1 In this appendix, examples of a SOLAS regulation
II-2/17 assessment are presented with the ambition to guide the
Administration in what to require from an assessment involving FRP composite
structures. The following three general principles are used:
-
.1 protect internal structures against exposure to an indoor fire;
-
.2 protect against fires exposing external surfaces; and
-
.3 document performance of fire protection using primarily established test
procedures.
2 A design preview meeting is typically held between the client, coordinator of the
assessment and the Administration prior to the start of the assessment in order to
clarify the scope, objectives, process and roles of stakeholders. Then a design team
is selected to mirror the complexity of the task, in the sense that the members
should together possess all the necessary competencies to perform the assessment of
fire safety. For example, experts in FRP composite materials, fire safety, fire
testing, fire safety engineering, risk assessment, fire safety regulation, ship
design and operation may all be relevant to include. Even if the design team should
be formed at the beginning of the project, it may be necessary to expand it further
along. The whole design team will not be part of all parts of the process but it is
key that the design team is well represented at the hazard identification. It is
also recommendable that the Administration is included in the hazard identification,
as well as at key review meetings, as a witness to gain insight or to provide direct
feedback on preliminary results.
3 An effective hazard identification requires that an investigation of potential
challenges to regulations has been performed (see section 3 (Important factors to
consider when evaluating FRP composite structures with starting point in the
regulations of chapter II-2)) on the basis of a base alternative design,
which must be well-defined at this stage. A base alternative design is the fire
safety design and arrangements which all trial alternative designs have in common,
including the introduced novelty and pre-determined safety measures. Different
combinations of safety measures (risk control options) added to the base alternative
design defines the trial alternative designs to be evaluated (see example
below).
4 As an example, consider the deck house in figure 15 with FRP composite sandwich
panel structures (composition as in figure 1). In the base alternative design the
inside of the FRP composite surfaces are covered by thermal insulation to achieve 60
minutes of fire integrity according to the 2010 FTP
Code, part 11. The fire integrity is maintained at openings and
penetrations.
Figure 15: Example deck house structure of FRP composite
5 Since the structures are not made of non-combustible material, deviations to
prescriptive requirements are found in SOLAS regulations
II-2/9 and II-2/11. Regulation
II-2/9 requires A-class divisions with 0 to 60 minutes fire insulation
capability and regulation II-2/11 requires deckhouses to be constructed of steel or
other equivalent material. By evaluation of the purpose statements, several more
challenges are identified, particularly when considering the unprotected external
surfaces (see section 3 (Important factors to consider when evaluating FRP composite
structures with starting point in the regulations of chapter II-2)). Affected functional requirements may be identified
as:
-
.1 restrict ignitability of combustible materials;
-
.2 restrict the amount of combustible materials;
-
.3 restrict the fire growth potential of combustible materials;
-
.4 limit the quantity of smoke and toxic products released from combustible
materials during fire;
-
.5 boundaries shall provide thermal insulation and integrity with due regard
to the fire risk of adjacent spaces; and
-
.6 materials used in the ships' structure shall ensure that the structural
integrity is not degraded due to fire.
6 During the hazard identification a number of potential ignition sources and fuels
may be identified inside and outside the deck house. With a deterministic
(worst-case) approach, two design fire scenarios are defined to evaluate the fire
safety of the deck house: a flashover fire in the generator space and a significant
exterior hydrocarbon fire. Trial alternative designs are said to be formed by adding
any combination of risk control measures (RCMs), identified as:
-
.1 provision of stiffeners on the inside of the exterior bulkheads (to
provide structural integrity along with the unexposed laminate in case of an
external fire);
-
.2 use of double sandwich panels (triple skin sandwich panels), where only
half are necessary to carry the design load (to structural integrity along
with the unexposed laminate in case of an external fire);
-
.3 provision of a drencher system covering the external surfaces;
-
.4 redundant supply unit for the drencher system;
-
.5 provision of low flame-spread characteristics on external surfaces;
and
-
.6 automatic surveillance of closure of doors.
7 In quantifying effects on safety there are different approaches (see appendix D
(Recommendations regarding the assessment)). Here, an approach where the ambition is
to perform at least as well as a prescriptive design in all areas where fire hazards
are introduced independently is exemplified; this implies that a sufficient safety
margin is sought which by conservative safety measures allows to keep the complexity
of the assessment at a minimum.
8 Regarding ignitability, this is only considered affected at external surfaces. A
fire test is performed in accordance with the standard ISO 11925-2, a test method to
evaluate the ignitability of building products when exposed to a small flame, which
shows that ignitability is not a problem. A full-scale experimental fire test is
performed for different RCMs applied to an FRP composite panel constructed as one of
the external sides of the deck house. They show that RCM e above or pre-activation
of RCM c prevents ignition during 20 minutes of significant fire exposure during the
fire tests. This is argued sufficient with regard to the potential for external fire
exposure and the organization of manual firefighting. Tests according to part 11 of
the 2010
FTP Code are used to demonstrate that fire integrity, fire growth
potential and smoke production are managed in case of a fully developed fire inside
the spaces. Event tree analysis shows that function of door closing devices is key
to prevent an interior fire to grow and spread. It is therefore included in all risk
control options (RCOs), which are now concretized as:
-
.1 RCO A: RCMs a + c + d + f;
-
.2 RCO B: RCMs b + c + d + f;
-
.3 RCO C: RCMs b + c + d + f + extended detection system;
-
.4 RCO D: RCMs c + d + f + extended detection system; and
-
.5 RCO E: RCMs e, f.
9 RCO A and RCO B require that the drencher system is activated if structural
integrity is not to deteriorate and therefore includes a redundant supply unit (RCM
c). Smoke production may not be a problem on deck but it can be argued that fire
growth is not properly managed. This may be handled by an extended detection system,
providing quick and reliable activation of the drencher system, which was therefore
added in RCO C. The question is then if the over-capacity in structural integrity
provided by RCM b is necessary, hence RCO D. Without further elaborating on these
issues, it is decided to present RCO E as the suggested final alternative design,
since it cost-effectively is considered to cost-effectively provide a reliable
solution. The performance criteria to better achieve safety functions where fire
hazards have been introduced are thereby considered to be met.