No idea why the S&L button wasn’t just pressed.
Quoting from the report: “He had not received any training involving the use of the LVL function.”
Obviously, reading the manual is not very popular any more. When I operate some machinery, I am always curious what the various buttons and levers are there for, but we are all different.
Doesn’t the Cirrus have a “CWS” or “TCS” button (control wheel steering or touch control steering, depending on the manufacturer) on the stick (or yoke?) which allows you to briefly disconnect the autopilot servos while preserving the selected modes? Almost every autopilot-equipped aircraft that I have come across has this feature.
On most systems, if you press AP, it will capture and maintain the current pitch and roll attitude.
I don’t know about the DFC90 but on King autopilots the roll angle limits are about 7 and 25 degrees. If you press AP below 7 degrees it will level the wings. If you press AP above 25 degrees, it will refuse to engage (you can test that on the ground, by pressing AP with the vacuum AI toppled off to one side).
Pressing AP will also capture the current pitch, with no attempt to achieve level flight. So even if you are just 2 degrees pitch down, it will capture that. Well, that’s what mine does…
Not knowing this is a really really basic omission but there is no legal requirement for systems training in GA. Anyway, IME, most instructors who are legally capable of a “complex signoff” don’t know any of this stuff either… I was in exactly this situation in 2002 with the brand new TB20 – no systems training, no usable manuals, no instructors who even knew what the HSI did. I downloaded the manuals, read them, and flew around at 5000ft over Kent while playing with the knobs 
until I worked it out. But, from having flown with many pilots with this level of system complexity, most of them are not going to do that, and I think many are not really interested in getting “deep into gadgets” – you need a slightly warped brain to enjoy that. But you do need a warped brain to understand modern (i.e. post 1980s) aircraft systems!
why he elected to exercise his IR privileges
True, though you can get an IR in a C150 with some gauges, and many have done. It doesn’t confer any systems knowledge. I agree he should have been a better basic pilot and should have been able to hold it the right way up in IMC, but that needs currency…
He should have disconnected the autopilot (which was doing stuff he didn’t understand) – as the absolute very first thing. There should be an obvious switch that does that. You should never manually overpower the autopilot and continue to try to fly that way. It should “work” (using the servo clutches) but that is for an all-out emergency only, and autopilot disconnection (immediately via the red button, which should remove power from the servo solenoids, and then as soon as you get around to it, via the autopilot master switch) is the correct procedure.
The alternative to the above is to re-engage the autopilot in the correct mode, or just HDG will do most of the time (you can fly any IAP in HDG if you really want to, and have a moving map), but in this case the pilot was too confused to do that.
Doesn’t the Cirrus have a “CWS” or “TCS” button
I am sure it does (mine does) but then you need to know what exactly the autopilot will do when you release CWS. It can be subtle, and varies with the mode. It can be used to change the captured pitch in PIT mode, the captured roll angle in ROL mode, the captured altitude in ALT mode.
No idea why the S&L button wasn’t just pressed.
I would ask his instructor(s). I bet the AAIB (or anybody else) will not do anything like that. But that is the bit which needs sorting out, with today’s aircraft systems. The alternative would be a formal type rating for every different system (which is basically what they have in jets) but a) do we really want that and b) very few instructors would be able to train it.
It’s a sad report to read, because it will give ammunition to various people who are looking for some, and a very pricey bit of hardware got trashed which will make UK insurers look twice at these propositions. One insurer jacked up the deductible about 5x following UK’s 1st SR22 chute pull; I wrote up the details here before.
I also wonder if Brize’s refusal to allow a descent was in Class G………. they do have a tendency, IME, to control Class G traffic rather a lot. I am honestly not sure if calling them up is the best idea if inbound to Gloucester, because they can result in you being unable to fly the approach (too close and too high).
The alternative would be a formal type rating for every different system (which is basically what they have in jets)
For all the modern avionics fits (like the one in this Cirrus here) the manufacturers supply some kind of computer based training, even for the 20+ year old Garmin GNS units there has been a simulator program available since day one. I know at least one class rating instructor for CitationJets and Mustangs who will only start flying with his trainees once they have completed their self-study systems training. Otherwise he will lose far too much precious flying time explaining buttons and knobs.
I am sure it does (mine does) but then you need to know what exactly the autopilot will do when you release CWS.
But exactly the same applies to disengaging and re-engaging the autopilot as this pilot did! He too obviously did not know how his autopilot would behave after re-engagement… Since many years, the one and only time during a flight when I press the disengage button is when taking over manually on final. All the rest is done via mode-selectors and TCS button. If only to spare the passengers the annoying (and possibly frightening) disconnection sound.
I am sorry but educating yourself on this stuff is straightforward. If you cannot properly deal with autopilot activation and modes then you shouldn’t be using it in IMC at least.
It isn’t about training it is about taking personal responsibility. And then not being able to correct a 30degree bank?
This is not an accident that inspires much sympathy I am afraid.
As an example yesterday I made a very bad mistake while on the ILS GS to Köln. In meaning to change freq of Nav2 I accidentally flipped NAV1. A/P disconnected and I had lost the ILS. It was stupid but there you go. Flipped it back. A/P on in pitch roll mode and by time reacquired hit APPR mode and we were good again and had never really left the profile. It all took about 3 seconds. You must be able to cope with things like this.
and a very pricey bit of hardware got trashed which will make UK insurers look twice at these propositions
I’m sure insurers will look on it more favourably than a similar priced IFR capable plane with no CAPS. I’m sure it’s cheaper for the insurance company to pay up for a wrecked aircraft with live uninjured pilots and passengers rather than pay up for a wrecked aircraft AND up to four dead people.
He should have disconnected the autopilot (which was doing stuff he didn’t understand) – as the absolute very first thing. There should be an obvious switch that does that.
There is. Depending on the model of Cirrus, it’s either a separate button on the top of the side stick or incorporated into the “Chinese hat” trim control which is also on top of the side stick, you disconnect the A/P by pressing down on it.
It’s a sad report to read, because it will give ammunition to various people who are looking for some, and a very pricey bit of hardware got trashed which will make UK insurers look twice at these propositions. One insurer jacked up the deductible about 5x following UK’s 1st SR22 chute pull; I wrote up the details here before.
Contrast that with the attitude of insurers in the US where most will waive the deductible if you pull CAPS on the basis, to quote one of them: “we’d rather keep you as a customer than pay out a fortune to your estate”
I’m sure insurers will look on it more favourably than a similar priced IFR capable plane with no CAPS. I’m sure it’s cheaper for the insurance company to pay up for a wrecked aircraft with live uninjured pilots and passengers rather than pay up for a wrecked aircraft AND up to four dead people.
That is the standard marketing line but I wonder if it is really true, because it sure as hell isn’t reflected in insurance premiums – both here and AFAIK in the USA relative to other IFR / performance types. I phoned up UK’s biggest insurer after one of the discussions here and reported the details, which didn’t support that position at all.
Firstly the insurer will not pay a penny to the estate of a dead pilot – unless he happened to have life insurance with the same company (which is very unlikely).
So, with 3rd party damage so very rare, and with so many people flying these machines solo (unlike renters, not many owners of IFR aircraft need to cost-share to fly), the payout on a totally smashed aircraft will be the same – just the insured value of the hull, plus maybe a bit more for cleaning up the site.
Secondly, most flights which end up on the ground from the enroute phase (but off-airport) do not result in injury or death. Even an engine failure usually leads to an OK-ish forced landing.
Thirdly any proper crash (e.g. a CFIT) will not benefit from the chute. Also if you pull it at Vne plus X% (which an SR22 will reach in not many seconds, in a spiral dive) it will just come off.
So it really depends on how many passengers were in there and what sort of a crash it was.
In the UK, passenger liability requires the establishment of the pilot being negligent. I have no idea how often that happens but obviously it is a very desirable objective for the passengers (or their estates) to work towards.
I really have no idea what the payout distribution is on these things, but the definite feeling I get is that insurers don’t like to think that – to use a bit of a caricature – if a non-BRS pilot gets into trouble he will have a big incentive to sort it out (and the payout might be forced landing damage i.e. ~5k-50k depending on damage) whereas a BRS pilot is just going to pull the chute and land the insurer with a 200k-400k payout (it’s true that some have been repaired).
If you think this caricature is way off then speak to some insurers and try to gauge their attitude. They may be wrong but they are human and they have actuarial data we don’t have.
I am sorry but educating yourself on this stuff is straightforward. If you cannot properly deal with autopilot activation and modes then you shouldn’t be using it in IMC at least.
It isn’t about training it is about taking personal responsibility. And then not being able to correct a 30degree bank?
I am not sure I agree with that. If I take myself for example, OK its only a PA28 that I fly, but it has a working A/P (rare, I know) and it has a GNS430. I have read the manuals for both, and am fairly competent in using them with their day to day functions and practise as far as I can on FSX (I downloaded the full GNS430 trainer product). But no-one has given me formal training on using a GNS430 with approach procedures, and it is only coupled to the A/P with via the HDG and OBS/GPS being aligned, it requires some manual adjustment everytime you make a track change.
But as I found out and described here once, there are in some cases two approach procedures (cat A,B, and cat C,D), and on the GNS430 it is unclear which is which (both have the same menu name). I found this out in the air, but got clarification by later asking a question on here. At the time I had a lookout pilot (instructor) with me but it was not formal training and he didnt know which approach was which either. My point is that you cant just read the manual, or try it on a simulator. It is hard (IME) to find local formal training with accrediation for using such systems, so there are people out there who could find themselves behind the plane, and it’s systems, particuarly in the low to medium ends of GA. I wouldnt fly a SR-22, which is a massive jump up from my PA28 SEP without embarking on a formal course, even if it was the other end of the country, but thats just me, and AFAIK not mandatory either (though I appreciate insurance would be hard to attain without some kind of differences training signoff).
Of course, if you havent had the formal training, or have absolute 100% knowledge and competency in any system complex or not, then you do need to make sure you can, and are prepared to, fly a plane manually. But as humans, I guess we just get flustered sometimes, and it then goes pear shaped in the worst case.
But as I found out and described here once, there are in some cases two approach procedures (cat A,B, and cat C,D), and on the GNS430 it is unclear which is which (both have the same menu name). I found this out in the air, but got clarification by later asking a question on here.
But let’s say you didn’t know, would it affect safety of the flight? Yes systems have quirks. But knowing how to turn it on, off and what the various modes do is pretty key. Sounds like you have done exactly the right thing by learning it. And if it doesn’t do what you expect, turn it off, fly the plane and work out what is wrong.
And I am not saying don’t take training, I think it is critical. But even after training on complex aircraft there are some things you have to work out for yourself. You get out the manual and say, it did X, I expected Y, I wonder why. And you work it out.
Not knowing this is a really really basic omission but there is no legal requirement for systems training in GA.
I think that’s a holdover from the steam gauge era. Having never flown a glass cockpit, for me the flight to Spain with Jason was – as I have written here already – very instructive. Simply put, you don’t fly the airplane and monitor the gauges, but fly the system and monitor the airplane. It’s a training deficiency (at PPL level), that’s slowly being recognized in the US, not sure over here. Personally I would certainly get some proper training before flying a glass cockpit in anything but the most benign VFR conditions.
