Applying a control parameter

Employing a method that allows incremental increases in the intervention time ensures the safe conduct of this type of trial. The most obvious way to do this is to introduce a small time delay between the failure being initiated and the pilot taking recovery action, with the delay time being increased gradually until specification compliance is achieved or an aircraft limit (or test limitation) is approached. In practice, however, it is extremely difficult to use time as the means of controlling the test. This is due to the inevitable inaccuracies that are generated by the neuro-muscular lag inherent in an observer calling actions on a stopwatch and the pilot reacting. It should also be remembered that there might not be a linear relationship between time and aircraft reaction; a small increase in intervention time may have a dramatic effect on the aircraft flight path.

A better and safer approach is to use a controlling parameter which can be presented more easily to the pilot and which is more directly related to the state of the aircraft. Consider a trial to check that a single engine helicopter can successfully establish autorotation following an engine failure with a specified intervention time. In this case the test pilot will primarily be interested in the value and rate of change of rotor speed so this should be chosen as the controlling parameter. The intervention time can then be increased gradually by initiating recovery at incrementally lower values of NR. Likewise for a trial evaluating AFCS pitch lane runaways, aircraft pitch attitude may be the most appropriate controlling parameter. Whichever parameter is used the methodology is the same; small increments are chosen initially with the aim of obtaining trend information. The test team need to establish the relationship between changes to the value of the controlling parameter at the point where the pilot takes recovery action using the flight controls and the value of any parameter which approaches a limiting condition during the recovery. Once a trend has been established informed predictions can be made about the effect of any increases in delay time and the test can progress, usually in reducing increments, to specification compliance or a limit. It is obvious from the above that the control parameter must be presented to the test pilot in such a way that precise values can be seen. This then may require the installation of specific test instrumentation. Living with the failure

As part of the trials planning process it will be necessary to agree with the aircraft operator the requirements that the aircraft will have to meet following a system failure. For example, a failure which leaves an aircraft controllable but very difficult to fly may be acceptable if the aircraft will only be required to recover to base in these circumstances. However, if the requirement is to continue and complete the mission then the same post-failure characteristics may be unacceptable.

Where systems are multiplexed, a failure of one system may have no direct effect on the capabilities of the aircraft, however the loss of redundancy and the possibility of further failures is always considered. This often requires the definition of a post-failure operational flight envelope (OFE) that may be considerably more restricted than the normal OFE. The considerations here are the probability of a subsequent failure and the effect it would have on the aircraft or crew. Loss of the remaining hydraulic system in a duplex installation is an example of a subsequent failure that has such serious implications that the post-failure OFE is usually restricted to landing as soon as it is safe to do so. In many other cases restrictions on speeds and heights or a warning to aircrew on the effects of subsequent failures may suffice. With failures of an engine in multi-engine aircraft the performance available OEI is determined and published for aircrew use.