I mentioned this several weeks ago, suggesting that increased frequency of severe weather conditions may be taking the probes beyond their original time of certification design limits.


The system design should ensure that no set of redundant components can degrade to a single point of failure, yet this is exactly what appears to be happening in the case of the probes. For a function that is so critical in terms of other systems that depend on it's data, why has this problem not been fixed ?. An engineering design issue that needs only the will and resources to find a solution. ie: If there's not enough probe heat at limit, redesign with more effective heating, perhaps with better insulation on the outer surfaces to prevent heat loss. If the probes are corroding due to excessive heat on the ground, redesign the probe heat controller to prevent this. If the drain holes are blocking, redesign. If there is a need to make the drain holes bigger, correct for this in the adc lookup tables etc.



The crew may recover in most cases, but should never be placed in a situation where they need to if the problem can be engineered out. After reading this thread for several weeks now and getting a feel for the status quo, the apparent industry complacency about this issue just beggars belief, imnsho… :ugh:


Chris

Probes replacement on AF Airbus fleet?
 

The spokesperson of the SNPL union has back-pedalled yesterday evening :

Air France: faut-il changer les sondes Pitot fabriquées par Thales*? - Libération

The SNPL trade union, which had asked the replacement of the Thalès Pitot probes on the whole Air France fleet, moderated Wednesday this request after having learned that the incident on a A320 in July had lasted only a few seconds. The SNPL union which is the majority one within the company, asks for the change of all the Thalès probes “if it is not an isolated event and if, in fact, it is well the probes that are in question", specified Wednesday evening to the AFP its spokesperson, Erick Derivry. But “one does not change a fleet, even on such a significant subject, an incident which lasted seven seconds”, he added, by stressing that it was necessary to continue the study of the circumstances of the incident.

I wonder what could add or delete to this statement an AF sposkesperson...

So, I take it you know exactly where the ice is forming on or in the probe, and can tell Thales and Goodyear exactly what is wrong with their designs so they can fix them post haste? Or does "only needs the application of some engineering" mean that engineers can instantly design a fix for a problem that they don't understand? (Certainly tens of thousands of engineering managers have demanded this over the years, so clearly it must be a functional engineering approach?)

Or perhaps it is that people THOUGHT they knew how pitot probes worked, and THOUGHT they understood icing conditions at high altitudes, and did the best they could to design to those specs. And with some 70 years of positive experience without failures, it appears that they did a pretty good job. But now suddenly what has worked reliably for 70 years no longer works. Clearly this means it is a trivial engineering job to simply redesign for this never-before encountered (and thus unknown and not understood) change.

Maybe (if you aren't a member of the government, where everything can be accomplished by fiat) it means you first have to start by understanding the problem. And despire the engineers and scientists being ordered by management to provide an instant fix, maybe it does take some time to figure out what is different...

Pitot probe defects
 

Hi there.
These Pitot problems have been known for decades. In the late 80s and the early 90s, there was already problems with water/humidity accumulating in the hoose between the probe and the ADR leading to biased airspeeds on Airbuses (and the Boeings may have had their problems too ?). Then Airworthiness directives in the early 2000s to replace the former Sextan (now within Thales) and Rosemount probes by the new models of Thales and Goodrich probes were released, without lasting results, the Pitot obstruction problems reoccurred despite the replacement by the new probes. And the last case between Paris and Rome suggests that a new replacement by the new Thales Pitot probes will not make it. Then let's do what the pilot unions demand, let's buy the new Goodrich probes while Thales prepares a new generation of Pitot probes specifically designed to cope with *increased anti-icing specifications*: if the existing Thales probes already met these specifications (and even more) and if the process of certification does not guarantee that the probes operate in a satisfying manner in real life, then let's review/alter the specifications themselves. The climate change also possibly impacts the high altitude. Maybe they should also include a criteria which says that the performances met during certification should be maintained over a XX % performance level, over YY month, with a ZZ % probability ? (given a specified maintenance process). This would be a way to specify a probe resistance to gradual pollution and obstruction.
The Pitot probes have suffered drain hole defects in the past, partly due to the manufacturing process and susceptibility to corrosion, making the probe vulnerable to pollution.
Similar problems were also observed on the fleet of Mirage 2000:
http://www.defense.gouv.fr/defense/c...2004-001-A.pdf
See page 55.
Within the Mirage 2000 fleet, the Pitot drain pollution was observed on 30% of the Mirages, 60% in hot/humid areas. Seemingly, this drain gradual obstruction by varied pollutions could make the probe more vulnerable to icing conditions. It was 5 years ago.

Now on he airliners we have at least two available probes:
http://www.goodrich.com/portal/goodr...t%200851HL.pdf

https://www1.online.thalesgroup.com/...sor_probes.pdf
One is designed to improve its anti-icing performance, the other is not.
Roughtly, Goodrich probes are on 2 planes over 3, and Thales probes on 1 plane over 3. Cathay's large fleet of planes has not known any problem with Goodrich (Cathay's saying) while Air France is struggling to make a census of its numerous unreliable speed past cases. Why does AF feel the need to establish an historical sensus of the "unreliable airspeeds" cases ? Is it the only way to get the attention of the manufacturers or the regulatory bodies ?
Jeff
PS) History: Technical Standard Orders from 1948 & 1951 about Pitots
Current Technical Standard Order
Current Technical Standard Order

Re “… increased frequency of severe weather conditions may be taking the probes beyond their original time of certification design limits.”

Also, consider any changes in the way in which we operate, the routes and altitudes now flown, and perhaps lighter weights in search of economy, all of which could place the aircraft in high risk areas. Add to that, the way in which humans react to bad weather – ‘not been a problem previously’ – ‘it will be OK this time’, 'can deviate too far - must save fuel'.
And especially where the ‘new’ technology radars provide an excellent picture – except the radar might not be able to show some of the hazards, or crews may not have been taught or be able to recall the hazards, i.e. high altitude ice.

You assume much, http://images.ibsrv.net/ibsrv/res/sr...ies/shiner.gif but in defence, did say in the first sentence that the probes may be working outside their original design limits. The design changes may not be trivial but it is, irrespective of how many years of ‘research’ and committee it takes, an engineering problem with a solution. An extrapolation of an existing, well proven and reliable technology. Thought though, is not good enough. One would expect that there would be a process of continuous research and feedback to identify problems proactively. One incidence of a problem noted, several incidences starts to show a pattern. As for the probe heat controller, it’s not exactly rocket science to design one to prevent overheating.

It seems like there have been many reports of probe icing failures over many years. So how many incidences over time does it take before some sense of urgency is attached to the problem. 1, 5, 10 years worth, or what ?. Not a criticism of engineering, more one of process and I do get the feeling that those who should be driving this are in no particular hurry…

Chris

Missing Rudder Static Dissipator Tufts
 

In some of the photos of the rudder being recovered, it was noted that several static dissipator tufts were missing. Since these are quite flexible, it seems unlikely that impact with the sea would shear them off. Has there been any analysis or discussion of why they were missing?

Guyana wash up of parts
 

A week or so ago a report was mentioned of some seemingly aircraft parts washing up on some isolated beach in Guyana. Has anything further come of this, as I've seen
nothing since?
Thanks.

Richard

rgb - I think somewhere previously it was 'expected' in France 'in August'

The problem really seems to be lack of a graceful failure mode of the pitot system and the potential severity of pitot failures on a fly-by-wire aircraft. THAT is the problem will yield best results to an engineering fix. Pitot tubes need to work, but most of the discussion regarding fixing pitot tube problems is really a discussion on the best way to put a band aid on the problem.
The computers that use airspeed need to have a speed input that cannot be completely corrupted by failure of any one class of input. Loss of pitot data should be substituted with SatNav or Inertial speed data in a logical voting progression. Then when pitot data failed, a realistic substitute piece of information could be inserted. This is an engineering problem that engineers can solve. Not cheap ultimately, but compared to the AF447 accident cumulative costs, reasonable.
Sid

Simulating a transonic buffet for a commercial transportation aircraft
 

If the BEA needs some aerodynamical/flight mechanics data for the A330 in the transonic regime (say between Mach 0.9 and Mach 1) according to different aerodynamical configurations (spoilers deployed or not or failed, landing gears lowered or not,...), ELSA-chimera can help:
http://img197.imageshack.us/img197/9...spoilerm02.jpg
http://img33.imageshack.us/img33/745...spoilerm03.jpg
http://img33.imageshack.us/img33/174...tairliner2.jpg
PS) the overspeed of a Learjet 24 (and inconsistant procedures about the airbrakes deployment above MMO)
http://www.airdisaster.com/reports/ntsb/AAR82-04.pdf

rgbrock, as of July 21, still not definitively identified as coming from either AF 447 or an Ariane

Stabroek News - Hope beach find still not positively identified

Regarding the news today of extending the search several months, assuming they can scan about 400 sq km a day, they are now about 10 days into this second phase, and no indication they have found the wreckage yet. And assuming they started the scan in the area they thought most likely to be the impact point, that's rather disappointing. This would now be an area equivalent to a square about 65 sq km on a side that's been scanned without apparent result.

SaturnV:

10 days into an operation that could take months is not long. I'm holding hope that eventually they find something. Until then, they need to scan, scan, scan and more scan
until they find it. The wreckage is down there. Somewhere. And it can be found.
The money and time in doing so will be immense. But, IMHO, well worth the effort
and expense.

machinbird, what on earth can satnav or inertial tell you about airspeed, which is the only relevant criterion?

I see 'ssk' reports a not dissimilar IAS issue on a A320 . Are we seeing the same issues on the 737/757/777 ie is it a 'new' wetaher phenomonen or is this somehow ABI common software related?

rgbrock, the second phase, the sonar scan, is to take 4-6 weeks and cover 16,000 sq km. The area west of the track is very mountainous, with some peaks that appear to be 2,000 to 3,000 meters high. Large pieces of wreckage that are resting on a steep slope or are in a crevasse would probably present with a smaller vertical profile to a scan, if not altogether hidden entirely.

This may be a very formidable task.

Short answer: not new. not ABI. At all.

Scenario:Sound like a familiar speculation from this thread ? Yet it as actually a quick summary of my understanding of this crash:Over a decade ago, boeing, not fly by wire, and that is just one example, there are more.

There is just no need to invoke the "evil airbus / fbw removing control from the pilots". Bad IAS (from bad pitot or static data) is a killer, we know that from past experience, and it looks like it still is.

Hi stepwilk,
It would not be pure satnav velocity or inertial velocity. That is where the engineering comes in. You generate a computer model of the wind vector, and generate an airspeed validation system (computerized like everything else on the aircraft). The system is looking for any input parameter that is changing faster than preset rates determined by experience. The rates can be adjusted for condition of flight.
The key thing you are trying to avoid with the system is things like a sudden drop off to 0 airspeed or surge to Mach 1.5. Just trying to keep things in the ball park.
Now does it make sense?
Sid

stepwilk;

The use of GPS groundspeed as well as the use of GPS altitude are both short-term recommended procedures for the loss of pitot-static data. Many of us who fly these aircraft have known this for years and have probably quietly made up our own "response" to loss of pitot-static data by calmly accessing the GPS Nav data on the MCDU, (Multi-purpose Control & Display Unit) on the Airbus or its equivalent on the 777 etc.

Others have made the observation that groundspeed is not airspeed and of course this is an immediately known and appreciated fact by those who fly these aircraft, but the airplane isn't going to do anything aerodynamically dramatic (like lose/gain speed) just because airspeed data alone is lost. The aircraft was fine moments before the loss and so the key is to do nothing, except take over manually and ensure a safe flight path with unchanged thrust levels. Then assess the situation, bring up the GPS page, note the groundspeed and maintain it while the other crew member sorts the event(s) out.

What brought 447 down will almost certainly be a combination of factors/failures rather than one or another system loss, which, individually, would not be mission-critical failures.

Given this, and given the solutions can be imagined as per machinbird's comments, I think this is a promising and already-extant (in use, however informally) solution. Getting to the bottom of a rash of pitot failures would be the first in the triage sequence, however.

Alternative speed scales
 

Hi there. There is already an alternative speed scale, the BUSS (Back Up Speed Scale). Not true airspeed though. I don't know how it works, maybe it relies on the interplay between the pich & thrust to derive a scale ? Maybe that the airspeed scale, when it is failing, can be replaced by other usefull scales (for the crew decision) like scales displaying an aerodynamic authority with the aerodynamic margins ?
Jeff