. The results clearly show that the fire was more severe on the left side of the specimen in all three tests.

Test 3 - plywood with cavity

Temperature in cavity

D1

(48)

Test 1 - inert facade

A3 A4 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

A3

(49)

In figure 53 is the mean temperature of the two thermocouples 100 mm from the surface shown for each test. The results are similar to the temperatures measured on the surface at the same level, figure 27, i.e. the lowest temperatures in Test 1, and highest in Test 2.

Test 3 - plywood with cavity

A3 A4 0 100 200 300 400 500 600 700 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Mean temperature 100 mm from surface

Mean test 1

(50)

4.8.2.2 Measurements 200 mm from the surface

Test 1 - inert facade

B3 B4 0 50 100 150 200 250 300 350 400 450 0 5 10 15 20 Te mp era tu re (° C)

Test 1 - inert facade

(51)

4.8.2.3 Measurements in the cavity

4.8.3 Air temperature 3450 mm above fuel source

4.8.3.1 Measurements 100 mm from the surface

The measured temperature 100 mm from the surface at a level 3450 mm above the edge of the opening of the combustion chamber is presented in figures 57 – 59, one figure for each test. Also on this height the results show that the fire was more severe on the left side of the specimen in all three tests.

Test 3 - plywood with cavity

Temperature in cavity

D3

(52)

Test 1 - inert facade

A5 A6 0 100 200 300 400 500 600 700 800 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

A5

(53)

In figure 60 is the mean temperature of the two thermocouples 100 mm from the surface shown for each test. The results are similar to the temperatures measured on the surface at the same level, figure 32, i.e. the lowest temperatures in Test 1, and approximately the same temperature in Test 2 and Test 3 except at the end of the test where Test 3 show higher temperature.. 0 100 200 300 400 500 600 700 800 900 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

A5 A6 0 100 200 300 400 500 600 700 Te mp era tu re (° C)

Mean temperature 100 mm from surface

Mean test 1

(54)

4.8.3.2 Measurements 200 mm from the surface

Test 1 - inert facade

B5 B6 0 50 100 150 200 250 300 350 0 5 10 15 20 Te mp era tu re (° C)

Test 1 - inert facade

(55)

4.8.3.3 Measurements in the cavity

lightweight concrete backing and the plywood. The test results at a level 3450 mm above the combustion chamber are shown in figure 63.

4.8.4 Air temperature 4200 mm above combustion chamber

4.8.4.1 Measurements 100 mm from the surface

The measured temperature 100 mm from the surface at a level 4200 mm above the edge of the opening of the combustion chamber is presented in figures 64 – 66, one figure for each test. 0 50 100 150 200 250 300 350 400 450 500 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Temperature in cavity

D5

(56)

Test 1 - inert facade

A7 A8 0 100 200 300 400 500 600 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

A7

(57)

In figure 67 is the mean temperature of the two thermocouples 100 mm from the surface shown for each test. The results are similar to the temperatures measured on the surface at the same level, figure 37, i.e. the lowest temperatures in Test 1, and approximately the same temperature in Test 2 and Test 3 except at the end of the test where Test 3 show higher temperature.. 0 100 200 300 400 500 600 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

A7 A8 0 50 100 150 200 250 300 350 400 450 500 Te mp era tu re (° C)

Mean temperature 100 mm from surface

Mean test 1

(58)

4.8.4.2 Measurements 200 mm from the surface

Test 1 - inert facade

B7 B8 0 50 100 150 200 250 300 0 5 10 15 20 Te mp era tu re (° C)

Test 1 - inert facade

(59)

4.8.4.3 Measurements in the cavity

The gas temperature was also measured in the cavity in Test 3, centrally between the lightweight concrete backing and the plywood. The test results at a level 4200 mm above the fire source are shown in figure 70.

4.8.5 Air temperature 5400 mm above combustion chamber

4.8.5.1 Measurements 100 mm from the surface

The measured temperature 100 mm from the surface at a level 5400 mm above the edge of the opening of the combustion chamber is presented in figures 71 – 73, one figure for each test. 0 20 40 60 80 100 120 140 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Temperature in cavity

D7

(60)

Test 1 - inert facade

A9 A10 0 50 100 150 200 250 300 350 400 450 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 2 - plywood without cavity

A9

(61)

In figure 74 is the mean temperature of the two thermocouples 100 mm from the surface shown for each test. The results are similar to the temperatures measured on the surface at the same level, figure 42, i.e. the lowest temperatures in Test 1, and approximately the same temperature in Test 2 and Test 3 except at the end of the test where Test 3 show higher temperature.. 0 100 200 300 400 500 600 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

A9 A10 0 50 100 150 200 250 300 350 400 450 Te mp era tu re (° C)

Mean temperature 100 mm from surface

Mean test 1

(62)

4.8.5.2 Measurements 200 mm from the surface

Test 1 - inert facade

B9 B10 0 50 100 150 200 250 0 5 10 15 20 Te mp era tu re (° C)

Test 1 - inert façade

Mean 100 mm from surface Mean 200 mm from surface

(63)

4.8.5.3 Measurements in the cavity

4.9 Temperature below the eave

The test method SP Fire 105 include two temperature measurements 100 mm below the eave, one 100 mm from the façade surface and one 400 mm from the façade surface, as described in the text below figure 4. These measurements were made with 0.5 mm thermocouples type K with a welded tip.

Figures 78 – 79 show the results obtained in the three tests. In the third test the

thermocouple at a distance of 400 mm was out of function, so there are no measurements in this case. 0 50 100 150 200 250 300 350 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Test 3 - plywood with cavity

Temperature in cavity

D9

(64)

Figure 78. Measurements 100 mm from the façade surface, 100 mm below the eave.

Figure 79. Measurements 400 mm from the façade surface, 100 mm below the eave.

Temperature 100 mm from the façade surface

Test 1 Test 2 Test 3 0 100 200 300 400 500 600 700 0 5 10 15 20 Te mp era tu re (° C) Time (minutes)

Temperature 400 mm from the façade surface

Test 1

(65)

4.10 Results from SBI test

The aim with the SBI test was to evaluate different ways to apply thermocouples on the surface for measurement and detection of surface flame spread. The evaluation of flame spread was made after the test. The test specimen was inspected and the charred area was mapped, see figure 80. In the figure is also the position of the thermocouples shown. It was not possible during the test to see the progress of the flame spread with any accuracy, and therefore the presented results are only based on the measurements from the

thermocouples during the test, and the visual observation of charring after the test.

Figure 80. Charring on the test specimen after the test. To the left is the large wing, and

to the right the small wing.

In figures 81 – 87 are the measured temperatures presented. In the legend are the

thermocouple number given as well as which wing they were mounted on. SW is thus the small wing, and LW is the large wing. The numbering is going from the corner outwards, i.e. the lowest number is the thermocouple closest to the corner on that specific wing.

Fire test of ventilated and unventilated wooden façades

Lars Boström, Charlotta Skarin, Mathieu Duny,

Robert McNamee

S

P

T

ec

hni

ca

l R

es

ear

ch I

ns

tit

ut

e of

S

w

ed

en

Fire test of ventilated and unventilated

wooden façades

Fire test of ventilated and unventilated

wooden façadesFire test of ventilated

and unventilated wooden façades

Lars Boström, Charlotta Skarin, Mathieu Duny, Robert

McNamee

Abstract

Fire test of ventilated and unventilated wooden façades

Contents

Preface

Summary

1

Introduction

1.1

Background

1.2

Limitations

2

Experimental setup – SP Fire 105

2.1

Test method

Fire Source - Heptane

2.2

Test specimens

2.3

Instrumentation and measurements

3

Experimental setup – SBI

4

Test results

4.1

General observations

4.2

Fire source

Test 1 - inert facade

Test 2 - plywood without cavity

Test 3 - plywood with cavity

4.3

Heat release rate

Mean temperature from each test

4.4

Thermal exposure on the façade above the

combustion chamber

4.5

Thermal exposure and heat flux one storey above

the combustion chamber

4.6

Thermal exposure two storeys above the

combustion chamber

4.7

Surface temperature

4.7.1

Temperature 1500 mm above the combustion chamber

Test 1 - inert facade

Test 2 - plywood without cavity

Test 3 - plywood with cavity

Test 3 - plywood with cavity

Surface temperature in the cavity