Is there anything that makes twins more prone to faliure of one engine than singles?
The apparent inability of some twin drivers to change their minds in the face of facts? Ok, ok, I’ll stop !
Could it be your choice of companies you use for the various tasks, Timothy?
The cylinder detachment was most likely over-tightening of the studs. This is easily done by somebody who is clueless. I don’t know about those particular bolts but e.g. the difference between wet torque and dry torque is huge. Or they may have re-used old studs. Regarding the 3rd major rebuild this year on the same engine – what is the story behind that?
During the 17 years I have been on various aviation forums I have been hearing various versions of the same story from twin owners: I have a spare engine so (a) I use the cheapest shop I can find and (b) I run them until they break. One doesn’t need to look all that diligently around the UK’s GA scene to see that is indeed the standard operating mode of the “classic” piston twin. I say “classic” because the DA42 has mostly gone into a different market (FTOs, and owners who wouldn’t be seen dead in say an Aztec, and no doubt that latter one is also largely true the other way round ).
Regarding the probability of an engine failure due to internal mechanical reasons, it is surely obvious that if you have 2 of them then you have 2x more chance of getting one. The only difference is in what happens afterwards. I recall Timothy writing numerous times that the chances are more than 2x higher, due to factors like more vibration, longer control cable runs, etc. But mechanical failures are rare so long as one doesn’t use a complete muppet engine shop.
I believe the MTBF of an IO540 has been calculated at 50000hrs. I would be surprised if this figure is accurate since most overhauls are done in the field (not by Lyco) which in turn is the great strength of these “old” engines and this is much appreciated by their operators.
I’ve just short of 9000hrs.
I’ve already mentioned two ‘failures’, two others were all intentional due to the engine exceeding limitations (oil T & P). Our problem is that we often operate at low levels in extremely high ambient temperatures combined with continuous power changes. Despite using more viscous oils and changing every 25hrs, the oil can go ‘off’ very quickly The TIO540 needs plenty of love and it is often better to shut it down rather than play around with power settings/mixture in such circumstance (during calibration we are rarely more than 12-15nm away from an airfield).
The fifth event happened immediately after maintenance when the engine threw away almost all of it’s oil; but that’s another story.
First, the accident rate both fatal and serious injury per 100,000 hours is higher than SEP, and several magnitudes higher compared to the safety of the C-172, PA28 -161/181 and DA40. Secondly, any accident in an asymmetric condition has 80% plus fatality rate.
Do you have credible source for those figures please ?
: I have a spare engine so (a) I use the cheapest shop I can find and (b) I run them until they break. One doesn’t need to look all that diligently around the UK’s GA scene to see that is indeed the standard operating mode of the “classic” piston twin
Peter – where do you get that from please ?
. My understanding which might be naive is that Regulations dictate the maintenance required – if you dont follow the regs you lose airworthiness cert.. There isnt any major discretion here (or is there) ?
Have provided the links to AOPA and Dick Collins. denota added Mike Busch. These are trusted sources. Also it stands to reason that with the higher kinetic energy, and arguably lower crashworthiness, of an MEP, in either a Vmc loss of control event or asymmetric CFIT/forced landing, the injury potential is higher. Take a couple of minutes to calculate the difference in Kj between a Cessna 182 forced landing and a Baron, the 182 having a higher probability of forced landing in control.
You could drill down the NTSB database using engine failure between types, as the AOPA study did.
My latest rebuild was at a thoroughly respected and respectable shop, and the engine has had to be rebuilt twice more in the first 150 hours of operation. I cannot say why, but it is not for the reasons Peter mentions.
However, Dave makes a very important point. In a twin you take the decision to make a precautionary shutdown much earlier that you would in a single.
For example, I have been in a single over the Scottish Islands when the oil pressure fell to zero, with no increase in oil temperature. I took the decision to fly to the Prestwick overhead (maybe an hour’s flight away) and then make a glide approach, rather than try to land on a hillside. It turned out to be, as I suspected, a gauge error.
In twin, I would probably have shut down that engine as a precaution.
Similarly, both the turbine blade fractures showed only as slight increases in temperatures. The engines kept going, and producing power. But we decided to shut down and divert. In a single we would have kept the engine running.
This is one of the reasons that the statistics are pretty much useless.
An observation, purely incidental, but when I owned an FBO and Maintenance shop, I observed that way too many multi engine aircraft were not maintained as well as single engine aircraft. Too many pilots would postpone maintenance or fly with a condition I would not fly with with a single engine aircraft. I suspect part of this was due to the costs involved and the wrong headed thinking that there are two of most systems including the engine/props and that the redundancy would compensate for poor maintenance. A well trained and current pilot using proper procedures flying a well maintained multi engine aircraft is safer overall in my opinion. But there are way too many poorly maintained light piston aircraft flown by less than current pilots.
Going back to the orange box in the above linked article, at the end are numbers indicating 16% of that study’s engine-failure sample (112 out of 700) were maintenance-induced, most overhaul and assembly/installation. “Repair” much less so. Preventative maintenance (and monitoring) needs to be done to keep an engine fit, but once one starts taking it apart and putting it back together, risk goes way up. Reaching TBO is certainly in itself not a valid reason at all to touch an perfectly running engine, EASA regulations notwithstanding.
A very quick search of the AOPA database searching fatal/engine failure for the Cessna 182 and Beech 58 (not TC or P), shows 22 accidents for the 182 and 4 for the 58, since 1983.
Around 23,000 182 types have been produced, and around 2,200 Beech 58/G58. Not sure what dummy the NTSB uses to estimate hours per year, but assuming an average of 100 per year gives you approximately 0.3/100,000 hours for the 182 and 0.5/100,000 hours for the 58 with respect to engine failure as a cause. Statistically not robust but helps explain the higher fatal rate for the MEP.
In the last five years there have been no fatal accidents linked to engine failure in the BE58, and 4 for the 182 (one linked to an EFATO/Stall/Spin during skydiving, which you would think would have a good outcome in a forced landing). This works out to around 0.4/100,000 hours for the SEP, and 0 for the MEP.
The BE58 is a good example of a personal MEP as it is well supported, still in production and used mainly for personal/business use by pilots with an IR. It would be fair to say that owners of a BE58 tend to be in a demographic where skimping on maintenance is unlikely, not least because they want to protect a hull value between $250k to $1.2million (for recent 2012 and on examples). They are also well built with good systems (this is the normally aspirated, non pressurised variant).
Two of the fatal accidents were related to the pilot feathering the wrong engine, another was a partial failure in a mountainous region, the fourth a classic EFATO loss of control.
My reading of this ‘quick and dirty’ is that the BE58, especially in the G58 variant, in the hands of a qualified and current pilot appears to have the edge over the SEP. Perhaps it is a closer run conclusion if the 182 was restricted to the later 182S and T variants, as there were no fatal linked to engine failure for the variant produced after 1998, and possibly no engine failure accident.