Wasps force two passenger jets into emergency landings

More proof that everything in Australia is trying to kill you: the nation's Transport Safety Bureau (ATSB) has published details of how wasps have caused two passenger jets to make emergency landings. The good news: the wasps in question are not some kind of giant mutants that can bat a plane from the sky and sting it on the …

[Voland's right hand]

Nutters

What I find worrying is that they tried to take off with faulty airspeed sensors. As if the Air France 447 accident never happened.

Doing this on a modern fly by wire passenger jet is in the realm of suicidal. Dunno which deity do they have to thank, but thankfully they managed to land it too (that, once again, without working windspeed sensors on a fly-by-wire plane is an interesting exercise).

As far as "trying to kill you", these guys (the wasps) actually try to kill what is trying to kill you. Spiders are one of their favorites and they happily take on spiders several times their size.

[wolfetone]

Re: Nutters

"What I find worrying is that they tried to take off with faulty airspeed sensors. As if the Air France 447 accident never happened."

Your worrying about the wrong bit.

Planes have 3 pitot tubes, and in the case of Air France 447 two of the pitots became blocked with ice. It was then the actions of the pilot who kept pulling the nose of the air craft up that caused it to stall and belly flop in to the sea. Pilots must keep the plane level and constant in such a situation until the problem is rectified.

What I find worrying is that the article says only one pitot tube was blocked with these wasps. The redundant system should still be working, so why the plane behaved like it did with it's back up system assumed to be working? That's the worrying bit.

[Voland's right hand]

Re: Nutters

Your worrying about the wrong bit.

The picture described in the report is as if two airspeed sensors have gone bonkers, not one. That is a fault which should not be "sign-able off" for a take off.

why the plane behaved like it did That is not the only question. The second question is - why did the captain decide on a take off despite it behaving like it did. I also stay by my opinion that they got lucky here. Why did the airspeed sensors not work correctly is unclear. It is a fact that they did not and landing a "heavy" passenger jet without correct airspeed reporting is in the realm of dice rolling.

[wolfetone]

Re: Nutters

I haven't read the report, keeps being blocked on my work internet for some reason so I can only reference the article.

The pilot rejected the first take off because of air speed, and the systems were checked and were obviously found to be normal. On the second attempt they kept going. The article doesn't say whether or not the plane made it past V1. If it did this before the airspeed became an issue then the pilots did the right thing. You can't slam the brakes on a heavy plane going faster than V1. It'll crash. However, if the airspeed warnings occurred before V1, then the pilots really were playing Russian roulette.

[Hugh McIntyre]

Re: Nutters

In terms of V1:

Based on the the ATSB article they were past V1 by the time it failed the second time, so continuing the takeoff makes sense. See https://www.atsb.gov.au/publications/investigation_reports/2013/aair/ao-2013-212.aspx:

"During the second take-off roll, the crew became aware of an airspeed discrepancy after the V1 decision speed and the take-off was continued"

In terms of 2 sensors:

Reading the other details, the blockage was apparently not detected on the ground when they checked the aircraft after the first failed takeoff, and they didn't have 2/3 redundancy for the second takeoff because "The aircraft was dispatched with the ADR part of ADIRU 2 inoperative (switched off) in accordance with the MEL". So arguably if they'd continued with takeoff #1 they would have had 2 working sensors, although not meeting the flight rules and would still have had the problem with the flaps.

[BlackKnight(markb)]

Re: Nutters

just add to the "why did they take off"

Shouldnt the preflight walk around have picked this up? you know where your taught in basic flight training to alway check the pitot and static vent because the pressure instruments are kind of important.

[Alan Edwards]

Re: Nutters

> so why the plane behaved like it did with it's back up system assumed to be working?]

If I've read the article right, they noticed one of the airspeed sensors was wonky but the backup was fine, so continued the takeoff. After takeoff something else went wrong and the flaps came up but not the slats, which triggered the return to Brisbane. Makes sense, the slats are only for low-speed flight.

I was expecting a giant wasps nest behind the slats to be the problem, but it sounds like something was reliant on the wasp-infested airspeed sensor and gave up the ghost. There's probably a safety system that stops the slats retracting if the computer thinks you're going too slow.

The Air France incident (if it's the one I'm thinking of) was a frozen attitude sensor. They were trying to test the anti-stall and had deliberately stalled it, but the computer didn't know and did nothing about it.

[JeffyPoooh]

Inventions - "Dime a dozen, cheapers in bales of twelve..."

Inventions - "Dime a dozen, cheapers in bales of twelve..." 28 Feb 2014 *

Given the ongoing problems with blocked pitot tubes, I hereby place these novel (?) inventions into the public domain.

First: Using valves, high pressure air, and appropriate pressure senders, periodically blast some high pressure air out the pitot tubes. Monitor the resulting pressure signal for irregularities. Integrate this novel Pitot Tube (itself!) Built In Test (BIT) feature with the Air Data system so that it can be done periodically during flight. The blast of high pressure air will simultaneously clean and test the actual tubes themselves. Any blockages would be fired out.

Next: Use photosensors to monitor ambient light entering the pitot tube itself. Compare it against other references. Use LEDs to add light if required. Darker than expected means blocked.

These concepts would provide the ability to actually test the Air Data system, including the Pitot Tubes themselves. They'll require very clever implementation to avoid increasing the overall failure rate. The photosensor concept is probably more harmless in this regard. But I like the blast cleaning concept.

Cheers.

*http://forums.theregister.co.uk/forum/2/2014/02/28/internet_of_things_new_enemy_hordes_of_angry_wasps/#c_2121343

[Vic]

Re: Inventions - "Dime a dozen, cheapers in bales of twelve..."

Using valves, high pressure air, and appropriate pressure senders, periodically blast some high pressure air out the pitot tubes.

You're going to take a sensitive pressure measurement instrument and inject high-pressure air into it?

Look at the failure modes. One moderately stubborn blockage - such as ice - means that your system would permanently destroy the pitot. A frozen pitot is trivially fixed using standard procedure, and an inoperative pitot - whilst quite a significant annoyance - isn't really a safety hazard. I have landed aircraft with the ASI gauge deliberately covered; it was part of my training.

Use LEDs to add light if required. Darker than expected means blocked.

Aside from the fact that the range of expected illumination would vary immensely, detecting a blocked pitot really isn't hard without your system. What matters is what you do about it.

Vic.

[Lars]

Re: Nutters

Referring to fly by wire has become a bit old fashion as it's no longer a Boeing/Airbus difference. I would expect the pitot tube to be redesigned as this is an old and silly problem.

[Chris Miller]

Re: Nutters

I'm not sure that a redesign of a Pitot tube is either called for or possible. It's basically a tube open at the forward end and sealed at the rear which measures the pressure generated by forward movement. If the front end is blocked (ice and small creatures are potential problems in this area), it won't work, which is why there are always at least three of them. But flying with 'unreliable air speed indication' is standard training for every pilot and shouldn't be too problematic, as long as you can recognise what's happening. It's the fact that two fully trained Air France pilots couldn't manage to do it that's worrying.

It's interesting that pilots learn to fly on 'indicated air speed', which is not directly related to true speed, either through the air or on the ground. A Concorde pilot once told me that indicated air speed at Mach 2 (and 60,000 feet) was between 500-600 mph.

[Known Hero]

Re: Nutters

@Chris Miller

Here have a quick read

http://www.theregister.co.uk/2012/08/07/laser_measures_airspeed/

would coincidentally, also be very easy to check if its blocked

[SkippyBing]

Re: Nutters

'It's interesting that pilots learn to fly on 'indicated air speed''

That's because that's the speed through the air being experienced by the wings which is more useful for the pilot to know than his speed across the ground. In the right aircraft with the right wind ground speed could in fact be negative.

The more accurate air speeds, from memory, correct for temperature and pressure, so represent the speed through an ideal atmosphere rather than the one you're actually in.

[Alan Brown]

Re: Nutters

"In the right aircraft with the right wind ground speed could in fact be negative."

Been there, done that.

In the wrong aircraft with the wrong ground speed in the wrong weather (microburst), air speed can also be negative. At that point the aircraft does a good imitation of a stone.

[Anonymous Coward]

Re: 'It's interesting that pilots learn to fly on 'indicated air speed''

@SkippyBing: what you're describing is the difference between (true) airspeed and ground speed. The original note concerned the difference between indicated and true airspeed and the fact that planes are flown by indicated airspeed. That's because it's not really airspeed that matters but dynamic pressure and that's what the pitot tube is actually sensing. The airspeed indicator is really measuring dynamic pressure with nominal airspeed figures (only good for standard atmosphere at sea level) painted on the dial.

[BlackKnight(markb)]

Re: Nutters

Indicated air speed makes sense.

Ground speed dictates your trip time

Airspeed dicates the lift generated from the wings.

IE if your stall speed is 60kn and you have a 10 kn tail wind you would need a ground speed of 60 to not fall out of the sky. as wind direction and speed constantly changing its easier just to remember 60 on the airspeed indicator and aim for that.

[Vic]

Re: Nutters

It's interesting that pilots learn to fly on 'indicated air speed'

Not really.

IAS is determined by the effect of the air on the mechanics of the aircraft. And it is this same effect of the air on the aircraft that determines whether or not it will fly. So IAS is directly and exactly correlated with the effectiveness of the flight surfaces.

TAS really doesn't matter one bit, except for navigation. That is why you calculate TAS (and adjust navigation accordingly), but measure IAS.

Vic.

[Herbert Meyer]

Re: Fly by wire ?

I am sure the Horsedrawn airplanes were much more reliable.

[Fungus Bob]

Re: Fly by wire ?

"I am sure the Horsedrawn airplanes were much more reliable."

Airplanes drawn by professional artists are much more reliable...

[JeffyPoooh]

Re: Nutters

Lars "...no longer a Boeing/Airbus difference."

Boeing's software typically includes some NASA-like 'Paranoid Programming Practices'.

Airbus's clearly doesn't. Far too many Airbus have been in excellent structural and generally flightworthy condition in the millisecond just before impact. Often, it's just some stupid little thing made worse by the computer plus crew. If there was a big net to catch it, then the fallen Airbus aircraft could be fished out, dusted off, computer rebooted, pitot tubes vacuumed out, manuals revisited, pilots spoken to sternly, software fixed, and dispatched immediately. They crash because of some stupid little thing, as nearly happened here.

Boeing accidents more often have serious damage before impact. Less often stupid glitches.

There's an indisputable difference caused by the software design approach, user interface, number of naive assumptions embedded in the code.

This post will result in heavy downvoting, but it's still true. Denial doesn't address the issue (or 'opportunity for improvements').

[Martin Gregorie]

Re: Nutters

Boeing's software typically includes some NASA-like 'Paranoid Programming Practices'.

Yeah, like running the flight-critical 737MAX MCAS system off a single computer connected to a single Angle Of Attack sensor. I know that the MAX carried two MCAS computers, each with its own AOA sensor BUT:

- only one MCAS computer and AOA sensor was active during a flight.

- each MCAS computer could only access its own AOA sensor

So, no cross-checking ability provided and the small red MCAS malfunction indicator was an optional paid-for extra, so not generally fitted because bean-counters.

IOW, Boeing was applying almost exactly the opposite of 'NASA Paranoid Programming Practices' as well as ignoring generally accepted aviation principles of multiple redundancy for known unreliable but vital sensors.

FWIW the aircraft I fly have no redundant flight instruments. However, part of the pre-solo training curriculum is to fly a circuit and land without either ASI or altimeter. I have subsequently done just that, but this time for real, when the ASI failed.

[Mi Tasol]

Re: Nutters

If you read the ATSB summary or report you will find the second take-off by Etihad was after the aircraft had reached V1.

Over simplifying, at that point the aircraft MUST lift off as the combination of weight and runway remaining makes that the only safe option. It is much much safer, even in this case, than a maximum energy abort that could end up off the end of the runway with brake temperatures well over 1500C.

The ATSB report is at http://www.atsb.gov.au/publications/investigation_reports/2013/aair/ao-2013-212/.

Air France 447 was caused because Airbus have installed inadequate pitot heating on many of their products starting with the A310 and then, in at least one model, ignore feedback from multiple customers until the regulators find out and force them to act.

[Vic]

Re: Nutters

Air France 447 was caused because Airbus have installed inadequate pitot heating on many of their products

I disagree with that.

AF447 happened because the pilots firstly did not follow procedure for suspected airspeed problems, and then compounded that error by failing to recover from a simple stall. This last was exacerbated because PNF couldn't keep his hand off the stick, even when asked directly about it...

Vic.

[Vic]

Re: Nutters

What I find worrying is that they tried to take off with faulty airspeed sensors. As if the Air France 447 accident never happened.

The AF447 incident was caused by the pilots failing to follow their training in several significant ways; the frozen pitot issue is merely what kicked off the incident chain. If that hadn't happened, somethig else would have.

Doing this on a modern fly by wire passenger jet is in the realm of suicidal

This is not true.

If the aircraft had passed V₁ when they had the warning, then not taking off would have been suicidal - in fact, it would have been criminally negligent, and had the pilots survived[1] the ensuing crash, they would have been culpable for all deaths and injuries sustained by their action. This is standard procedure - it's why you have a declared V₁, and why you have PNF calling it out as PF executes the take-off roll.

Dunno which deity do they have to thank

They have standard procedure to thank. They survived this incident because they followed it. AF447 crashed because the pilots did not.

Vic.

[1] This is quite unbelievably unlikely; an aircraft aborting after V₁ is most likely to smash nose-first into something that doesn't nove very well..

[akeane]

I'm torn...

On the one hand the existence of wasps proves there is no god, on the other hand they attack spiders...

[Simon 15]

Re: I'm torn...

Either that or they prove that if god (assuming christian) does exist he really just isn't *that* nice.... The genocide of the flood seems to support this supposition. I'm not sure how an all knowing being can keep making so many serious mistakes but then again he does work in mysterious ways!

[Anonymous Coward]

Re: I'm torn...

Australia was created by the old testament god, most of the rest of the world new testament.

[Simon 15]

Re: I'm torn...

Either that or they prove that if god (assuming christian) does exist he really just isn't *that* nice.... The genocide of the flood seems to support this supposition. I'm not sure how an all knowing being can keep making so many serious mistakes but then again he does work in mysterious ways!

Our fear of spiders seems to be irrational, they catch insects that we usually consider pests. I think people are really just jealous that they have so many legs.

[allthecoolshortnamesweretaken]

Re: I'm torn... / spiders

"Our fear of spiders seems to be irrational, they catch insects that we usually consider pests. I think people are really just jealous that they have so many legs."

Yeah, that's pretty much the thing I have about spders. And crabs, for that matter. Too many legs, and they move just wrong.

[Anonymous Coward]

Re: I'm torn...

Tell it to the White Anglo Saxon Protestants.

[JeffyPoooh]

Re: I'm torn...

"The Creator must have an inordinate fondness for beetles. He made so many of them."

- J.B.S. Haldane

"If one could conclude as to the nature of the Creator from a study of creation, it would appear that God has an inordinate fondness for stars and beetles." (alternate version)

[allthecoolshortnamesweretaken]

I didn't know that wasps eat spiders.

[Oengus]

The Wasps don't eat the spiders. They paralyse them with their sting and include them in the nest where they lay their eggs. The spiders are consumed by the wasp larvae when they hatch.

[Anonymous Coward]

Wonderful image over breakfast ☺

[Doctor Syntax]

"Wonderful image over breakfast"

The wasp larvae certainly think so.

[Triggerfish]

That's actually relatively nice, compared with some of the stuff the insect world get's up to. HR Giger would go away and have a cry.

Wasps are actually quite useful they hunt a lot of the things we consider pests. (Although they also hunt some of the things we do not as well).

[John Tserkezis]

"Wasps are actually quite useful they hunt a lot of the things we consider pests."

Too bad they can't be trained to hunt some very particular family members that are considered pests.

[Nathan 13]

Care!

Hope they double check that the covers are removed before flight. This sort of mistake has brought down planes before.

[Mark 85]

Re: Care!

The covers used to be a cloth tube (bright red or orange) with a "Remove before Flight" streamer. I remember back in the mid-60's that when a plane landed and while at the gate, they put those covers on. I'm guessing that someone figured it cost too much money to have some drag out a maintenance stand or ladder and put them on and take them off.

[Camilla Smythe]

Cunning Plan

Have some sort of internal mechanism that forces a pipe-cleaner or some other suitable rod up the pitot tube from the inside out to force any crud/wasps out of it.. or high pressure water/gas.

[Paul Crawford]

Re: Cunning Plan

Or coat the tips with some insect repellent every so often?

[bazza]

Re: Cunning Plan

From what I've heard most insects in Australia will happily gargle repellent before coming after you to get breakfast, lunch and dinner. Wanna avoid being eaten alive? Armour plate is nearly thick enough, but don't ever let one of the blighters get in...

[I ain't Spartacus]

Re: Cunning Plan

In the case of Australia and its wildlife, we should nuke it from orbit. It's the only way to be sure.

[Captain DaFt]

Re: Cunning Plan

"In the case of Australia and its wildlife, we should nuke it from orbit."

Australian wildlife always gets the hate, but what about South America? Way too much of the fauna there is quite happy to dig into you and feed on you while you're still alive.

The continent is a haven for horrible parasitic species that make politicians look almost benign. >shudder<

[bazza]

Re: Cunning Plan

That's because literally everything in Australia is nasty in some way. To borrow a line from Terry Pratchett, a list of harmless Australian animals would be as follows:

1) Some of the sheep

That's it. In South America there are at least one or two animals that aren't out to get you in some way or other as soon as you're on the same continent.

[Mi Tasol]

Re: Cunning Plan

Pratchett had a better name for the capital tooo.

Didyabringabeeralong is a far friendlier name than Cantberra (politicians leave out the t to fool the electorate into thinking that they are allowed to do some things.)

[6th]

Re: Cunning Plan

If we can survive down here this long what makes you think Nukes are gonna hurt us.

[Known Hero]

Re: Cunning Plan

have also posted further above, but is also relevant here :) Just add a windscreen wiper :)

http://www.theregister.co.uk/2012/08/07/laser_measures_airspeed/

[Anonymous Coward]

As a Pest Controller

i can assure you (in Blighty at least) that wasps are a spiders* food source.

Never seen a wasp with a spider but have seen plenty of wasps in spiders webs, all wrapped up ready for dinner.

*money spiders do not count!!!!

[Voland's right hand]

Re: As a Pest Controller

In Blighty too.

I regularly see wasps patrolling the spiderwebs in late summer and trying to take out the spiders in my garden. They are nowhere near as successful as some of the subtropical and tropical species (these are specialized spider killers), but still give it a go. Rather not surprising too - with all the pesticides being sprayed around, their primary food source (bees) is nearly extinct so they switch to other alternatives.

[jcitron]

Re: As a Pest Controller

We see them up where I live too and they can be a pain not because of stinging but due to where they put their mud daubs. Due to some vertigo issues I was unable to drive last summer.

When I finally got to drive, the locks on my Jeep were stuffed with mud daubs from the wasps.

[Anonymous Coward]

No such species

Just to be pedantic: in species names the second part should start out lowercase, not uppercase, thus the creature should be written as 'Sceliphron sphecidae'. However, there's actually no such species; Sceliphron is a genus within the Sphecidae family of wasps. So the correct way to refer to the creature would be 'Sceliphron sp.', indicating that it's 'some kind of mud dauber, but I don't know which one exactly'

[Simon Sharwood, Reg APAC Editor]

Re: No such species

I failed Latin. And biology. Actually I studied neither.

[Francis Boyle]

Shouldn't it be

'Sceliphron.*' if you don't know the species. (I never studied biology. The Christian Brothers thought it might lead to evolution.)

[Herbert Meyer]

plastic streamer

I hope there is a plastic streamer attached to the plastic cap over the pitot tube, that can be seen from the cockpit window, to ensure no plane tries to take off without removing the cap.

[Anonymous Coward]

Re: plastic streamer

I think you'll find that aviation has been aware of the possibility for years

http://www.panoramio.com/photo/37348192

In this case, it's the rapid work by the wasp that's the issue.

[Anonymous Coward]

White Anglo Saxon Protestants

At first glance, i thought that Kerry Katona and her mates had been on another beano to Marbella!

[deshepherd]

One of my sons is an avid watcher of "Air crash investigation" on NatGeo ... and I'm pretty sure one of the episodes he was watching involved a fatal crash where eventually they found the cause was incorrect air speed readings due to something like this.

[Updraft102]

Birgenair Flight 301.

[PassiveSmoking]

Not the first time

This sounds nearly exactly like Birgenair flight 301.

Sadly things didn't go so well in that case and nearly 200 people died. Good job these pilots were on the ball and able to get the plane back down (might have been a different story had it been night).

https://en.wikipedia.org/wiki/Birgenair_Flight_301

[DropBear]

Re: Not the first time

Well, you could always just install those covers they mentioned while stationary. Of course, then this happens instead:

https://en.wikipedia.org/wiki/Aeroper%C3%BA_Flight_603

[Anonymous Coward]

An age-old problem

This problem probably goes back as far the dawn of the pitot tune.

Back in the 60s (no that's not the dawn of the pitot tube) when I worked on Vulcans, we had a tech notice come around about the middle of May each year to the effect of controlling the 'june bug'.

Apparenty the 'june bug' was a tiny insect that liked laying its eggs in tiny orifices. One such orifice (and apparently a favourite of the 'june bug') was the static vent that was part of the airspeed system. We had to examine said vents each morning and wipe them with oil of citronella. I must admit I never saw any sign of these bugs - parhaps my citronella wiping skills were of an high order.

We also had to check and wipe the pitot tubes.

The Vulcan also had a couple of probes that stuck out perpendicular to the skin of the forward fuselage, one on each side. They were part of either the autoland system or the terrain following radar system (I can't remember) and I think were called 'air attitude sensors'.

They consisted of a stainless steel tub about 10-12cm long, with two rows of slots along their length. The slots faced forward and the two rows were about 60 degrees apart/

As the airstream moved over the tubes it would enter the slots, and depending of the attitude of the a/c there would be a greater pressure in one row of slots compared to the other. By the use of this differential, the tube would rotate the remove it, giving an indication of the nose-up/down situation to quite a high degree of accuracy. A blockage in one of these slots would be quite serious so these things had to be cleaned too.

To the poster who commented about the Concorde doing 500 at mach 2. Quite possible.

Mach number is constant and is compared to the speed of sound. However, the speed of sound changes with altitude. The higher you go, the lower the speed of sound. (that's why they use mach numbers - it's that static vent again).

[Marcelo Rodrigues]

Sharks with lasers!

1) Breed a tiny shark species.

2) Put a little 500mw laser on its head.

3) Build a (very) small pool inside the pitot tube.

4) Teach the shark to kill anything that comes inside.

5) Profit!

[Black Betty]

How about building a self test mechanism for blocked pitots.

Put a ball valve as the base of the tube, in one position normal operation, turn to apply compressed air to the open side of the tube. Test for overpressure, indicating a totally or partially blocked pitot tube.

And how about this for a totally bloody stupid idea?

Webcams cams cost a pittance, even pan-tilt-zoom cameras can be had for bugger all. So PTZ cams at the tips of each wing, horizontal and vertical stabilizers, pointing in at the plane. A selfie stick for looking at the underside. Fixed cameras aimed at critical systems inside cabinets, wheel wells etc.

Inside the cabin too. "What's going on back there?" is not a question any pilot should have to ask. Give him rear and side exterior views as well.

And finally computers smart enough to do most of his "seeing" for him. Chances are a computer would "see" another plane, or a drone long before a pilot might even if he was looking in the right direction at the right time.

The right software could spot the wasp doing its thing and alert the pilot or ground crew. With an exact, down to the last rivet, map of the plane's exterior (and major interior spaces) the slightest damage would be seen the moment it occurred. Stir in motion exaggeration algorithms and some polarising filters and you've potentially got real time vibration and stress maps of the entire airframe. MEMS sensors are cheap and small enough to be fitted to damned near every moving part, certainly the critical ones.

When the cost of installing a device to continuously monitor a mechanism is less than paying an expert to examine it once, then why the hell not? A computer doesn't have to know what the problem is, it simply has to be able to determine that a problem might exist and either scream for help or politely put up a flag and wait for attention.

That's the biggest selling point I think. Not safety, or general utility, but the degree of micro-maintenance it permits. Which should in turn extend the time between major services and complete teardowns enormously.

[graeme leggett]

Re: How about building a self test mechanism for blocked pitots.

You've just described a dozen more things to be added to an aircraft. that would need to be checked at maintenance intervals, that could go wrong and might lead to a plane/flight being grounded while investigated.

[Richard Pennington 1]

If a wasp stopped a plane in the USA...

... would they send in a SWAT team?

Checking my coat for unwanted insects.

[Dr Patrick J R Harkin]

Wasps force two passenger jets into emergency landings

And that's why I always fly Bee A.

I'm here all week, try the fish.