Thanks guys.
I know that the standard missed approach gradient is 2.5%.
So, does it means:
1.In case of one engine inop, as long as I can still have a climb gradient > 2.5% during missed approach, I can just follow the published missed approach procedure?
2.In case of one engine inop, and I can not have a climb gradient > 2.5% during missed approach, I need to follow the special engine out procedure instead of the missed approach procedure?
3.In case of one engine inop, and I can not have a climb gradient > 2.5% during missed approach, and there is no special engine out procedure, then I either maintain runway heading and climb to MSA and request radar vector, or I follow a SID which is appropriate?
BlackandWhite2000, in principle, you only ever fly the standard missed approach with all engines operating. Because this is what the procedures are designed for. First freaking paragraph in ICAO Doc 8168!!!
I don't know why people have so much trouble getting their heads around this. It is generally accepted not to fly a SID with an engine out. But most people seem to believe you have to be able to fly a G/A single engine. Some even believe you have to leave passengers or luggage behind, if not able.
Fact is, that you only have to meet the gradient with all engines operating.
So what do you do if an engine fails:
You ask your operator! Just like ICAO wrote:
Note.- Development of contingency procedures is the responsibility of the operator.
If your operator doesn't have an answer for you, the you report them to the authorities, refuse to fly and start looking for work at a serious company.
My operator specified engine out G/A procedures for every runway we fly to. Mostly they are identical to the takeoff engine out contingency procedure. Or they may be identical with the published G/A. In both cases the company from where we get our performance data, has calculated that it is possible to fly the procedure with one engine out.
In no case do you start to make calculations yourself (unless you want to make yourself a criminal). 2.5% may be a no-brainer. But what if the published missed approach has a gradient of e.g. 6% to 7000 feet, and the procedure contains several turns?
You have no chance of make a reasonable calculation yourself. Even from charts in the performance manual. Remember you gradient diminishes in a banked turn. So even if you are able to calculate an average, you still don't know if the mountain peek is just at the place, where your gradient is lower than the required average, because you are in a bank.