That’s funny, Silvaire – I was worried for the exact same thing! I was convinced we were gonna have a split flap scenario and roll uncontrollably into the ground…
How can one explain that such events occur twice as often in twins than singles, unless the probability of a serious accident after an engine failure is higher in a twin?
We’re running in circles here. Although this is ok question to ask and probably one can list many explanations behind the answer, it’s not the right question for getting the answer on single vs twin safety. To get full picture on this comparison one has to calculate in the number of OEI flights which ended without accident. This has been already stated in this thread and even neglected by many, this can’t be ignored.
I had only one engine failure in twin (DA42, lost coolant in left engine due to broken hose clamp) and after gradually reducing power, I made precautionary shutdown to avoid complete engine failure. With 4 occupants on board I continued the flight with feathered engine at OEI ceiling and landed perfectly safe.
So such events simply are not recorded in accidents databases (at least not for GA) thus we can’t get fully compensated and non-biased statistics on twin safety after one engine outage.
Emir wrote:
We’re running in circles here. Although this is ok question to ask and probably one can list many explanations behind the answer, it’s not the right question for getting the answer on single vs twin safety. To get full picture on this comparison one has to calculate in the number of OEI flights which ended without accident. This has been already stated in this thread and even neglected by many, this can’t be ignored.
Sorry. I may be dense, but I really don’t see how that’s relevant. In the end what we mean by “safety” is avoiding injury or death. So how can the rates of anything else than injury and death be relevant? I fully accept that the the mission profiles, risk compensation etc. for SEPs and MEPs are different and so MEPs may be more exposed to “difficult” situations that SEPs.
I’m sorry, but why does everyone dance around the fact that if a single and a twin had the exact same amount of engine failures, regardless of hours, or conditions they fly in, then which one has statistically the higher possibility of making an emergency landing? 100% in the case of the single. Somewhere between 0-100% for the twin. That makes statistically the twin safer, no matter how you slice it.
I think that the assertion that MEP owners don’t maintain their aircraft is a similar on to the one at the head of the thread.
It is as well that I don’t have time or the inclination to dig through old posts on other forums, or access posts on forums I no longer have access to 
Unfortunately the attitude referenced is widespread in some parts of the “classic” piston twin arena, and in some cases the owners happily talk about it.
That makes statistically the twin safer, no matter how you slice it.
I think that definitely holds true for incidents which do not involve loss of control. The issue seems to be that LOC is much more common on a twin, and it almost never happens in cruise; it happens close to the ground. LOC close to the ground is very likely to be fatal, and even if the passengers are OK it takes little to damage the wing tanks and start a fire.
I wish it was as simple as that. Safety is about prevention and consequently you need to identify risk. The fact that a OEI event often results in a safe outcome doesn’t make that a safe event. One element of ‘redundancy’ has been removed, just like a vacuum pump or magneto failure. To be safe you have to recognise trends and events that have started to undermine safety mechanisms/layers.
I made precautionary shutdown to avoid complete engine failure. With 4 occupants on board I continued the flight with feathered engine at OEI ceiling and landed perfectly safe.
One hopes you actually complied with the AFM and “… land(ed) as soon as practicable at the next available/suitable airfield”
AdamFrisch wrote:
one has statistically the higher possibility of making an emergency landing? 100% in the case of the single. Somewhere between 0-100% for the twin. That makes statistically the twin safer, no matter how you slice it.
Not quite. There are even worse scenarios than a (properly conducted) emergency landing. As always – it depends.
Another example? IF you have to do an emergency landing, you’d rather make it in a DA20 than in a DA42……….
AdamFrisch wrote:
I’m sorry, but why does everyone dance around the fact that if a single and a twin had the exact same amount of engine failures, regardless of hours, or conditions they fly in, then which one has statistically the higher possibility of making an emergency landing? 100% in the case of the single. Somewhere between 0-100% for the twin. That makes statistically the twin safer, no matter how you slice it.
But, as we keep saying, the premise is wrong. Twins have more “failures” for a number of reasons, but mainly because
That probably only scratches the surface of reasons.
There is no question in my mind that my failure rate on twins is far greater than on singles. I have very approximately 25/75% on singles/twins and my failure rate is about 8/92%. That doesn’t make statistics, but I don’t believe it to be untypical of the fleet.
Dave_Phillips wrote:
One hopes you actually complied with the AFM and “… land(ed) as soon as practicable at the next available/suitable airfield”
Oh dear. Another religious war?
How can the author of the AFM possibly have done your TDODAR for you sitting in an office 40 years ago?
Just blindly accepting that AFM edict can lead you into all sorts of problems, some safety related, such as runway length (yes, 800m is “suitable” for my type, but surely 1500m ten miles further on makes more sense?), weather (nearest has R800 which is “suitable”, ten miles further on has 3000m vis) and safety facilities (the nearest suitable has a Landrover with a fire extinguisher, 10 miles further on they have a full AFS with three appliances), others practical (nearest suitable has no engineering facilities, ten miles further on is my maintenance base, nearest suitable has no Avgas, ten miles further on does, the nearest suitable has no ground transportation, but ten miles further on is an airport with buses to the railway station.)
Part of our job is to take the decisions, not to blindly follow the AFM.
Dave_Phillips wrote:
We achieved a climb rate of about 30-40ft/min and diverted to another airfield some 25nm away on our departure track having decided to maintain straight ahead until we achieved a sensible altitude.
A very wise decision. 
Airborne_Again wrote:
So how can the rates of anything else than injury and death be relevant?
Rate of survival is also relevant. In SEP engine failure accidents you count walkaways, injuries and deaths. Why wouldn’t you count the same outcomes in case of MEP engine failure accidents? So safe OEI landing has outcome of all occupants walk away without injuries.
Dave_Phillips wrote:
One hopes you actually complied with the AFM and “… land(ed) as soon as practicable at the next available/suitable airfield”
Of course. At that moment I had few options – it was night and closest airport was VFR. So the best feasible ones were turning back to LDRI airport (20 NM on island and behind 5000 ft mountain) and my base airport LDZA (50 NM ahead in the flatland and 2.5 km runway in my heading). The second one seemed to be overall safer choice so I landed at my base.
Timothy wrote:
How can the author of the AFM possibly have done your TDODAR for you sitting in an office 40 years ago?
That’s why the author of these things says nearest practicable airfield, not nearest airfield. It leaves it up to the PIC to determine what’s “practicable”. This may be for example diverting to an airfield 15nm further away with a world class maintenance facility, H24 fire cover, ATC and the rest, over an unattended airfield that’s perfectly long enough but has no emergency services, no ATC and no maintenance facility.
