One engine out procedure
 

dear collegues , if I were in an airport and I didn't have the one engine out procedure available, what are the construction criteria that a pilot must consider by him/herself? that is to say ? if I take the OACI type A obstacle chart for that airport , what margins should I consider for obstacles to be safe and legal during a IMC takeoff .?????

your obligation is to ensure your flight path meets the obstacle clearance requirements which are broadly 35' at the DER + 48'/nm thereafter, with a 200'/nm slope (3.3%). That is very simplified and is found in both ICAO PANS OPS Volume I Flight Procedures (ICAO Doc 8168 Vol I 3.1.4) PANS OPS Volume II Construction of Visual and Instrument Flight Procedures, TERPS, and the FAR AIM (see (5.2.9 (e)(1). ). The operator is responsible for contingency procedures. 8168 1.2 states:

For a SID if published, if the 3.3% slope is infringed, a gradient requirement is included as a requirement. The SID is for normal operations, and does not itself give any guarantee of contingency suitability, that is up to the operator who is presumed to have the training and data to confirm that it is acceptable or another alternative needs to be applied.

One simple catch all solution where there is no obvious geographic solution is to remain VMC within the circling area until reaching an MSA or a definite ability to meet a spot height distance down track. This is not necessarily efficient, but it is very safe as an alternative. For all engines, this would essential be met by a climb overhead in a race track within the circling area. As PIC, or the operator, it is up to you to have a supportable justification for what you flew. The departure is the main point where this issue arises, for enroute and arrivals it is far clearer. For G/A's there are some occasions where an OEI gradient will result in a higher DA/H, and that is up to the operator to determine what the capability of the aircraft is on the day. Flying the reverse path of an approach, provides some information as to terrain clearance, but if you cannot meet a crossing then it will get untidy, you would be obliged to do something about that and deviation from the approach path would place you outside of known areas of protection fairly quickly.

A straight out departure can be calculated for required gradients, some airports will provide such surveys. Others need some digging around to use official surveyed data to achieve a solution. [Did this once for both departures at VQPR/PBH/Paro Intl in the hills of Bhutan. It took a lot of time to determine a manner to achieve a departure, and to set limitations for weather conditions. It was so complex for the aircraft involved, it was easier in the end to decline the task. On a good day it was fun, on a bad day, you would be tracking a winding path for around 55nm to achieve a safe height for normal tracking. As the visibility down the valleys could not always be guaranteed then much of that tracking would be based on GPS positioning with GPSS and better than RNP0.3, the terrain was continuously just outside of the max obstacle area width, and that is barely better than 0.3nm ]

PANS OPS are available with some searching from the internet, or from your management, they should have a copy. Vol II part 1 Section 2 Ch 1.3 covers obstacle clearance and thereafter gives examples.

Second segment gradient charts are fine for a simple close in obstacle, but are predicated on T/O thrust, so after the time constraint you would be looking at other options for your analysis.

Why not just take a cup or two, or more, of coffee, sit down and run the AFM analysis ?

Unless you have a fairly critical situation matching a given Type to a runway, the work is not difficult, although often tedious. In the event that you need to run curved departures to keep away from awkward terrain it does get a tad more complex but many airports are fine with a straight ahead departure.

I am never all that happy with endeavouring to fit a visual departure for an OEI takeoff with much tiger terrain as the gradients are far too shallow for any sort of eyeball reliability.

tks FDR..Good clear answer