EPS as standard substrate for roof coverings in ENV 1187 test method 2

(1)

SP Swedish National T

esting and Research Institute

Ingrid Wetterlund Ulla Fridh

EPS as standard substrate for roof

coverings in ENV 1187

test method 2

SP Technical Notes 2004:29 Fire Technology

(2)

(3)

Abstract

Rules for direct field of application of test results according to ENV 1187 test method 2 are given in the classification standard EN 13501-5. The rules allow direct classification for a roof covering to be applied on mineral wool based only on test results where the product has been applied on expanded polystyrene. It has been found that this in many cases may give a direct application on unfounded reasons. Roof coverings, which fulfil the criteria of EN 13501-5 when applied on expanded polystyrene, do not necessarily meet the criteria when applied on mineral wool.

There is a need for both the pre-standard for testing ENV 1187 and the classification standard EN 13501-5 to be revised with respect to the listing of standard substrates and the rules for direct application of the results. None of them should include expanded polystyrene as a standard substrates for direct application of test results. Expanded polystyrene should instead be regarded as a non-standard substrate.

Key words: Fire classification, small-scale fire tests, roof coverings

SP Sveriges Provnings- och SP Swedish National Testing and

Forskningsinstitut Research Institute

SP AR 2004:29 SP Technical Notes 2004:29 Borås 2004

Postal address:

Box 857,

SE-501 15 BORÅS, Sweden

Telephone: +46 33 16 50 00

Telex: 36252 Testing S

Telefax: +46 33 13 55 02

(4)

Sammanfattning

Regler för direkt applicering av provningsresultat från provning enligt ENV 1187 metod 2 finns angivna i klassifikationsstandarden EN 13501-5. Reglerna tillåter att en taktäckning direkt kan klassificeras för läggning på mineralull enbart baserat på provning av produkten lagd på expanderad polystyren. Det har visat sig att detta i många fall ger en klassifikation på osakliga grunder. Taktäckningar som uppfyller kraven enligt EN 13501-5 när de applicerats på expanderad polystyren uppfyller inte nödvändigtvis kraven när de appliceras på mineralull.

Både förstandarden ENV 1187 och klassifikationsstandarden EN 13501-5 behöver revideras med avseende på vilka standardunderlag som listas och hur reglerna för direkt applicering av provningsresultat är utformade. Ingen av dem bör ange att expanderad polystyren är ett standardunderlag som kan leda till direkt applicering av

provningsresultat. Expanderad polystyren bör i stället betraktas som ett specialunderlag (tidigare betecknat ”särligt underlag” i NKB’s produktregler 15 [1]).

(6)

1 Direct application of test results from

ENV 1187 test method 2

Fire performance classification of roof coverings is performed according to procedures given in EN 13501-5 [2]. The classification standard refers to the pre-standard for testing ENV 1187 [3] where three test methods are listed. Both the pre-standard for testing and the classification standard contain requirements for how to prepare the test specimens and what substrates to be used. A number of special conditions as well as instructions

regarding direct field of application are given in EN 13501-5 [2] for each of the test methods.

The special conditions for test method 2, which is a test method with burning brands and wind, are the four different standard substrates listed in Table 1.

Table 1 Standard substrates for test method 2 of ENV 1187

Standard substrate Density, kg/m2 Thickness, mm

Wood particle board 680±50 19±2

Combustible substrate Expanded polystyrene 20±5 50±10 Fibre reinforced calcium silicate board 680±50 10±2 Non-combustible substrate Mineral wool 150±20 50±10

The rules for direct application of the test results with respect to the substrate are given in clause 6.4.3.3 of EN 13501-5 [2]. The list below is an extract from that standard:

b) Test results obtained for a roof covering attached to a substrate apply only for the roof covering on substrates having a density greater than or equal to 0,75 times the density used in the test.

c) Test results obtained on a non-combustible standard substrate apply only for non-combustible substrates which also comply with 6.4.3.3b).

d) Test results obtained on a combustible standard substrate apply to combustible and non-combustible substrates which also comply with 6.4.3.3b).

(7)

2 Direct application is based on the worst case

The fire behaviour of a roof covering depends very much on the thermal inertia of the substrate. The lower the thermal inertia is the more heat will remain in the roof covering itself and the flame spread may therefore increase. Thus, the worst case of testing a roof covering is to attach it to the combustible standard substrate having the lowest density, which in this case is the expanded polystyrene. Hence, applying the direct application rules in clause 6.4.3.3 of EN 13501-5 [2] leads to a full classification on all the other standard substrates.

Everything would be straightforward if the expanded polystyrene had remained intact during test. This is unfortunately not the case. The polystyrene melts away from the fire zone by the heat. A big hole is created under the fire zone. This hole either causes the insulating capacity to be highly reduced or induces air currents that cool the roof covering. The result is that the flame spread in a roof covering attached to expanded polystyrene is much shorter than when it is attached to mineral wool.

Using expanded polystyrene as substrate for roof coverings and believing that this would give a judgement based on the worst case is to be deceived. The results in 3 clearly show that the rules for direct application do not have a sound base.

(8)

3 Comparison of test results where the

substrates expanded polystyrene and

mineral wool have been used

A number of roof coverings have been tested at SP on the substrates mineral wool and expanded polystyrene. The test specimens for each of the roof coverings were all taken from the same production batch. For some of the products the comparison was done only at one air velocity. The results of the comparison are given in Table 2 and Table 3. The results of the tests where the roof covering was attached to expanded polystyrene were of two types of typical fire behaviour:

1) the fire never went through the covering to the substrate (products, 1 – 6). 2) the polystyrene caught fire (products 7 – 9) or

Case 1 above illustrates the problem with using polystyrene as standard substrate. In this case the fire never went through the covering. Instead the big hole in the polystyrene reduced the insulation capacity of the substrate and the damaged length in the roof covering was shorter than when tested on mineral wool. The damaged length when the products were tested on polystyrene was approximately 20 % shorter as an average, in some cases even up to 50 % shorter. For some products in fact a different classification would be received for the two substrates i.e. the longer flame spread on mineral wool would end in a failed classification on that substrate. This was the fact for the product in Table 2 and also for products 1, 3, 4, and 5 in Table 3

The only case when the polystyrene substrate gave the worst result was when the polystyrene caught fire (case 2 above) and the entire test specimen was involved in flames so that the test had to be interrupted (the fire was extinguished). Of course, this case is no good for determination of the fire behaviour on other substrates since it is a fail result.

Table 2 Damaged length of a PVC roof covering tested on mineral wool and expanded polystyrene. Fail results on mineral wool are given as bold figures.

Product no

Length of damaged material in roof covering at air velocity 2 m/s mm

Length of damaged material in roof covering at air velocity 4 m/s mm Mineral wool 150 kg/m3 EPS 20 kg/m3 Mineral wool 150 kg/m3 EPS 20 kg/m3

Mean Mean Mean Mean

1 320 350 305 362 339 710 805 305 410 352 356 11) 290 305 310 302 470 340 565 458 1)

A glass fibre fleece 120 g/m2 was mounted between the product and the EPS. This fleece acts slightly in the same way as the mineral wool. At 4 m/s the damaged length is prolonged compared to the tests where no fleece was added.

(9)

Table 3 Damaged length of bitumen roof coverings tested on mineral wool and expanded polystyrene. Fail results on mineral wool are given as bold figures.

Product no

Length of damaged material in roof covering at air velocity 2 m/s mm

Length of damaged material in roof covering at air velocity 4 m/s mm Mineral wool 150 kg/m3 EPS 20 kg/m3 Mineral wool 150 kg/m3 EPS 20 kg/m3

Mean Mean Mean Mean

1 622 530 525 559 520 503 510 511 NT1) NT1) NT1) NT1) 2 505 420 435 453 321 341 352 338 285 388 363 345 287 319 321 309 3 631 593 699 641 538 405 416 453 589 585 523 566 415 486 414 438 4 NT1) NT1) NT1) NT1) 585 595 585 588 515 560 500 525 5 540 535 500 525 375 385 370 377 550 545 590 562 545 495 455 498 6 485 530 555 523 630 455 450 512 NT1) NT1) NT1) NT1) 7 565 410 475 483 Ex2) Ex2) Ex2) NT1) NT1) NT1) NT1) 8 NT1) NT1) NT1) NT1) 500 410 404 438 Ex2) 9 NT1) NT1) NT1) NT1) 532 570 535 546 Ex2) Ex2) Ex2) 1)

Ex = extinguished due to specimen was completely involved in flames. Thus, no data regarding flame spread recorded.

2)

(10)

4 Conclusions

Classification based on test results from ENV 1187 test method 2 where roof coverings have been applied on expanded polystyrene may in many cases give a direct application for mineral wool on unfounded reasons. It has been shown that roof coverings, which fulfil the criteria of EN 13501-5 when applied on expanded polystyrene, not necessarily do so when applied on mineral wool.

Both the pre-standard for testing ENV 1187 and the classification standard EN 13501-5 should therefore be revised with respect to the listing of standard substrates and rules for direct application of the results. They should not include expanded polystyrene as one of the standard substrates for direct application of test results. Expanded polystyrene should be regarded as a non-standard substrate.

(11)

5 References

1 Produktregler for Brandmæssigt egnede Beklædninger. NKB Produktregler 15. Copenhagen (1990)

2 European Standard – Fire classification of construction products and building elements – Part 5: Classification using data from external fire exposure to roof tests. EN 13501-5:2003 (E). Brussels (2003).

3 European Pre-standard – Test methods for external fire exposure to roofs. ENV 1187:2002 (E). Brussels (2002).

(12)

Appendix

In this Appendix are given the changes in ENV 1187 and EN 13501-5 that are necessary for avoiding that direct application of test results will be based on false facts. Some additional changes are also given.

Changes in ENV 1187

Clause 5.4.2.2 point a): Delete the following text:

or expanded polystyrene (EPS) of density of (20 ± 5) kg/m3 and thickness (50 ± 10) mm.

Clause 5.8.1:Delete the following points:

where required for regulatory purposes or for classification, the damaged area of the roof covering and the substrate, respectively, expressed in square mm;

where required for regulatory purposes or for classification, the maximum depth of the damage on the specimen, expressed in mm.

Changes in EN 13501-5

Clause 6.4.3.2: Delete point b)

expanded polystyrene (EPS) (not fire retardant treated), density (20 ± 5) kg/m³, (50 ± 10) mm thick;

(13)

SP Swedish National Testing and Research Institute Box 857

SE-501 15 BORÅS, SWEDEN

Telephone: + 46 33 16 50 00, Telefax: +46 33 13 55 02 E-mail: info.sp.se, Internet: www.sp.se

technical investigation, measurement, testing and certfi cation, we perform

research and development in close liaison with universities, institutes of technology and international partners.

SP is a EU-notifi ed body and accredited test laboratory. Our headquarters are in Borås, in the west part of Sweden.