Peter wrote:
That’s an excellent point, highly relevant to the UK “maintenance” scene where incompetence is rife, often with the tacit agreement of the owner.
I think this view of UK maintenance is rubbish. There is absolutely no evidence that UK maintenance is any worse than anywhere else in Europe or the world. You say it a lot. Please provide evidence more than your personal single data point.
In speaking with people I know who have done European maintenance, there are just as many probs in Germany.
I get the technical fixes by changing the approach aids, but the human factors bit, I think, needs more looking at.
If he were flying what he perceived to be a difficult approach, one he hadn’t flown before, or in a “hard” plane, he’d be more alert, using his “hard disk” based memory and skills to fly the thing. He’d be alert, working very hard, and aware of every little thing (think about when you learnt to fly, everything is v real, and you are very focused)He’d be tired afterwards.
Problem was, he was the opposite with respect to the plane, the airport, and the approach. So his “working ram” was used instead, and one feature of this is its ability to predict (guess, if you like) what is going to happen next- because it is more efficient that way. Now these cognitive guesses are not necessarily under conscious control, but they have worked multiple times in the past and are really very hard wired- things like knowing that rocks are hard, because they have always been so in the past- a very reasonable guess, but potentially wrong in certain circumstances. His cognitive plan didn’t pick up the NDB/airport location mismatch on this occasion, basically because he may have been on the psychological equivalent of autopilot and on this occasion this lead to an incorrect course of action (turn too early, descend too early, whatever)
Im sure there may well have been a horrible few moments (infact, I hope not) where the cognitive guess was proved wrong, and hard disk replaced ram. But by that point it was too late (a less experienced pilot may not have had the hard disk stores in the first place- thats what experience is)
It wasn’t his fault, he was just exhibiting normal human traits, novel, new things need more processing than mundane things, but are greedy on resources and energy, so only get used when learning something new, or consciously “thinking”
I guess the questions to ask, are why this didn’t happen everytime someone flew this particular approach-if it were indeed down to the wrong naiads, there would be many instances of this particular accident chain -I think there are a handful.
If we change the approach then we will certainly mitigate the chance of this happening again, but we will have failed to learn the underlying issues- and that of course is a big vicious circle, because as sure as day follows night, at some point in the future, there will be another accident, where someone following a GPS approach gets it wrong, and crashes. What do we do then- design another approach which is even more intuitive – because that leads to more ram based processing, and less alertness.
Or, as AirbourneAgain concludes, this is human factors. Lets talk about approach aids.
Human factor contributors to an accident take on many forms. Because they involve humans, we are less willing to openly discuss the shortcomings of anyone’s applied abilities in a situation, than we are of technical failures. As a civilised species, respect and dignity to each other are usually always shared. This respect closes the door on freely exposing what may be additional contributor(s) to the cause(s). We can damn an NDB and a Garmin 530 without the risk of inadvertently offending. That is therefore as much as I can offer over my knowledge of the subject accident of this thread.
However, my own experiences of Human Factors in light aviation is something I am more than willing to share. In the airline industry, we explore and coach the non-technical skills of pilots including Leadership, Situational Awareness, Decision Making and Communication. The pilot has less resource in a single crew environment, thus it demands additional care and less haste. This is something I am starkly aware of when chopping and changing between two-crew commercial and single-crew private flying. The distractors in the latter are arguably higher; with no recognised sterile cockpit philosophy, often a passenger’s contribution can be more of a hindrance than a help. In a single crew environment, the risk is higher of pressing on when not safe to, whereas in multi crew the pilots are theoretically trained to the same standard of situational awareness where one pilot should intervene before the situation becomes dangerous. Turbulence, lack of familiar automation, avionics and displays, and in particular, recency on type, are other human factors that serve to risk undermining the capacity of a single crew pilot. Now in my 25th year of flying, I have deliberately slowed down my single pilot operation to a pace of which some have dismissed as lethargic; but this is because I am hugely conscious of the limits of my capacity.
By way of explanation, during my PPL days I was receiving radar vectors to an ILS in Norway in an E55 Baron, having crossed the north sea using a Trimble IFR GPS superimposed on the HSI. It was only when the controller informed me I was not tracking the localiser that I suddenly realised I had not flipped the CDI switch from GPS to NAV. I remember fortunately becoming VMC shortly after and having to dive when I saw the runway, as I had remained at platform altitude in IMC since the GS bar had not appeared on the HSI. I recall that heart-sinking feeling of overwhelming confusion to this day even though it was 20 years ago. Lessons are never more effectively (but reluctantly) learned than when one gets a fright; this taught me to never, ever use the HSI to display GPS information again. I simply haven’t trusted my own capacity to deal with this function since.
Just a quick point on SOPs, which were mentioned earlier in the thread.
The SOP that most people would adopt is to load the procedure in the GNS/GTN, which provides a GPS track “for monitoring” the NDB indications on the outbound track and while turning base.
The SOP adopted by the accident pilot, apparently, was to use OBS mode and set the outbound and inbound track manually on the HSI.
If only used for monitoring (which, alas, it presumably wasn’t) this handmade SOP would give almost exactly the same HSI guidance as the procedure SOP.
However, I can think of 2 critiques of the second SOP:
Both these points reduce the safety of the handmade SOP, hence the procedure SOP is clearly preferable.
The point I want to make is that the correct SOP for transitioning from GPS to VLOC navigation during a procedure is not part of the IR exam, at least not here in the UK. The exam here is based on using only the ADF for navigating beacon outbound on an ILS procedure – you are not allowed to load the procedure on the GPS, even for monitoring.
The result is that a key base-turn check (select VLOC) is barely meaningful in the training/exam context, because you always selected VLOC about 20 miles from the airfield so you couldn’t be accused by the examiner of cheating.
A couple of weeks ago I flew my first NDB approach with a GPS overlay, simply because I realised it was something I had never done. Surely that can’t be right in today’s training environment? Surely it’s time to introduce training in, and testing of, the use of GPS in non-GPS approaches, and real-world use of GPS overlays to monitor a conventional-navaid approach?
Experience has shown me that complete reliance on the GPS data is a recipe for disaster. I’ve seen incorrect databases, signal dropouts, interference/jamming (it is not a myth) and various other failures. I’ve even seen two GPS units disagree.
Dave, please can you elaborate. What you don’t say above is whether you received annunciations about these errors. If you did, then I don’t see the issue as that is the way the system is designed to work, you get alerted to a loss of integrity and you cancel the approach and go somewhere else that has an ILS or VMC.
But if you are saying that you have had erronous GPS positioning, with no annunciation to this effect, that is an entirely different issue, especially if you are talking about WAAS receivers.
FWIW, I would fly the outbound using HDG mode (if an AP is used) and twiddle the HDG bug to achieve the track specified on the plate.
Then turn, also using HDG, towards the LOC, and when the EHSI starts to move, press APR ont he AP (to avoid false LOC lobes). Or the manual equivalent.
Using OBS to fly the outbound is more slick but needs more buttons. I instead use the OBS to set up the LOC inbound, for situational awareness relative to the ILS itself, and have a checklist for that, because it is so easy to screw it up.
Rich,
the first part of your post supports my point. In order to improve GA IFR safety, IAPs need to be made (as far as possible) foolproof, because humans just have the habit of making mistakes every now and then.
The last bit of your post is remarkable. Are you saying that so far, you have always been flying NDB approaches (possibly in IFR conditions) on the ADF alone, despite having a perfectly cabaple GNS530 on board?
need to be made (as far as possible) foolproof, because humans just have the habit of making mistakes every now and then.
IMHO, that’s the only way to see it. The more simple and the more foolproof, the less pilots will die in CFIT accidents. Had we asked the accident pilot before his death if he thought it was possible he woud make such a mistake,… we know the answer. And since most of us are far from being 4000 hour ATPs… we are never safe from making similar mistakes.
That is actually the reason why i don’t fly IAPs without the active procedure on the MFD. The green airplane symbol on the procedure is what gives me peace of mind. And still: The GPS could fail, the MFD could fail. That’s why I have the same data on the iPad.
The downside is that all this terribly expensive. You don’t want to know how much i pay p.a. for the MFD data. I can well understand if people don’t buy the charts (you still have the procedure and the track on the MFD) … or don’t want to/cannot afford to install an expensive MFD … I can only speak for myself: I would not fly IFR approaches in IMC without that stuff, because I am fully aware that I am not good enough to safely fly the procedures under all circumstances. Sure I can do it when everything is working fine and the weather is ok. But let the autopilot fail, do it in really bad weather plus a system failure.
I myself sometimes call my IFR flying “IFR light” and I don’t have an ego problem with that. A long time ago i swore to myself that I will never do something in an airplane because others can do it. While i find it impressive that others fly hard core IFR in ice and through storms, it is not for me. That, and the ability to say NO will keep you alive, and not even then there’s a guarantee.
except if you make things too simple, people get bored, and don’t bother to monitor stuff diligently.
The trick is to keep the arousal up, yet present the information in an easy to assimilate form.
Flying in hard IFR is inherently more dangerous than light IFR, but some of the increased risk is mitigated by being more cognitively aroused by the workload.
I think what falls out of all this, is that there are many error modes in the human brain, many more than in a GPS or NDB and to try and understand them goes some way towards avoiding them.
boscomantico wrote:
The last bit of your post is remarkable. Are you saying that so far, you have always been flying NDB approaches (possibly in IFR conditions) on the ADF alone, despite having a perfectly cabaple GNS530 on board?
No 
– I’m lucky enough to have ILS and RNAV approaches at home base, and so far I’ve not needed to fly anywhere that only has an NDB approach. Why would you choose an NDB approach if others are available? So I’ve only done them during reval, without using the GTN650 (which, for the purposes of this thread, works in the same way as the GNS).
