SEIFR Minima

[Fuji Abound]

I don’t think Krallu is concerned with approach minima. I think he is asking us for our opinion on what en route cloud base we would be comfortable with for SE ops.

Clearly there are those who will accept any en route cloud base presumably on the basis the risk of an engine failure is acceptably small. Then there are those who will apply an en route minima on the basis that in the event of an engine failure with their selected minima they feel they have a reasonable chance of setting up for a forced landing. If you consider the second group Krallu may be interested in what minima we feel would give a realistic possibility of setting up for a FL in a reasonably controlled way form cloud break.

[Wrong Stuff]

Originally Posted by bookworm

So, 1/4 of 15% i.e. 4% of fatal IFR accidents is engine failure.

I'm not sure that's the whole story. Bear in mind that these are flight under IFR, including those in CAVOK conditions. If you look at flights with, for example, ceilings less than 500 ft or visibilities less than 1 mile, they might tell a different story.

I recall, while I was flying SE IFR regularly, checking out the NTSB site to find engine-failure-related fatal accidents to prove just how safe SE over low ceilings and poor visibilities is. I stopped after finding an uncomfortable number.

Bookworm, that's a fair point, but I can't quite see where you're getting the doom-and-gloom numbers from.

I've had a quick look at the raw data figures that the NTSB provide on their website for 98, 99 and 2000 (shame they don't provide more, really). The engine failure numbers there look pretty much in line with the rough numbers I quoted from Richard Collins's book.

Of the 70 single-engine IFR fatal accidents, 5 have a first occurrence cause of engine failure for mechanical reasons, 2 for fuel exhaustion and a further one for undetermined reasons. Of the 5, one had a contributing factor of failure to perform an engine overhaul and a second was an inadvertent stall attempting to clear trees on a go-around with a perforated piston, so I'm not sure that should be included either. That leaves us with 3 or 4 fatal accidents with the primary cause of engine failure - which is within spitting distance of the 4% quoted earlier.

It is interesting though that the raw data generally includes the actual weather conditions at the time. Including all the loss of engine power IFR accidents, 8 were in IMC, 2 were unknown and 51 were in VMC. This is almost exactly the same IMC/VMC proportion as the twin engine failure accidents (5 IMC, 25 VMC - fewer accidents, but more often fatal), implying that the single engine aircraft data isn't skewed because they make up a greater proportion in soft IFR than hard IFR. This was also the finding of an analysis Flyer magazine undertook, where they found the proportion of singles remained around one-third of the IFR piston engined aircraft both on good and bad weather days.

So, provided you don't particularly avoid flying in good weather and hunt out the bad, it would appear the original statistics are reasonably accurate. Bear in mind there's a lot of data, so this is only a rough analysis. Also bear in mind that, being a SEP IFR frequent flier, I'm biased and that may be reflected in my analysis. I'd be happy to have any errors pointed out.

[S-Works]

Bah.

My mimima is the same IFR single or twin. if the donkey quits in IMC it does not matter if you have 200ft or a 1000ft you are going in and either you survive or dont......

I do have to say flying to Menorca this weekend IFR in the twin did make me "feel" like I would have longer for a mayday before the crash into the pyanees......

[Wrong Stuff]

Just before deleting the downloaded NTSB data off my hard drive, I just thought I'd run one more quick analysis for my own interest. I noticed yesterday that none of the fatal mechanical engine failures happened in IMC, so I was wondering if you looked only at the IMC forced landings (including both mechanical engine failure and fuel exhaustion) what the outcomes were. Of course, this doesn't tell you the degree of IMC-ness (ie cloud base & visibility) but I thought it might prove interesting anyway.

For the 7 forced landings the worst reported injury percentages came out as:

Fatal - 14%
Serious - 14%
Minor - 29%
None - 43%

Of course, that's only for 7 data points, so treat it with caution.

What's obvious, though, trawling through all that accident data, is that spontaneous engine failure is a relatively unlikely way to kill yourself, and it doesn't seem to matter very much whether you're in a twin or a single. Most of the time people just seem to make a bad decision, fly badly and hit stuff really hard.

[IO540]

To me, the most telling thing is that there are just 7 data points, among the vast amount of GA activity going on in the USA.

Most pilots would die of old age before getting this situation - that's assuming they could fly 24 hours a day for the rest of their lives.

Same applies to a mid-air (except possibly at Stapleford on a sunny Sunday )

[Krallu]

Wow what an interesting thread my question turned into.

Very interesting to see all of your ideas and suggstions.

Well, I think my question was more into the enroute cloudbase for a chance to spot a field before hitting something in case of emergency landing. For the approach you of course use the plate minima if you have to. But the planning minima could be higher if you arent comfortable flying a 200 ft approach or for safety reasons??

So, what do you guys use for enroute planning minima? And/or approach planning minima?

[IO540]

Depends if going VFR or IFR, and whether the VMC-on-top option is available.

It's really easy to get a list of TAFs, and METARs before going, for every flight, along the route, so why not? You get an idea of the cloudbase, and also the forecast should indicate bad weather like TS or TSRA, CB, etc. Every pilot should do this, for a selection of airfields along the route. I do this even for flights at airway MEAs e.g. FL150. Of course one has to look at other sources for enroute weather but that wasn't what you asked about.

This will also give you an idea of the cloudbase en route.

It's rare to fly a route of any length with it being OVC002 all the way along underneath. Obviously one could not do this VFR with a UK PPL - well some people will always try flying under OVC002 It's an IFR or VMC-on-top issue really.

I think that overflying a 30nm area on which fog or very low cloud is forecast is OK. It amounts to an exposure lasting only minutes.

Overflying the Alps if covered in solid cloud is something else. There you have a 1hr or so exposure, with your only escape route being high enough, and having a GPS running a decent topo map so you can glide into a valley. The valley is unlikely to have cloud all the way to the bottom. I don't think I would do a flight like that again though It should be SCT or better, and a satellite pic is a good way to check that.

Approach planning minima is what is says on the plate!