Appendix 2 — Test specification rationale

ED Decision 2018/007/R

(a)      Introduction

The overall risk of capsizing within the 5-minute exposure period consists of two components: the probability of capsizing in a given wave condition, and the probability of experiencing that wave condition in an emergency landing on water.

If it is assumed that an emergency landing on water occurs at random and is not linked with weather conditions, the overall risk of a capsizing can be established by combining two pieces of information:

(1)      The wave climate scatter table, which shows the probability of meeting any particular combination of Hs and Tz. An example scatter table is shown below in Figure 1 — Example of all-year wave scatter table. Each cell of the table contains the probability of experiencing a wave condition with Hs and Tz in the range provided. Thus, the total of all cells in the table adds up to unity.

(2)      The probability of a capsizing in a 5-minute exposure for each of these height/period combinations. This probability of capsizing is different for each helicopter design and for each wave height/period combination, and is to be established through scale model testing using the method defined above.

In theory, a model test for the rotorcraft design should be performed in the full range of wave height/period combinations covering all the cells in the scatter table. Clearly, wave height/period combinations with zero or very low probabilities of occurrence might be ignored. It might also be justifiably assumed that the probability of capsizing at very high wave heights is unity, and at very low wave heights, it is zero. However, there would still remain a very large number of intermediate wave height/period combinations that would need to be investigated in model tests, and it is considered that such a test programme would be too lengthy and costly to be practicable.

The objective here is therefore to establish a justifiable method of estimating the overall 5-minute capsize probability using model test results for a single-wave condition. That is a single combination of Hs and Tz. Such a method can never be rigorously linked with the safety objective, but it is proposed that it may be regarded as a conservative approximation.

(b)     Test methodology

The proposed test methodology is as follows:

The rotorcraft designer selects a desired significant wave height limit Hs(limit) for ditching or the emergency flotation certification of his helicopter. Model tests are performed in the sea condition Hs(limit) Tz(limit) (where Tz(limit) is the zero-crossing period most likely to accompany Hs(limit)) with the selected spectrum shape using the method specified above, and the 5-minute probability of capsizing (Pcapsize) established in this sea condition.

The way in which Pcapsize varies for other values of Hs and Tz is not known because it is not proposed to perform model tests in all the other possible combinations. Furthermore, there is no theoretical method to translate a probability of capsizing from one sea condition to another.

However, it is known that the probability of capsizing is related to the exposure to breaking waves of sufficient height, and that this is in turn linked with wave steepness. Hence:

(1)     the probability of capsizing is likely to be higher for wave heights just less than H_s(limit) but with wave periods shorter than T_z(limit); and

(2)      the probability of capsizing will be lower for the larger population of wave conditions with wave heights less than H_s(limit) and with wave periods longer than T_z(limit).

So, a reasonable and conservative assumption is that on average, the same P_capsize holds good for all wave conditions with heights less than or equal to H_s(limit).

A further conservative assumption is that P_capsize is unity for all wave heights greater than H_s(limit).

Using these assumptions, a comparison of the measured P_capsize in H_s(limit) T_z(limit) against the target probability of capsizing (P_{capsize(target)}) can be performed.

In jurisdictions where flying is not permitted when the wave height is above H_s(limit), the rotorcraft will have passed the certification criteria provided that P_capsize ≤ P_{capsize(target)}.

In jurisdictions where flying over waves greater than Hs(limit) is permitted, the rotorcraft will have passed the certification criteria provided that: Pcapsize ≤ Pcapsize(target) – Pe, where Pe is the probability of exceedance of Hs(limit). Clearly, in this case, it can be seen that it would not be permissible for the rotorcraft designer to select an Hs(limit) which has a probability of exceedance greater than Pcapsize(target).

Figure 1 — Example of all-year wave scatter table

[Amdt No: 27/5]