obviously when engine failures are the number 3 reason for fatal accidents that is not the same as “every third”. My mistake.
It is still not impressive that engine failure is such a big factor.
I made no claim that Diesel engines fail less often. We were in a general discussion about the quality of our engines.
Peter wrote:
Flyer59’s assertion that every third fatal GA accident is caused by a failing engineis totally absurd because it means every 3rd forced landing kills somebody, which is obviously wildly wrong.
(Sorry to nitpick … :-) ) Ufff, this is logically incorrect.
Imagine that all fatal accidents are CFIT and forced landings due to failed engines.
Imagine that in a year there are 200 CFIT, 1000 failed engines with 1000 forced landings, of which 1/10 is fatal. Still, 1/3 of fatal accidents is caused by failing engines.
I have no idea if the statistic is wrong as I would have thought far more than 66% die due to pilot error.
Still, 1/3 of fatal accidents is caused by failing engines.
Yes, but I find that statistic weird because an engine failure followed by a fatal accident (IOW a failed forced landing) is a very odd scenario.
Firstly, engine stoppages are very rare. With an MTBF of the order of 50k hrs, the majority of pilots will never see an engine stoppage in their entire flying career.
Secondly, the great majority of forced landings are not fatal.
I have read many many accident reports and hardly any suggest this particular sequence of events.
And if Thielert diesels were 2x more reliable than Lycos (a figure which I dispute as operationally correct anyway, above, especially as so many Thielerts are doing dead easy profiles in the low level training business) that doesn’t mean much for such a rare accident scenario.
Peter wrote:
MTBF of the order of 50k hrs…
Secondly, the great majority of forced landings are not fatal.
Do you have sources/data for this?
I suggest a search here on MTBF.
The forced landing stats are just really very evident.
If it was such an “odd scenario”, how could engine failures be the # 3 reason for fatal accidents?
Or do you think that FAA statistic is wrong?
How is an IFSD a non-reportable non-event if it’s a twin?
By the same logic you should massively increase legacy engine failure rates because an IFSD on a Seneca is “a non-event”.
Peter wrote:
And if Thielert diesels were 2x more reliable than Lycos
Do you have other data to back up your claim that the figures by the BEA are wrong? 
When looking at the data you need to look at the period during which the in-flight shutdowns in the statistical sample occurred, because the fact that the Thielerts at one point did have their very generous share of in-flight shutdowns is not contested!
At the risk of bringing this conversation back to where it started, I do wonder if actually one of the biggest problems is simply the number of constraints you are imposing on the design of a diesel engine such that it could in theory be transplanted into existing airframes. It might be that this is practically impossible.
For example one of the reasons the Tecnam P2006T works is that the Rotax allows for a much smaller engine nacelle than would be possible with similarly powerful conventional engines. Making a water-cooled diesel engine which fits under the cowling of an air cooled Lycoming doesn’t make much sense to me.
A diesel engine, almost as an irrefutable fact, will have to be heavier than a similarly powerful conventional petrol engine due to the higher cylinder pressures. They require larger displacement and / or turbo-charging (vs turbo-normalization) to produce the same power output. However, this is traded off to some extent by the lower fuel requirements for a given mission. This in turn obviously has some impact on the CG range which impacts the placement of the wing and the wheels…
It makes me wonder whether the only way this would work is if someone produced an engine and an aircraft which is able to take advantage of it. I for one like the idea of a deltic engine layout in an aircraft 
LondonMike wrote:
It makes me wonder whether the only way this would work is if someone produced an engine and an aircraft which is able to take advantage of it. I for one like the idea of a deltic engine layout in an aircraft
You have heard about the EcoFlyer? Retrofitted C182s? There are ways of retrofitting. The problem is that it is expensive so the best bet is OEM.
It is not a question about engine qualities or reliability. It is financial and cultural. There is no demand in the U.S. market where diesel traditionally has a bad reputation and AVGAS is cheap and available. Even outside the U.S. a lot of people get bad vibes when diesel is mentioned as airplane propulsion. Nevertheless Diamond and Robin are quite successful with their diesel airplanes.
