I just saw in the AAIB monthly list this one, which is close to home:
The weather on the day (which I remember well, because my Dad phoned me to make sure it wasn’t me who had gone in, after hearing on the radio that a light aircraft from the Isle of Man had crashed in the sea, but with no other details) was very marginal in the morning and utterly terrible by the afternoon (bad enough that a bus finished up on its back in the Laxey River, and there was widespread flooding).
There’s of course a lot of emphasis on engine failure while over the sea, but I’m not sure that’s necessarily the greatest danger to a VFR single. Even in reasonable weather the conditions can turn into essentially IMC while legally VFR. While the above flight was in very marginal conditions (since the aircraft was below 3000 feet and 140 knots, VFR minima were 1.5km visibility and clear of clouds, and the lowest ceiling was over 1000 feet), I suspect had this been over the flatlands rather than the sea the flight may have successfully landed. I know a few pilots who “scud run” over the sea – after all, there are far fewer things that you can crash into while doing so – but I get the feeling that pilots who may not have experienced the “milkbowl” effect in poor visibility over the sea may not realise how hard it can be if they have no instrument experience. Even in what would be reasonable VFR over land – for example, not long ago on a flight to Ireland where all weather stations were reporting >9999 and celings over 6000 feet, an easy flight over land – turned into a milkbowl over the sea with the featureless sea blending into a featureless sky: I might as well have been inside a cloud even though it was technically decent VFR. I think VFR pilots need to be prepared for this eventuality if they are contemplating going over the sea – being able to confidently maintain a track and altitude and keep a look-out in these conditions is just as important as having the life raft and jacket, I think.
That report, especially the stuff on the radio, doesn’t make good reading.
I guess this is the “Kennedy” scenario. However the first thing which strikes me reading this is that someone buying a Commander 114
should really get some sort of instrument training, even if nothing legally useful. These ~150kt planes are not easy to fly the way one flies a C152. They need thinking ahead, etc. Even a little bit of instrument training makes you a much better VFR pilot and would enable getting back on the ground.
Also training on the installed avionics. I reckon this pilot (aged 75, FWIW) didn’t even know how to use the autopilot, because with it he could have flown an ILS hands-off all the way to the tarmac. The first thing one does if in deep sh*t is to put the autopilot on (if not on already) and sort things out in peace. However I know many pilots fly “advanced” planes without knowing the systems. I don’t believe this plane had no autopilot and if it was there but wasn’t working then whoever allowed that to continue paid a price for it.
Peter wrote:
I reckon this pilot (aged 75, FWIW) didn’t even know how to use the autopilot, …
He could have been flying on autopilot. After all, he was able to establish himself on the extended centerline from something like 10 miles out in almost zero visibility and tracked it while looking outside for visual clues. All the while he made a contolled descent down to 200ft. Maybe he simply forgot to add power after arresting his descent? Quite a few experienced and IFR rated pilots have made that mistake before him (with things as big as a B777).
Interesting to note that the airfield was technically in VFR conditions. But those 2000 metres of vis, with cloud all around can certainly be very challenging, not only for navigating, but also for sheer control of the aircraft, for a VFR-only pilot.
As often, we’ll never know exactly what happened and why.
All true, but my point was that even if this pilot could not fly and just knew how to engage the AP to fly the ILS, he would have been OK, because a coupled ILS down to 200ft makes a landing dead easy, in vis of 1km-2km.
For a dirty but acceptable job, all you need to do is to watch the IAS and twiddle the throttle, on the way down.
Yes, but for us very current and experienced pilots, it is often all too easy to say things are easy. We’ve largely lost the plot on what is going at the sheer PPL level. Let’s face it: most VFR pilots (not those on euroga 
) can’t even fly properly under VFR conditions, let alone work the avionics on board. So, to say “you merely have fly certain headings and altitudes, then when that needle comes in, push some buttons, then, when the other needle comes in, reduce power to 15 inches and then wait until you see the runway” is just not realistic. At least not for someone who has only had a one-off training on this, say for a few hours, some 17 years ago..
It takes a good skill basis and at least somewhat regular training, i.e. exposure to these things.
Yes, I think so too. There’s many things you have to make right to fly an ILS correctly: Have the right frequency and mode on the autopilot, know the final course, initial altitude, check altitudes … have the chart. Many autopilots cannot fly the glide slope automatically. Configuration of airplane at FAF, gear, flaps, trim for the right speed, stay on the glide, talk on the radio … There’s many ways to mess up an ILS approach.
Also many autopilots in older planes do either not work, or not work right. The repair always costs thousands, so many are never repaired again once they fail. The school in which I did the IR had no airplane at the time with a working autopilot … (“we don’t use it in training”).
PA28-140 from 1975, yes, Commander 114 from 1995?
Basically, you are saying it’s OK to fly into near-IMC without having working automation or without knowing how to use it? Sure it is OK, and many do it, but you have a good chance of getting killed.
Pilots of these (relatively) advanced planes need to understand the systems. I they don’t, they are flying with one arm tied behind their back. I know many disagree… and there is a problem that the logical conclusion of this line of argument is a Type Rating for each type of aircraft – which nobody wants.
Not sure about the non-working automation. From the report:
The aircraft had a continuous and comprehensive maintenance history and its most recent annual inspection was carried out and certi ed on 27 April 2015. Although the aircraft was already fitted with a modern avionics suite with a moving map GPS and autopilot, the pilot had made arrangements for an avionics upgrade to be carried out on the aircraft in Guernsey.
Sounds like the gentleman was keeping the avionics in good nick. Also, seemed to have at least had an interest in instrument flying. Again from the report:
until the following morning when a member of the public reported finding a small shoulder bag containing various items including an instrument flying text book
My emphasis in both quotes.
Talking of automation and VFR-into-IMC, if the pilot in this case knew how to use the undoubtedly very good autopilot in the Cirrus, he and his passengers would have had a much better chance of being alive today.
He could have engaged the autopilot, freeing his mind to navigate, talk to ATC, find better weather conditions etc.
