U.S. orders airlines to replace cockpit displays on 1,300 Boeing airplanes

[underfire]

"SUMMARY: We are adopting a new airworthiness directive (AD) for all The Boeing
Company Model 737-600, -700, -700C, -800, -900, and -900ER series airplanes, and
Model 777 airplanes. This AD was prompted by testing reports on certain Honeywell
phase 3 display units (DUs). These DUs exhibited susceptibility to radio frequency
emissions in WiFi frequency bands at radiated power levels below the levels that the
displays are required to tolerate for certification of WiFi system installations. The phase 3
DUs provide primary flight information including airspeed, altitude, pitch and roll
attitude, heading, and navigation information to the flightcrew. This AD requires
replacing the existing phase 3 DUs with phase 1, phase 2, or phase 3A DUs, and for
certain replacement DUs, installing new DU database software. We are issuing this AD
to prevent loss of flight-critical information displayed to the flightcrew during a critical
phase of flight, such as an approach or takeoff, which could result in loss of airplane
control at an altitude insufficient for recovery, or controlled flight into terrain. "

https://s3.amazonaws.com/public-insp...2014-23231.pdf

[BOAC]

Hmm - just as restrictions on radiated signals from the myriad on-board mobile devices are relaxed?

[Skyjob]

How does the FAA get the figures?
According to the report half those affected units requiring replacement are in Ryanair and the other half in the rest of the world of 737's? Seems a bit strange.
Either way, a major financial setback on each operator requiring replacement of the DUs over the next 5 years.

[Denti]

The results of the initial tests that shows the problem are pretty old, interesting that that AD got published this late. We were aware of that for several years now and made sure we didn't have phase 3 DUs in our fleet, although inflight WiFi wasn't implemented nor approved.

[Ian W]

Honeywell are a little miffed as there has been no example in flight of a DU failing and they feel that the tests were not realistic as the displays do not operate standalone but bolted into the cockpit instrument panels and are not affected in the same way.

[EEngr]

the displays do not operate standalone but bolted into the cockpit instrument panels and are not affected in the same way.

This is true. But the certification process has to account for an aging fleet. Where the grounding path between cockpit panels and the airframe may be degraded by dirt, oxidation and repeated removal/reassembly for maintenance. So the system design gives no credit to the shielding characteristics of cockpit structures. The alternative would be a repair and reassembly procedure that would include cleaning and treatment of each panel and LRU to ensure proper RF shielding. Plus a procedure to check this following any maintenance. So the DUs have to operate in 'free air' even though the structure does provide some level of shielding since it isn't an easily controllable factor.

[captplaystation]

Empirical evidence suggests that wifi devices, even when used in the cockpit (with the engine start levers "not in cut off") do not appear to have affected up to 1200 pilots for at least 3 years . . . . . .

[tdracer]

Honeywell are a little miffed as there has been no example in flight of a DU failing and they feel that the tests were not realistic as the displays do not operate standalone but bolted into the cockpit instrument panels and are not affected in the same way.

But that can't be! The FAA is in the pockets of the manufactures - Al Jazeera said so The FAA would never act contrary to the manufacturers!


OK, sarcasm off...
In all serious, I've never known the FAA to accept 'service experience' for HIRF/Lightning compliance. For HIRF, the regulations (or in this case Special Conditions) say the system must withstand X volts/meter at Y Hz without upset - either you pass or you don't, no gray area. Claiming that the requirement isn't representative of the actual threat gets you absolutely nowhere.


BTW, this AD is simply mandating what Boeing recommended two years ago.

[captplaystation]

Exactly, total Horlicks. . .a "company" in Europe offers free wifi to all of its pax, can you imagine the crew don't wanna bit of the action ? so, in he "real world" no issues.

[Jwscud]

Captplaystation tells it like it is. I was chatting on Facebook with a friend at NAS a few months ago when he said he had to go as it was time to brief the approach

[Ian W]

MSBBarratt

Who cares if fixing the displays into the instrument panel makes the problem go away? What cannot be determined or measured is the degree of immunity gained by doing so.

Actually, you should care. Of course tests can be set up in a realistic cockpit installation. Indeed, I would say that there must be system level tests. It may be that there is some level of anomalous propagation rather than shielding when the equipment is bolted into the actual flight deck. If those system level tests show a huge reduction in interfering signal strength then insisting on more internal shielding may be a nugatory exercise.

The question is: "at what point do you stop?" If you go inside the instrument panel shielding you get interference. True; and if you go inside the shielding on the equipment you can get interference too. So perhaps there should be another set of shielding just in case the first is damaged with one of the small scratches you quote?

At some stage one has to step into the real world and look at the entire airframe and test that. Then look at the vanishingly small chances of:
(1)Display blanking for a few seconds (no cases at all in operational service in millions of hours)
(2)at precisely the wrong moment (must happen at critical phase of flight and
(3) Simultaneously to BOTH PF and PNF displays) and
(4)Simultaneously with a standby systems failure (ever seen standby instruments fail?)
(5) IMC
(6) with a pilot who cannot cope (vanishingly small chance )
-- and accept that the probability of all these holes in the cheese lining up is too small to be concerned about as there are so many other hazards that are routinely accepted that have a far far higher probability of happening.

Should we ground all aircraft with any engine cowling fasteners and demand all aircraft engine cowlings be redesigned?
Should we stop operating if there are _any_ birds within the airport boundary or flying within the area of the SIDS/STARS
etc etc etc

Note that the above examples all actually cause accidents incidents

[BOAC]

One major error there -

with a pilot who cannot cope ( increasingly likely chance)

[calypso]

(1)Display blanking for a few seconds (no cases at all in operational service in millions of hours). Or a few minutes or until landing


(2)at precisely the wrong moment (must happen at critical phase of flight. Or at any time at cruising altitude. Have you tried flying visually at 41000 feet?

(3) Simultaneously to BOTH PF and PNF displays) and

(4)Simultaneously with a standby systems failure (ever seen standby instruments fail?) Yep, I have.

(5) IMC see (2)

(6) with a pilot who cannot cope (vanishingly small chance ). Some can hardly cope WITH the displays working.

[tdracer]

I have my issues with the FAA/EASA defined HIRF "threat" (there is one particularly high frequency where the threat is defined as, IIRC, 18,000 volts/meter. It's so high that we have issues generating the field strength in a lab environment (and if encountered in the real world would probably mean you'll never have any more children, at least not 'normal' children ).
But that being said, the FAA/EASA consider HIRF to be a 'common mode threat' - i.e. redundancy doesn't help because the whole airplane is going to be exposed to that threat. In other words, if one display blanks due to HIRF, ALL displays will blank - and I can't dispute their reasoning. Further, it's assumed that the external HIRF threat is greater closer to the ground (since that's where the high power radar and similar HIRF generators are located). Again, hard to dispute their reasoning.
So as far as the FAA/EASA is concerned, the threat is that all flight deck displays will malfunction at the same time, at the worst possible time during final approach.
Given that recent experience has demonstrated that some flight crews have a hard time landing a 777 on a 2 mile long runway on a perfect summer day with everything working correctly, it's not exactly unreasonable to assume that some flight crews might have a bad day if all the displays blank while doing a CAT III landing under extremely challenging conditions

[cockney steve]

Surely, the real issue is , A POS, unfit -for-purpose, inferior display was CERTIFIED AND APPROVED and on that basis, customers bought it

Now, they're being told, you've been conned, bend over and we'll screw you again, and, no, you don't have a choice and we dictate the price.

It's a damned disgrace, I bet there's not a single member here who would tolerate a similar situation if they.d bought a car which had an inherent design fault in the brakes or steering.
"warranty" period is an irrelevance....fitness for purpose for the design-life or expected service-life is what's important.

[FullWings]

it's not exactly unreasonable to assume that some flight crews might have a bad day if all the displays blank while doing a CAT III landing under extremely challenging conditions

Actually, that’s probably one of the less problematic scenarios. The autopilot(s) will be engaged, so whether the display units are functioning correctly or not has no bearing on the flight path of the aircraft.

Even if you lost all the EFIS on rotation, you could engage the autopilot and fly the aeroplane through the MCP and CDUs. It’s unlikely that you’d lose the standby instruments, which are quite different to the primary ones and mostly self-contained.

I’ve got >10,000hrs on the 777 and have never experienced nor heard of anything like this happening. I have had a few DU failures but given the redundancy available, this is a non-issue. Obviously, spending a lot of time sitting in front of them at night/IMC, I’d like the displays to be as hardened as possible to any interference but it does seem to be a chain of “what if?” down to ridiculously low probabilities...

[Aluminium shuffler]

Why an Air Worthiness Directive now?

So my understanding is that in an ideal world any pilot would be able to land their aircraft safely using nothing but the basic electro-mechanical attitude, altitude, airspeed and compass indicators. Surely in an ideal world all the electronic displays going blank shouldn't be a problem at all (an inconvenience at worst).

I, as SLF, was kind of assuming that an air worthiness directive meant that something had to be done to ensure safety of flight. I appreciate that there is likely to be more to it than that. But does it imply that use of the electronic displays now officially considered essential rather than being simply very convenient?

Shame you added this to an otherwise excellent post. Unless you have tried it, then you will not realise how tricky flying on the standby instruments is in weather. They're tiny, offset, suffer badly from parallax error and display in a format we have been trained out of by modern EFIS formats. They really are a last resort for landing in moderate conditions - I wouldn't fancy shooting an ILS to minima on them in windy or gusty conditions. I do appreciate your confidence in us, though - it makes a welcome change!

[172_driver]

Surely, the real issue is , A POS, unfit -for-purpose, inferior display was CERTIFIED AND APPROVED and on that basis, customers bought it

Now, they're being told, you've been conned, bend over and we'll screw you again, and, no, you don't have a choice and we dictate the price.

True - but with that logic should we just scrap AD's all together? They provide a safety net for those unexpected design flaws that come to light as result of experience.

Just a question of who foots the bill..

And whether the risk involved warrants an AD, which the FAA seem to have deemed.

[Ian W]

Originally Posted by calypso

(1)Display blanking for a few seconds (no cases at all in operational service in millions of hours). Or a few minutes or until landing


(2)at precisely the wrong moment (must happen at critical phase of flight. Or at any time at cruising altitude. Have you tried flying visually at 41000 feet?

(3) Simultaneously to BOTH PF and PNF displays) and

(4)Simultaneously with a standby systems failure (ever seen standby instruments fail?) Yep, I have.

(5) IMC see (2)

(6) with a pilot who cannot cope (vanishingly small chance ). Some can hardly cope WITH the displays working.

You miss the point - all 6 have to happen at precisely the same time.

Now just calculate the chances even if you put them at low highish probabilities. The probabilities almost certainly meet the required level of safety.

Re 2 - yes I have; there was considerable invective on this board against the crew of AFR447 who lost instruments and were unable to fly for a few seconds without their main instruments - it was felt that any competent pilot should have been able to maintain control. However, you make an excellent case for unmanned commercial aircraft.

[John Farley]

However, you make an excellent case for unmanned commercial aircraft.

I know you may be jesting here, but we don't want unmanned commercial aircraft. What we want as I have said over and over is autos that are fail safe (like stuctures etc) so that the crew - which we must have to operate the aircraft - do not have to steer the aircraft.

All the pilots I know (including me) have from time to time made a bad job of steering an aircraft. Autos don't have those sort of bad days. However the idea of autos operating the aircraft - that is making all the decisions about a flight is plain silly.