E TA

TECHNICAL REPORT

Rules for extended

application of test results
of fire resisting elements

TR 35
Edition July 2009

EUROPEANORGANISATIONFORTECHNICALAPPROVALS
 Resistance to fire

RULES FOR EXTENDED APPLICATION OF TEST RESULTS OF FIRE RESISTING

ELEMENTS
This Technical Report can only be used for products covered by the EOTA ETAG 003 on Internal
Partition Kits and is not suitable for the assessment of products in other product areas.
It refers to the state of the art as laid down when the Progress File for this ETAG was endorsed at EOTA
level in 2006.

Part 1 - General requirements for extended application

1. Scope

This document and the rules contained within its Parts have been developed by the Group of Notified
Bodies - Fire Sector Group in response to a request for the European Fire Regulators Group to facilitate
extended application in advance of the extended application standards being produced by CEN TC127.

Notified Bodies will use these rules to increase the scope of products that can be CE marked above what
is defined in the relevant European Technical Specification or to widen the family of products for which
CE marking is claimed.

The products currently covered in the annexes are:

Part 2 Partitions
Part 3 Glazed screens
Part 4 Penetration seals
Part 5 Pedestrian doors and shutters (timber and steel)

2. Extended application

CEN TC 127 has defined extended application as:

The outcome of a process (involving the application of defined rules that may incorporate calculation
procedures) that predicts, for a variation of a product property and/or its intended end use application(s),
a test result on the basis of one or more test results to the same test standard.

The methodology adopted in the prediction of fire resistance performance is based on universally
accepted rules, some of which incorporate calculation methods and some of which may be agreed
expert opinion based on common technical appreciation between experts. The basis of development of
the of rule varies according to the different philosophies employed for different elements.

Consequently, in order to reduce the likelihood of disagreement between Notified Bodies and to increase
the acceptability of an extended application, any judgement outside the rules or agreed expert opinion
cannot be considered as part of extended application.

3. General requirements

3.1 Supporting information

3.1.1 Primary Information

In order to undertake any extended application, suitable test evidence shall be supplied. Primary test
evidence shall satisfy the following conditions:

It shall have been generated by a body notified under the Construction Products Directive as
capable of testing or certifying such materials/products against the relevant standards.
It shall have been generated from full tests carried out solely to European standards
3.1.2 Secondary information
Secondary evidence e.g. indicative tests to ENs, or full tests to existing national standards may also be
used in support of EXAP in certain cases. The acceptability of using indicative tests is given in the
relevant section and varies according to the different philosophies employed for different elements.
General principles for the use of existing national tests are given in GNB-FSG document N 159. The
default condition is to use only data produced by tests to European standards.

3.2 Permitted changes

3.2.1 General
The changes permitted in 3.2.2 and 3.3.3 are applicable without the need for justification. However,
where the general principle cannot be applied e.g. for a particular product or product property, the
restriction will be specified.

3.2.2 Interpolation
Interpolation between two measured fire performance parameters for a given range of a product property
is permitted for all product properties where the relationship between the two measured fire performance
parameters is linear. For non-linear relationships interpolation is not allowed.

3.2.3 Extrapolation
Extrapolation of any measured fire performance parameter is permissible for all product properties in
either direction where the relationship between the two measured fire performance parameters is linear.
For non-linear relationships extrapolation is not allowed.
3.2.4 Multiple changes
The number of changes that can be made using these rules varies according to the complexity of the rule
and the type of construction being considered. For some constructions and some rules, multiple simple
changes can be undertaken without any detailed consideration. However, for other types of construction
and/or for more onerous changes, then sometimes only one change can be made. If more than one
change is required then a further evaluation will be needed e.g. by a calculation. Some examples taken
from the current annex on partitions are given below.

An example of multiple simple changes that can be carried out without further evaluation would be to
increase the dimensions of a stud (width, thickness, but not length) in a partition up to 4 metres high,
whilst also adding mineral wool insulation. These two changes do not interact to reduce the fire
resistance of a partition and so both can be carried out.

An example of multiple changes that would required further evaluation would be to increase the thickness
of the facings of the partition by up to 50% and to change the shape (cross section) of the stud. The
ability of the modified studs to support the extra weight of the faces would need to be considered and a
calculation method is provided in the annex on partitions for this purpose.

The extent to which each type of construction can have single or multiple changes applied is given in the
appropriate annex for that product.

4. Procedure for undertaking extended application

The following steps must be followed:

Collate the evidence from the various sources outlined in 3.1

Apply the rule(s) from the appropriate annex
Use linear interpolation/extrapolation as appropriate/required
Report the extended application according to the requirements of section 5 below.

5. Contents of the extended application report

 Name and address of the Notified Body that is undertaking the extended application
Name and address of the applicant and details of the extended application request
Date, reference number and the testing laboratory for each test considered
Calculations used in the evaluation process
Drawings and specifications of the product variations approved
Statement to indicate that the extended application has been carried out in accordance with
these rules
Specific rule(s) from the appropriate annex used to confer the extended application
GROUP OF NOTIFIED BODIES - FIRE SECTOR GROUP

RULES FOR EXTENDED APPLICATION OF FIRE RESISTING ELEMENTS

Part 2 Rules for Partitions

1. Scope

This Part contains the extended application rules applicable to partitions of the stud and sheet type
whereby a lightweight metal framework is covered with a layer(s) of board material on each face.
Gypsum plasterboard is most commonly used for this purpose, but this annex is not limited to
plasterboard faced partitions.

Partitions made from complete discrete units that are fixed together are outside the scope of this
document.

Glazed screens consisting of a framework of timber or metal filled with panes of glass are also outside
the scope of this document. These are covered in Part 3.

If the partition being considered includes glazed panels, then the rules for partitions given in this annex
are applicable only to the partition part of the construction. Rules for extending the application of the
glazed part of the construction are given in Part 3.

Reference should be made to the Fire Sector Group document: Rules for Extended Application of
Test Results of Fire Resisting Elements which should be read in conjunction with this annex.

2. Permitted changes

2.1 General
The types of changes permitted given in the Fire Sector Group document: Rules for Extended Application
of Test Results of Fire Resisting Elements are also applicable to partitions.

2.2 Multiple changes

The number of changes that can be made using the rules in this annex vary according to the complexity
of the rule and the number of modifications being considered. The rules given below have been divided
into two types: simple and complex.

For the simple rules (no shading) multiple changes can be undertaken. It is possible to make one, some
or all the changes without any without any detailed consideration or calculation.

For the complex rules (shaded grey) then a calculation method needs to be applied. If more than one
change is required, then the calculation method must be applied to the partition after all the proposed
changes have been made.

Where the required changes include a mixture of simple and complex rules, then the calculation
method must also be applied after all the proposed changes have been made.

Note: A new calculation method has recently been submitted from CTICM. TG-5 needs to decide to use
this new method, the existing method or allow both to be used. The method is not yet included in this
document.
3. Rules

ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/

CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

CHANGES IN COMPONENTS

Increase in size Full EN test of partition Rule The dimensions of the section of a stud (vertical Increase in size allowed automatically
of stud section with stud on which member) may be increased without limitation
EXAP is based

Change in shape Full EN test of partition Rule incorporating The shape of the section of a stud (vertical Change in shape or size shall neither result
of stud section with stud on which calculation method member) may be changed provided that the new in any decrease in bending stiffness nor in
EXAP is based section is shown by calculation method to be as any increase in thermal induced bowing.
including extra strong as the stud tested Calculation method is applied at elevated
temperature temperatures.
measurements as
required by the
calculation method

Increase in Full EN test of partition Rule The thickness of the facings of a partition with Consideration may be needed on ability of
thickness of with board on which non-metallic faces may be increased by up to partition framework to support extra load
board/facings EXAP is based 50% provided the length of the fixings for the of thicker facings hence limit of 50%
facings (boards) are increased pro-rata. Maximum thickness increase and 3m height. Metallic
height of partition restricted to 3m. Metallic facings excluded due to increased
facings excluded. propensity to induce bowing.

Existing national test data may be used to justify Existing national test data must
increases in excess of those given above and demonstrate that changes have no
metallic faces detrimental effect.
 ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/
CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

CHANGES IN COMPONENTS

Inclusion of Full EN test on Rule Partitions tested without mineral wool insulation Height limited to height tested (normally
mineral wool partition without infill may have mineral wool added without 3m)
insulation (up to insulation restriction.
4m height)

Inclusion of Full EN test on Rule incorporating Partitions tested without mineral wool insulation Calculation shall address stiffness, self
mineral wool partition without calculation method infill may have mineral wool added provided they weight and bowing of partition at elevated
insulation (>4m insulation (to be agreed) satisfy the requirements of the calculation method temperatures
height) given at the end of this document

Inclusion of Full EN test on Rule Existing national test data may be used to justify Existing national test data must
mineral wool partition without mineral wool infill for heights in excess of 4m up demonstrate that inclusion of mineral wool
insulation (>4m insulation to height tested. has no detrimental effect.
height)

INCREASE IN HEIGHT

Increase in height Full EN test of partition Rule incorporating Partitions may have their height increased above Calculation shall address:
(above 4m) system calculation method 4m provided they satisfy the requirements of the stiffness, self weight & bowing of
(to be agreed) calculation method given at the end of this partition at elevated temperature
document possible increased bowing due to
restraint of thermal elongation

Increase in height Full EN test of partition Rule Existing national test data may be used to justify Existing national test data must
(above 4m) system heights increases in excess of 4m up to height demonstrate that increase in height over
tested. 4m has no detrimental effect.
Calculation method for evaluating partitions with increases
in height above 4,000 mm and components

The methodology given below allows the maximum height of a stud and sheet partition system to be
established based on data taken from a test fully in accordance with EN 1364-1:1999. The test data
required must be taken from a test, which falls within the direct field of application of the proposed
construction. Before commencing the exercise the assessing authority should ensure that the partition
meets the direct field of application lateral deflection requirements during a 3000mm high test. Data taken
from a test conducted below 3000mm is not allowed.

Step 1

Calculate weight of wall per square metre i.e. kg/m2 x 9,81, and select a test height L mm.

It is essential to start with a low height such as 4000mm and work upwards in 1000mm increments. Once
the partition becomes unstable reduce the increments to 100mm until the critical height is found for the
duration in question.

Step 2

Obtain stud thickness t mm, stud spacing m mm, stud depth d mm, flange width fw mm (Note 2). The yield
stress should be taken as the value appropriate to the quality of steel being used e.g. 210, 280 or 350
N/mm2

The Youngs modulus E is assumed to be 205000 N/mm2.

Step 2

Obtain hot flange temperature Th in deg C and cold flange temperature Tc at the time at which the stability
is to be checked. (Note 3)

Step 3

Calculate the following terms:

kh = 1 - Th /800
kc = 1 -Tc /800
k = kc - k h

Step 4

Calculate the neutral axis yn from the cold end:

yn = fw .( kh . d + kc . t/2) + d2 .(k/6 + kh/2)

( (kh + kc ) . (fw+ d/2))

Step 5

Calculate e = d - yn

and calculate the second moment of area of the stud Ih at elevated temperature:

Ih = t.( fw . (kh . e2 + kc . yn2 ) + e3 .(kh /3 + k . e/(12.d)) + yn3 .(kc /3 - k . yn/(12.d)))

Step 6

Calculate the Euler height under the above conditions:

Le = ((2. 2 . E . Ih . 106 )/( . m))0.3333

Step 7

Check that height L is a lower value than the Euler height Le.
If it is not lower then the partition will be unstable regardless of any thermal bow. Otherwise proceed to
step 8.

Step 8

Calculate the moment capacity Ms of the stud at elevated temperature:

Ms = t. (fw .( kh . e + kc. yn )+ e2 .(kh /2 + k . e/(6.d))+ yn2 .(kc /2 - k . yn /(6.d))).

Step 9

Calculate the moment capacity Mc of the stud frame per metre width:

Mc = Ms .1000/m

Step 10

Calculate the load Pe:

Pe = 2. E.Ih.1000/(L2 . m)

Step 11

Calculate the thermal bow b:

b = ((14x10-6 ) . L2 . (Th-Tc))/(8.d)

Step 12

Calculate the total weight P per metre width:

P = . L/1000
Step 13

Calculate the additional bow be due to self weight from

b e = b /(2.Pe/P -1)

Step 14

calculate the term:

= 2 . tan-1 (2 . (b + be )/ L ) ( in radians)

Step 15

Calculate the moment M produced by the self weight acting eccentrically:

M= (. L2 . . (1 - cos ) ) / (4 .1000. sin2 )

Step 16

Now compare the moment M with the moment capacity Mc.

If M is greater than Mc then the wall will be unstable and collapse.

If the wall is stable return to step 1 and increment the test height repeat this process until the maximum
fire rated height is established to the nearest 100mm.

Note 1: Wall comprising metal C or I-studs lined each side with plasterboard. Plasterboard assumed to
provide no extra stiffness to wall but its full weight to act downwards at all times. Wall assumed to be free
at each side edge and with unrestrained head.
Note 2: Where the flange has a turn down this is included in fw and where an I-stud has a double
thickness due to folding, this is included in fw.
Note 3: Temperature information can be obtained from a normal fire resistance test and should be
available at every minute during the test. The calculation will need to be done at all time increments in
case conditions exist whereby stability failure occurs. A number of studs will need thermocouples
attached the data should be used from the most onerous stud position it is not correct to average this
data. In the calculation, hot flange temperature Th must be assigned a value of no greater than 800 deg C
since metal strength is effectively zero above this value. If information is available on metal temperatures
then this model assumes a linear temperature distribution across the web and the data must be
persuaded to fit the model.

Design Aspects of Fire Rated Partitions within Buildings

There are various design aspects which must be considered such as the total movement due to the bow
and the movement at the head of the partition.

Dealing with the total bow first, this can be calculated by adding the result from Steps 11 and 13 i.e. the
thermal bow and the self-weight induced bow.

The extent of the head movement depends on the net result of the vertical upward expansion due to
temperature and the drop caused by the thermal bow.
The thermal expansion coefficient of steel is approximately 14 x 10-6 (between 100 and 700 degC) and
therefore the upwards expansion at the head yu mm of the partition is:
 yu = ( 14x10-6 ) . L . Ts

Ts = stud average temperature in deg C (above ambient)

If the cladding material does not protect the steel studs adequately, then quite large expansion at the
head is possible and clearly therefore, it is best not to allow the metal to get too hot!
This upward deflection however could be reduced by the effect of thermal bowing (and self weight) of the
partition and when a large bow is expected then the head detail might need to accommodate the partition
dropping at the head and yet still be adequately restrained laterally.

The head drop yd mm caused by the bow is:

yd = L . ( - sin ) / sin

where = 2 . tan-1 ((2 . bt )/ L ) ( in radians)

and bt = the total bow in mm (addition of Steps 11 and 13)

The resulting net movement at the head is yu - yd.

The assessment must evaluate the extent of the head drop if any so that the partition designer can adapt
the partition design to accommodate the anticipated head drop.

As with the maximum height calculation the head drop is required to be calculated for every 1 minute of
the test. It is likely that the maximum head drop will occur prior to the duration of interest.

Calculation Check

To aid assessors who are preparing spreadsheets to conduct the calculations the following verification
data has been provided for the partition height calculation.

Input data:

350
400
L 8000
t 0,5
m 600
d 146
fw 39
Th 700
Tc 200

Output data:

kh 0,125
kc 0,75
k 0,625
yn 43,589
e 102,109
Ih 105057,513
Le 12088,103
Ms 673658,153
Mc 1122763,588
Pe 5535,404
b 383,562
P 3200
be 155,943
0,268
M 873638,358

Wall is stable since Mc >M

GROUP OF NOTIFIED BODIES - FIRE SECTOR GROUP

RULES FOR EXTENDED APPLICATION OF FIRE RESISTING ELEMENTS

Part 3 Rules for glazed screens

1. Scope

This Part contains the extended application rules applicable to glazed screens consisting of a framework
of timber or metal filled with panes of glass.

Where a glazed screen is part of a partition, then the rules given in this annex are applicable only to the
glazed part of the construction. Rules for extending the application of the partition part of the construction
are given in Part 2.

Reference should be made to the Fire Sector Group document: Rules for Extended Application of
Test Results of Fire Resisting Elements which should be read in conjunction with this Part.

2. Permitted changes

2.1 General
The types of changes permitted given in the Fire Sector Group document: Rules for Extended Application
of Test Results of Fire Resisting Elements are also applicable to glazed screens.

2.2 Multiple changes

The number of changes that can be made using the rules in this annex vary according to the complexity
of the rule and the number of modifications being considered.

For glazed screens, the rules are all simple and do not require further evaluation e.g. by calculation.
Consequently, multiple changes can be undertaken in one extended application.

However, for the rules concerned with aspect ratio and pane size, the effect of applying them together is
too onerous and consequently this needs to be limited. The limitation is given in 2.3 and to caution
Notified Bodies, the relevant rules in clause 3 subject to the limitation are shaded in grey.

2.3 Limitation on aspect ratio v. pane size

For < 30 minute applications, the rules for pane size variations can be applied in addition to the rules for
adverse changes in aspect ratio.

For > 30 minute applications, either the rules for pane size increases can be applied or the rules for
adverse changes in aspect ratio can be applied.
3. Rules

ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/

CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

ASPECT RATIO

(< 30 Mins Full EN test of system Rule The aspect ratio may be varied without restriction
insulating and on which EXAP is
non-insulating based
glasses)

(> 30 Mins non- Full EN test of system Rule The aspect ratio may be varied without restriction
insulating on which EXAP is provided that the limitations for sizes of panes
glasses) based and the longest dimension of panes are not
exceeded

Existing national test data may be used to justify Existing national test data must
increasing the longest dimension above that demonstrate that increasing the longest
tested (but retaining the pane area limitation) dimension above that tested has no
detrimental effect.

(>30 Mins Full EN test of system Rule The aspect ratio may be varied without restriction
insulating on which EXAP is provided that the limitations for sizes of panes are
glasses) based not exceeded
 ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/
CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

PANE SIZE

Timber Full EN test of system Rule The areas of panes may be increased by up to There must be test evidence that
frames on which EXAP is 25% according to the time of overrun e.g. 10% demonstrates ability of transoms to support
insulating & based overrun allows 10% increase in pane size. extra weight of larger panes
non-insulating (Overrun is length of time over the achieved
classification period). The test evidence that demonstrates ability
of transoms to support extra weight of
larger panes may be EN or existing
national test data.

Steel frames Full EN test of system Rule The areas of panes may be increased by up to As timber frames.
insulating & on which EXAP is 25% according to the time of overrun e.g. 10%
non-insulating based overrun allows 10% increase in pane size.
(Overrun is length of time over the achieved
classification period).

Note: For < 30 minute applications, the rules for pane size variations can be applied in addition to the rules for adverse changes in aspect ratio. For > 30 minute
applications, either the rules for pane size increases can be applied or the rules for adverse changes in aspect ratio can be applied, but not both.

ORIENTATION

Angle of Full EN test of system Rule The result of a test may be applied to an identical
installation on which EXAP is system oriented at + 15o from angle tested
based provided that there was 15% overrun of integrity
(all criteria) in the test Existing national test data must
Existing national test data may be used to justify demonstrate that increasing the angle
increasing the angle from that tested under the above 15o has no detrimental effect.
EN
 ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/
CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

ASYMMETRY

Validity of result Full EN test of system Rule The result of a test on an asymmetrical glazed Worst side equals:
on both sides when tested from the screen tested from one side only is applicable to glazing beads on fire side
with test from worst side both sides if it was tested from worst side double glazed units with fire glass on
one side only exposed face

NB glasses designed to provide greater fire

resistance from one side can only be used
in that side

Existing national test data may be used to

determine the worst side if not covered
above.

DIMENSIONS OF SCREEN

Increase of Full EN test of system Rule Height may be increased pro-rata with percentage Original EN screen must be tested at 3m
height time of overrun e.g. 10% overrun allows 10% high.
increase in height
Increases in screen height may be
accommodated by increases in pane sizes.
However, these shall be limited by the
rules for pane size and aspect ratio.
Extension of Full EN test of system Rule Unlimited extension of width is permitted Only permissible provided one vertical
width edge tested unfixed and unrestrained

ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/

CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

NUMBER OF PANES

Increase in Full EN test of system Rule The number of panes may be increased. The test specimen shall incorporate shared
number of panes mullions/transoms as appropriate.

Decrease in Full EN test of system Rule The number of panes may be decreased. The pane area shall not exceed maximum
number of panes allowable under these rules. The aspect
ratio shall not be more onerous than that
allowed under these rules.

EDGE COVER

Change in edge Full EN test of system Rule Edge cover may be increased for all glasses TG-5 to decide on what constituted edge
cover except edge sensitive types. sensitive e.g. modified toughened soda-
Edge cover can never be decreased lime glasses.

Existing national test data may be used to

determine which glasses are edge
sensitive.
 ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/
CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

REBATE DETAILS

Variations in Full EN test of system Rule Centres of fixings may be decreased, but not
glazing bead increased

Pinned V. screw Full EN test of system Rule A test on glazing beads that are pinned (nailed) in converse does not apply
fixed timber with pinned on glazing position covers an end-use where the glazing angle of fixings cannot change by
glazing beads beads bead are screwed in position. more than 10o
extent of fixing crossover cannot be
reduced

Existing national test data may be used to

justify >10o change in angle.

Stud located V. Full EN test of system Rule A test on steel glazing beads that are stud located Permitted provided there is no:
screw fixed steel with stud located (clip-on type) covers an end-use where the weakening of retention
glazing beads glazing beads glazing beads are screwed in position. decrease of any shielding effect
increased exposure to radiant heat

FRAMING/SUPPORTING CONSTRUCTION

Changes to Full EN test of system Rule The result is applicable to any supporting The maximum fixing centres shall not be
supporting construction of at least the same fire resistance as exceeded and the minimum number of
construction/fixi that tested fixings cannot be reduced
ng of frames The fixings used to fix the glazed screen may be
varied
 ITEM TO BE SUPPORTING TYPE OF EXAP RULE OR CALCULATION METHOD LIMITATIONS/
CHANGED EVIDENCE TO BE TO PERFORM COMMENTS
REQUIRED UNDERTAKEN EXTENDED APPLICATION

GLAZING MATERIALS

Glazing Full EN test of system Rule Substitution of glazing materials (including the Separate tests required for each type
materials with plus test evidence of glass) is permitted providing the limitations given of non-insulating glass
different glasses new glass and/or in the limitations/comments section adjacent are
glazing material to be adhered to. Tests on monolithic wired glass cover
incorporated toughened borosilicate glass but not
vice-versa

Tests on modified toughened glasses

validate use with toughened
borosilicate and ceramic glass but not
vice-versa

Insulated glasses are limited to the

thickness tested if there is only one
test

Linear interpolation is permitted

between different thicknesses of
insulating glasses of the same type

separate tests will be needed to

include doors