Flight Tester.

Consider your motorcycle turns a little deeper.
Initially you have to START a turn with a force on the handlebars into the turn which initiates a bank/roll of the machine. The bank immediately and progressively causes turning instability as a result of its deliberate design and if left to itself would automatically tighten to a spin out. Once into a turn you have to use an opposite handlebar force to counter the instability. Motorcycle riders seem to prefer this type feel.

I wouldn't care to approve such instability for an aircraft!!

Milt,

You must ride some strange motorcycles or ride them strangely. I know I certainly don't prefer any feel of instability and I've been riding for near on thirty years.

Nigd3I find your comments related to testing that has demonstrated a pilot preference for the Russian O-I ADI very interesting. Is there a published report you might be able to direct me towards; perhaps something online at Royal Aeronautical Society and/or Society of Experimental Test Pilot?

RL

A good source to start is "The Outside-In Attitude Display Revisited" by Fred H Previc and William R Ercoline. This was published by the International Journal of Aviation Psychology, online in 1999.

Another interesting read is "Methods for Validating Cockpit Design" by Gideon Singer, which details testing of roll reversal errors when using different roll index formats.

The current attitude indicator sky pointer was mandated by the SAE Large Aircraft Cockpit Working group. Most people don't even know they exist.

Airbus have their own Cockpit Evaluation Group and Cockpit Integration Group, and develop equipment requirements. By the time most designers down the supply chain (at equipment manufacturers for example) get to see the requirements, many of the decisions (colours/layout/symbology) are already made.
Remember the glass cockpit displays have multiple layers and overlays possible so ensuring consistency is not that easy when multiple suppliers are developing the equipments.

hugel

I didn't think it was legal, let alone sensible to have totally electronic instruments in any cockpit, especially airliners ?

I know Russian fighters tend to rely on the HUD, ( I've photographed M-29's & SU-27's cockpits, which didn't go down very well at the time ) but they still have standby instruments and very good ejection seats !

Nigd3 Thank you. I found both the documents you recommended.

Shawn Your post confirms my understanding for transport aircraft. I am also aware of how Sperry, Bendix and Collins drove early instrument design.

I have not been aware of studies indicating a clear preference for O-I v I-O displays. My personal preference based on testing of both US and Russian equipment would be contrary, though I find both designs functional. I would like to understand the test conditions of studies related to an O-I v I-O preference. I am involved in a new design/test effort for a high workload cockpit. We are attempting to think outside of the box and if a significant improvement in situational awareness and/or reduced workload can be substantiated, or even if a clear preference has been demonstrated in testing, I would like to include O-I in planned testing using both test and operational pilots.

I've seen a new aircraft with all EFIS instruments including the standby's - I have no idea how they were driven, but certainly all the instruments were all flat screen.

Interestingly I heard of an accident with a Diamond TwinStar (I'm pretty sure that's what it was - new light twin though.) It has FADEC to control the engines - the alternators on the engines are not quite powerful enough to run everything when the gear retracts (electrically), so it robs power from the battery for a bit while the gear comes up. The problem was the battery had gone flat in the accident aircraft and had not been charged in accordance with the POH. The upshot was that on gear retract, there was no power to rob from the battery, and not enough oomph to drive the other bits and bobs, including the FADEC, hence the double engine failure that ensued.

Some of these all singing all dancing systems sometimes leave me a bit concerned, things like the Airbus incidents linked above do show that sometimes we are not quite as forward thinking as we would like to be. (Without wanting to sound like a luddite, I love technology, but only when it helps).

Jetstream I mostly agree with you, but mostly because I am a throwback to a by-gone era. The real problem is that nobody designs single task flight instruments anymore. We can argue the superiority of Inside-Out, Outside-In, Grapefruit (Arc-Segmented), Highway In The Sky, Integrated Flight Path Vector and other displays but must do so within the context of the entire information system. If we consider a transport aircraft there is one solution, for a fighter aircraft there is another solution, for helicopters there are so many tasks that one elegant solution is a design dream never to be realized. Then there is the mission. If the mission is to fly IMC along established airways to a high category landing I might prefer one design over another. If the mission is a high speed, low level ingress in zero/zero to attack a target I will never see visually, then I will certainly want another design.

It is impossible to design one perfect system for all applications, but it is possible to merge the best of different designs when there is task similarity.

A real problem for me is that data from simulator studies are often very different from aircraft studies. The cost of a test in an aircraft is much higher than a simulator study so more and more of this type of testing is being conducted in simulators; yet because of limited visual and other cues, I might prefer one display in a simulator and strongly prefer another for an aircraft. I think of video games. I almost always prefer and choose an Outside-In display because these provide the most cueing for the gaming task. I feel the same about simulators, because even the best of full motion simulators are cue deprived environments. In the aircraft I have a preference for Inside-Out displays. The question I am asking myself now is do I have this preference only because I am used to that type of display, or are there other more compelling reasons for the preference?

We are increasingly aware of an age biased flight display preference. The XBox generation see things much differently than the pre-computer generations. Although both can learn to adapt to each others systems, when asked they seem to demonstrate a primacy preference.

One of the best pieces of advice I heard regarding designing instruments is "Data is not information, unless it can be interpreted". If a pilot has to stop and think "what the hell does that mean", as stated previously regarding the caption ABN, then it is not an intuitive design. This obviously has the caviat "with reasonable training".

It sometimes can be difficult not to be distracted and objectively assess something in terms of does it do the job, even when presented with multiple/a significant failure, rather than, it looks flash and works wonderfully when all is operating ok, in normal flight profiles.

RL - maybe you should update the much used/abused phrase of "thinking outside the box" to "thinking outside the xbox". :):D:) .......I'll get my coat.
With regards your preference for the I-O ADI when actually flying, the publications I recomended may go some way to explaining this. When you lose certain sensory cues, that you have been extensively conditioned/trained to interpret, that are not represented in a sim or on a PC game, the brains natural preference regards the O-I as the more intuitive presentation. This is my view anyway.

As part of one of our Human Factors Workshops we were shown a video of an F-16 pilot in combat being locked up and fired upon by multiple SAM sites (he escaped). The interesting thing about the whole sequence (which lasted for around two minutes) was that all he had to do to escape the Missile Engagement Zone was take up a southerly heading - that was it, no requirement to pop chaff and flares, no altitude changes, no power changes - just fly South for thirty seconds. His wingman was yelling at him over the radio to follow him and fly south.

The pilot of the F-16 had reached a point of sensory overload, adrenalin was flowing, breathing rate had gone up and gotten shallower and he was scared - result, the higher brain functions had shut down and the guy was flying a fast jet using what the neuroscientists call "the caveman path". Didn't matter what he'd been conditioned/trained to interpret - as a species we have had millions of years to evolve but we are still hard wired to avoid being eaten by large animals on the African savannah - instinctively we know that the best way to avoid the large 300 pound tawny cat bearing down on us is to jink because we know it can outrun us in a straight line. As a result of his hard wired response, the pilot elected to jink his aircraft left and right but remain in the MEZ with missiles leaving contrails all around the aircraft. He eventually got a grip of his caveman instincts and flew south out of the MEZ.

It was a salutory reminder that even highly trained and conditioned people can reach overload under stress and that cockpits need to be designed bearing those situations in mind.

Nigd3[QUOTE][RL - maybe you should update the much used/abused phrase of "thinking outside the box" to "thinking outside the xbox". /QUOTE]Clever. There's a giggle in that thought.

First, I accept that I am extensively conditioned but I like to believe that my training and experience allows me to overcome this conditioning in the conduct of research and test. I have noted that there are several researchers and pilots who seem to share your preference for the O-I dispaly; yet my reading to date, admittedly limited to your recommend publications and about 10 other papers, seems to provide no real concensus for a clear Outside-In preference. It is early in my research so I have yet to reach any personal conclusions, but I am beginning to feel as if a simple comparison of design concepts will not be possible as there are too many variables in the published research to form a useful compartive matrix. I am, however, seeing a preference in some of the more recent documents for the Arc-Segmented Attitude Reference Display, which I am sure you are aware is classified as an Inside-Out concept. My real difficulty in comparing the various displays is that the published researched I have read to date are not conducted using a single standardized research protocal. Further, the tests never seem to evaluate more than one display and many use old technology or adapted computer displays. For instance, one uses a panel mounted display, another legacy instrument panel displays, another a computer screen, another HMD fixed or non-fixed, another full face visor (fixed and moving frame of reference), and still another point-in-space/infinity projection - yet most researchers use only one display for their tests and this introduces an obvious test bias. One might conclude using a panel mounted flat screen display that O-I is preferred while another using a helmet mounted HMD with superimposed and fused FLIR/IR imaging underlay and flight and navigation symbology overlay that an I-O is the preferred display protocal - which then becomes the design standard when both displays are available for the pilot to use? Sometimes the design choices are not so clear. Most aircraft companies understand that testing an infinite matrix is a bit cost prohibitive and that is why introduction of any new standard requires clear and convincing proof that the new standard as far superior to the old.

RL - Firstly apologies. I have no real knowledge of your flight training or test and evaluation experience, hence making assumptions on your capability to adapt to different situations is unfair. In my defence, it was intended as a generalistic viewpoint, rather than trying to identify an individauls (your) behaviour, although it doesn't really read that way.

It would be very interesting to hear if you do assess an O-I ADI, or any other formats and any results that you are allowed to publicly detail.

Another problem that is stated in one of the publications I listed, is that most research on O-I ADIs was undertaken in the 1970's and not a lot has happened since.

Rich - I agree with you too, I cannot imagine how a fast jet pilot would be able to accomplish a mission with the sort of instrumentation I look at every day. There is a real difficulty in specialising instruments for a given job though, and that is when someone transfers from one to the other.

I've never flown an Airbus, but have discussed at length some of the the differences with various colleagues who have. Plenty of stuff makes me think "why on earth did they do that?" and after a few conversations and different views I begin to understand what the designers were getting at (at least some of it, some of the ideas still seem totally bonkers to me!). I am sure it would be the same the other way round. The fact is that people change from Airbus to Boeing and from fast jet to airliner, so some common ground does help.

We also then get stuck in ruts - if we standardise everything, then when the new technology comes along it is suddenly non standard, so do we stay in the "dark ages" all the time, or improve? It is such a difficult situation, that often I think we need to be thankful that humans are so adaptable. No cockpit design is ever going to get it right, but that is no excuse for making a bad one!

There is so much that seems obvious and intuitive to me - but a lot of that is due to my upbringing and training. Red is lucky in China, so probably not the best colour to use for a warning caption. GEAR NOT DOWN - lucky us!

Nigd3 There was no offense taken in anything you have written and no need for an apology. There is merit in your opinion and in this discussion.

I am, as others in this and other forums have at times noted, a rather thick-skinned and witted American who only rarely understands he has been insulted - even when such insult is intentional. The truth of it is, if I didn't listen to those who disagree with me I would never learn anything. You will also note that when someone says something monumentally stupid in any forum other than Jetblast, I usually don't bother to answer. I have discovered in those cases a reply is a bit like wrestling with a pig in the mud; sooner or later you realize the pig likes it.

You raised a very interesting point and given the comments of test pilots I have come to respect in this forum that have agreed with your opinion, you have caused me to evaluate my own bias with, as Jetstream Rider so eloquently stated in the last post, a hope of finding the more elegant solution rather than just doing the same thing because that is the way we have always done things in the past. Your comments have already been very beneficial as I have questioned my core beliefs and as a result of the research have a better understanding of the evolution of the various designs. As some bard once said, there are many paths to the top of a mountain.

I've just got out of the bath and I've been thinking. In a sim session one day we were discussing at the briefing what we would do if we have an engine failure on a North Atlantic Track - we both came up with reasonable ideas and then jumped in the box. When we were actually presented with the scenario some time later, we did something different. Both ways were acceptable, within SOP's, but nevertheless different. My psychology isn't up to much, but I reckon we were probably using different bits of our brains.

I see the same thing with the endless questionairres we are sent by various people and businesses - what you actually write may be very different from your view. Questions such as "Are you proud to work for xxx?" are easy to identify (yes and no, so where do I tick?), but others less so. When you introduce lack of cues, it compounds the problem, when carrying out the survey, you are usually deprived of the cues you need like a dirty table top in a restaurant. I wonder how much of a factor this is in aviation research? Would the outcomes of some of these O/I v's I/O studies for instance actually be a long way in error?

That might sound like I like one type and am sticking to it and only finding things to support my view - but I promise that's not the case! I mean in general terms, rather than specifically. After all there are instruments out there that have been through many meetings and tests, yet still produce erroneous results.

Was thinking of a HUD too - airliners are getting them soon - does what works best for a HUD, work differently for a panel? Having not flown with a HUD, I have no idea. Do we have multicoloured ones now? How do green lines compare with glorious technicolor?

Just finished development flying for certifying HUD on one of our large business jets. Short answer is yes. There are things that you want on the panel that you don't want on the HUD and vice versa - nice feature on the HUD is the flare cue - bit like a FD pitch target. Wouldn't want it on the panel display though. No colour but green on the HUD, so you have to be more inventive with the failure indications - either strikethroughs or boxes - sometimes not as intuitive as a yellow or green indication. Likewise, with the Flight Mode Annunciators.

In terms of distraction the HUD can be a major headache - on the HDD having the flight path acceleration cue bouncing up and down during taxi is not a major distraction, when it's collimated on infinity and doing it right in front of your face in symbology that's "apparently" 6 feet high it a real bugger!

We've had HUDs on airliners for well over twenty years that have got progressively more & more sophisticated but still with only one color - green. That's because the reflection HUD uses a single frequency holographic mirror.

Given the main theme in this thread,the HUD is possibly the one modern display that doesn't do Unusual Attiudes (UA) very well. You can have sky pointers, declutter at set Pitch & Roll angles and include ladders and lines but it's still not easy to decide which way is up - especially when you're in a total whiteout with no outside clues.

The other problem when you're only allowed a single color is that it's relatively easy to draw something that actually will "pop" on you such that the shape goes inside out as you blink.....

Rich Lee asked: "On 10 January 2000, at 16:54:10 UTC, in darkness, on runway 28 of Zurich airport, the Saab 340B aircraft of the Crossair airline company, registered HB-AKK, began its scheduled flight CRX 498 to Dresden. Two minutes and 17 seconds later, after a right-hand spiral dive, the aircraft crashed on an open field near Au, Nassenwil ZH."

The commander was a Moldovan citizen who had flown nearly all his career in Soviet aircraft.

"The following factors may have contributed to the accident:

• The commander remained unilaterally firm in perceptions which suggested a left turn direction to him.

• When interpreting the attitude display instruments under stress, the commander resorted toa reaction pattern (heuristics) which he had learned earlier.

• The commander’s capacity for analysis and critical assessment of the situation were possibly limited as a result of the effects of medication."

http://www.bfu.admin.ch/common/pdf/1781_e.pdf

John Maris of Marinvent has patented Dynamic Non-Linear Display formats that provide information about the complete altitude and airspeed range of the aircraft (as just two examples) on tape instruments. Ideal for unusual attitudes and dynamic maneuvers like emergency descents. Anyone want to try them in an R&D setting can send me a PM.

AI design
 

The Flash Airlines B737 accident at Sharm el Sheikh on 3 Jan 2004 was partially attributed to spatial disorientation of the Commander and one of the factors was allegedly his Air Force background (Mig 21). Note that the Egyptian CAA report and the NTSB and BEA comments do not fully agree. See:
BEA
and go to Reports/Search 2004.

F-111 correction
 

Nigd3 - The F-111 wing sweep lever was not such a brain stretch and was never redesigned. It remains as ever, wing sweep lever aft = wings aft. The relationship between the F-111 park brake handle and the hook handle, now that's another story.

Bentstick

Interesting update/correction on the wing sweep lever. It is contradictory to the HMI thesis I read but it sounds like you have first hand knowledge of the lever and its operation, so I'm not really in a position to dispute with any real strength.

Whats the story with the park brake and hook handles?

You can confirm it for yourself by looking at cockpit photos of F111s (any model, any era) when parked on the ramp. You'll notice that the wing sweep lever is always fully forward (16 deg).

The Hook and park Brake handles were of similar(ish) shape and size and mounted about 10cm apart, just fwd of the wing sweep lever(Park break was grey, hook handle was yellow) . Many a Pig driver blew the hook down while attempting to engage the park brake.

Hi
 

But did u ask them if they were happy flying the Airbus??

Engineers and pilots
 

hello
I think it is a wrong approach to want to change instruments. We've saddly lost the first SF34 in ZRH because of this you're trying to tell me?
I think that you're trying to solve the "how" this happened while you should try to fix the "why".
But for heck of it ...I hate this in cockpits:
A) Clicking. Easy,787, 380, 777, E145,E170...Never heard about touch screens like in jet fighters? Why do I need to see Windows like menu in an aircraft.
B) FAULT lights with partial system working. Is this thing working or not????
C) Poor audio radio quality. Should be digital (less fatigue) with auto frequency switching
D) Control laws: Too complicated and there are always exceptions. Why? The engineer will give you 10x10000 reasons why, but he's not flying the thing.
E) Virtual MCDU keyboards. :{
F) Limitations. Do not operate the gear if pressure is below 1450 PSI. Do I care? No, I need to raise of lower the drag generator. End of story. Fadec can compute several parameters per second but I still need to keep an eye on the oil pressure when starting the engines. Thanks automation.
G) Switches versus push buttons. I prefer flush, pushed in for normal, LED bulbs. Green= Works ON. Red = failed stopped operation. Not lit = OFF.
H) Flap handle. Should be kept for backup only and a new logic incorporated in speed modes allowing flaps extension for slowing down. EGPWS will yell at you TOO LOOW GEAR/FLAPS but no piece of automation here...still manually have to lower the gear. Even 3 radio altimeters not sufficent to tell that we can safely raise the gear automatically. We have auto slats but no auto flaps.
I) Garmin 1000 on VFR (mainly) machines. So you have to operate VFR while spending most of your time punching buttons head down on the displays of a single pilot operated machine? Watch the other guy.
J) 0/0 capabilities. Engineering dream. How do I taxi if I cannot legally rely on the moving map of my EFB?
K) Airbus controls (stick and throttles) not showing movements.
L) Pure engineering concepts. 'Kmon have you ever seen a concept car commercially available? No, that's why it is called a concept. Concept and practical don't go along (use is practical but manufacturing is nightmare).
If you're french and born in Blagnac then you can understand Crew Operating Manual. (How to operate the crew right?). If you're from KBFI then you can understand Aircraft Operating Manual (How to operate the ship).
M) Vertical speed mode. This thing is on since 800 BC on ships and no one came up with something smooth.
N) HF radio. Yeah, I still have my 56k modem for surfing the web!

I love:
A) EVS, HUD
B) EICAS/ECAM message auto arrange
C) OEI auto pop up for SID, Climb...
D) TAC on 777
E) Antiskid (everybody thinks it is granted like on cars. Remember 40 years ago...)
F) Wx radar mapping by FL.
G) FANS, ADS, Satcom...
H) Electronic checklist
I) EFBs vs laptops.

To conclude:
The why vs the how...
It is very complicated for a pilot to write down a robust spec to an engineer.
It is very easy for the engineer to implement the IF ...CONDITION then ACTION from the spec, but it is awfully complicated to write a ELSE if nothing is given in the spec. Most of the problems are coming for this missing statement (unspecified). And it is getting worse because of commercial pressure( shorter coding, testing, integration, robustness test time), more and more combination of conditions are remaining untested or deliberately put into the class "unlikely to occur".
"This failed because these conditions where never tested". "This plane crashed because pilot was tired". You would like to design something that will tell a tired pilot he's making a mistake but next thing management will ask is to have our flight duty extended and fly with more fatigue.
I do beleive that making these 2 worlds (Engineers/Pilots) "talking" together is an art. I love to write sequencers in C versus "Cool" realtime applications in Java with embedded windows XP. But between you and me: KISS (Keep It Stupid & Simple).

You have raised some very interesting points. The one I have quoted above is another instance of the Automation vs Control debate. Not everything can be developed from the pilots’ use-case. Redundancy, reversion and BITE data-validity checking may be going on of which you are not aware. What level of abstraction is reasonable ? At what level would you want to interrupt the automation workflow if a fault develops (assuming it is safe to do so) ?

The important thing is to avoid latent failures. It is very difficult to establish the impact of partial loss of inputs to a complex system, in order to allow continued operation. Which functionality of the equipment is still OK ? It is safer to declare the unit non-functional and switch to an alternative. So I agree a “partially failed” indication as mentioned elsewhere in your post is not very helpful !


It is not possible to cover testing of all combinations of input scenarios, but the classification DO-178B (Airborne Software Considerations) identifies categories of software an applicable levels and type of testing required. A hazard-analysis bottom-up (what-if ?) and a Fault Tree top-down (what-could-cause-this-top-level-event ?) should cover hazardous events and identify mitigation to reduce to risk of the outcome.

Getting users, even highly trained and intelligent-ones like pilots (!) to express their requirements is difficult and is often based on “I want it like this - but better”. It needs to be related to a specific aspect of an existing system of an available prototype to demonstrate the concept.


I would be interested to know your views of data-fusion, and integration of equipmnet displays and data sources.


hugel

The BROWN-UP instrument
 

I understand the thread moved past this particular item, but perhaps someone may still find little info on the horizon interesting. The part number is LUN 1202, electrically powered through voltage changer. Best picture I could find is here:

http://pwdt.virtualskies.net/images/lun1202.jpg

Used on 70's design eastern trainers (Zlins), simple design now obsolete. I completed my initial instrument and night training with these.

The mechanical principles are:
Aircraft symbol is fixed in bank. The left knob serves for adjustments in pitch designed to fine tune the vertical position of the aircraft symbol so that in level flight with different speeds pilot could "zero" the indicated pitch and maintain precise reference. I was told not to use it, nor felt it would be required.

The coloured ball inside (not the side-slip indicator) maintains position in 3D with the vertical divider aligned to true horizon.

How to read:

The picture shows a slight nose down attitude, for basic recognition the upper part over which the aircraft symbol is now portrayed is painted brown with a label "descends". Suppose you would like to return to neutral pitch: look at the ball's vertical divider and "pick" it up to align with (fixed) aircraft symbol. The ball's divider needs to go up - pull back on the stick.

Bank is read by comparing the top-down line on the ball to the scales at bottom. The gyro-ball is again stabilised and the whole instrument (and aeroplane) revolves around it. For a right 15 deg bank situation, the scales would turn clockwise over the stabilised background gyro ball. I.e. you need to pull the top-down line drawn on the gyro-ball out from the neutral position behind the scales to the first mark on the right side. As you apply right rudder to keep the sideslip indicator centred you will see the turn coordinator located right in between these to sway to right, all conveniently located at one spot. So once we had banked to the right you would see the top-down line on the reference ball aligned behind the right side of the (fixed) scales on the bottom of the instrument. To return wings level, you push the top-down line back to centre with a left input on the stick. Or, as I was told on the first day of the training "kick" the gyro back to neutral position - with your knee kick the stick.

Maybe different teaching would be required with yoke equipped aircraft, but for Zlins this worked seamlessly.

Yours,
FD (the un-real)