US alters air crash report after French protest

WASHINGTON, Sept 24 (Reuters) - U.S. safety investigators have deleted parts of a report into the 1994 crash of a French-made plane in Roselawn, Indiana after protests from France, according to a letter made available on Tuesday.

The letter from the National Transportation Safety Board, dated Sept. 13, told plane-maker Avions de Transport Regional (ATR) and the head of France's civil aviation authority, DGAC, that it was changing contested language in the report into the Oct. 31, 1994 crash of an ATR-72 that killed all 68 people on board.

The crash was blamed on ice forming on the plane, but the NTSB said it was withdrawing language suggesting ATR and the DGAC already had detailed knowledge that ice could cause control ailerons to move in an uncontrolled way.

"The safety board wishes to clarify, by way of this response, that such an interpretation is unwarranted," the letter said.

But it did say it would have been prudent for ATR to examine prior incidents more carefully and that there was ample data that should have spurred the DGAC to require this additional work.

"ATR and the DGAC should have conducted further research and developed more effective means of preventing such incidents," the NTSB said.

It said it had decided to change its 1996 report into the crash of the turbo-prop commuter plane operated by American Eagle, the regional carrier for American Airlines, after considering petitions from ATR and DGAC, and counter petitions from the Air Line Pilots Association.

Efforts to reach an ATR spokesman about the NTSB letter were unsuccessful. ATR is a 50-50 planemaking joint venture between EADS (EAD.PA) and Finmeccanica (SIFI.MI).

In-Flight Icing Encounter and Loss of Control, American Eagle Flight 4184, Roselawn, Indiana

Chris, This is an absolute outrage. This is exactly the thing I am worried about that will happen to the sr111 report. What if the Swiss and the Americans for that matter (they certified and installed the IFEN) don’t care for the fact that the IFEN is blamed for the crash? I’m pretty sure that they won’t. It would be very easy for them as they go through the draft and pick it apart, (throw politics into the mix) to alter the final outcome. In fact one Swiss reporter that we were in touch with casually mentioned that there would be a big difference between the draft and the final report. Politics have no place in aircraft investigations. In the case of sr111 it is already suspicious. The manufacturers and all involved parties get to comment on the draft. As we speak they are scrutinizing it and are allowed to submit their comments. I’m just so sure that we will see the report the way it should be presented. The NTSB by doing this has almost done us a favor. They’ve made it very clear that accident investigations are a complete farce. Well if you ask me, you’ve heard it from the Swiss press the IFEN was responsible for the crash in the draft. Now let’s see what we get with the final report. If it doesn’t appear to be the same I think we as the victims of this terrible tragedy should be outraged. It is extremely important that we the families of sr111 see that draft report if we really expect to ever understand why our loved ones are dead.

Just wanted to add that I don't think Jim Hall would have done this. Isn't the new head of the NTSB from the FAA? If so I think that says it all. Pretty frightening.

Chris or Cecil, Mark has told me that he believes there may have been a counter-petition filed by pilots. Do either of you know where we could find a copy of that? We've already checked the NTSB site and have not located it. Thanks.

Barbara

Barbara...

The link that I provided below will take you to the NTSB abstract. There are two .pdf documents available from there - the final report (showing the original text and amended text) as well as a "Response to Petition for Reconsideration".

In the response, there is mention of submissions by ALPA and American Eagle disagreeing with the DGAC. The response runs to 37 pages and quotes from the ALPA submission in many sections.

Barbara... It is not unusual for Petitions for Reconsideration to be submitted to the NTSB after
investigation reports have been published. The key here is that the report IS published, and is
only amended AFTER each individual case has been reviewed. In the USA you will not see a
report amended through the �draft� process, but you do from time to time see the reports changed
after parties successfully petition the Board. I agree with you regarding the fear of change prior
to public release of the findings by the CTSB. I too am troubled over that.

Thanks Chris, here is something someone sent me regarding ALPA & this particular investigation:
http://www.alpa.org/internet/alp/1999/octHallSpeech.htm
During the investigation of the 1994 crash of an ATR 72 in Roselawn, Ind., ALPA actively supported the tanker icing tests that led to changes in the design of that aircraft and in our understanding of flight in icing conditions. ALPA assisted us with engineering expertise to develop factual information about the airframe icing-related problems with the ATR 42 and 72. You provided the Safety Board with valuable documentation that identified an icing problem with the airplane and expressed the concerns of your membership to the FAA about certification and operation of the airplane in icing conditions.
Just as that investigation was beginning, the Safety Board completed a safety study on commuter airlines, and concluded that they should be included under the same operating regulations as the major airlines. ALPA coined the phrase, "One Level of Safety," and ALPA's support of the NTSB's efforts enabled us to achieve significant changes in commuter air carrier regulations. I count passage of the new regulations that adopted this concept as one of the highlights of my years as chairman.



ALPA Supports NTSB Findings in American Eagle Accident
... News Release. Release #96.30. July 9, 1996. ALPA Supports ... Safety Board's findings
in the October 31, 1994 crash of American Eagle Flight 4184 near Roselawn ...
www.alpa.org/internet/news/1996news/NR96030A.htm - 3k - Cached - Similar pages
[ More results from www.alpa.org ]
NTSB Identification: DCA95MA001 . The docket is stored in the (offline) NTSB Imaging System.



14 CFRPart 121 operation of Air Carrier AMERICAN EAGLE AIRLINES

Accident occurred Monday, October 31, 1994 at ROSELAWN, IN
Aircraft:ATR ATR-72-212, registration: N401AM
Injuries: 68 Fatal.

The airplane was in a holding pattern and was descending to a newly assigned altitude of 8,000 feet when it experienced an uncommanded roll excursion and crashed during a rapid descent. The loss of ccontrol was attributed to a sudden and unexpected aileron hinge moment reversal that occurred after a ridge of ice accreted beyond the deice boots. The manufacturer failed to disseminate adequate warnings and guidance to operators about the adverse characteristics of, and techniques to recover from, ice-induced aileron hinge moment reversal events; and failed to develop additional airplane modifications, which led directly to this accident. The DGAC failed to require the manufacturer to take additional corrective actions, such as performing additional icing tests, issuing more specific warnings regarding the aileron hinge moment reversal phenomenon, developing additional airplane modifications, and providing specific guidance on the recovery from a hinge moment reversal, which led directly to this accident. The ability of the FAA to to monitor, on a real-time basis, the continued airworthiness of the ATR airplanes was hampered by the inadequately defined lines of communication, the inadequate means for the FAA to retrieve pertinent information, and the DGAC's failure to provide the FAA with critical airworthiness information., because of the DGAC's apparent belief that the information was not required to be provided under the terms of the Bilateral Airworthiness Agreement.

The National Transportation Safety Board determines the probable cause(s) of this accident as follows.

the loss of control, attributed to a sudden and unexpected aileron hinge moment reversal that occurred after a ridge of ice accreted beyond the deice boots because: 1) ATR failed to completely disclose to operators, and incorporate in the ATR 72 airplane flight manual, flightcrew operating manual and flightcrew training programs, adequate information concerning previously known effects of freezing precipitation on the stability and control characteristics, autopilot and related operational procedures when the ATR 72 was operated in such conditions; 2) the French Directorate General for Civil Aviation's (DGAC's) inadequate oversight of the ATR 42 and 72, and its failure to take the necessary corrective action to ensure continued airworthiness in icing conditions; and 3) the DGAC's failure to provide the FAA with timely airworthiness information developed from previous ATR incidents and accidents in icing conditions, as specified under the Bilateral Airworthiness Agreement and Annex 8 of the International Civil Aviation Organization. Contributing to the accident were: 1) the Federal Aviation Administration's (FAA's) failure to ensure that aircraft icing certification requirements, operational requirements for flight into icing conditions, and FAA published aircraft icing information adequately accounted for the hazards that can result from flight in freezing rain and other icing conditions not specified in 14 Code of Federal Regulations (CFR) Part 25, Appendix C; and 2) the FAA's inadequate oversight of the ATR 42 and 72 to ensure continued airworthiness in icing conditions. (NTSB Report AAR-96/01)



Full narrative available

Index for Oct1994 | Index of months



Mr. HALL. We are presently involved in an icing related accident that occurred at Monroe, Michigan. Testing is under way, and Dr. Loeb can bring you up to date, Congressman.

Mr. LOEB. We made a number of recommendations following the ATR–72 accident at Roselawn, Indiana. The report was completed last year. Some of those recommendations have been acted upon favorably by the FAA; and research has been undertaken.

Then as the Chairman indicated, the accident at Monroe occurred. It raises additional questions about icing. Some of them are different from those that were raised at Roselawn. In fact, some action has been taken on the particular airplane that was involved, an Embraer–120. But research is continuing regarding the central issues that go to the certification of the airplanes and the conditions under which untoward events could happen in icing.

There is a lot of research that is being done in conjunction with the FAA, and at NASA Lewis in Cleveland.

We expect to bring our report to the Board on the Comair accident—that will probably, undoubtedly, have a number of additional recommendations attached to it—this spring, but there is a lot of work left to be done. I think the FAA has taken a number of steps, but there is more to be done.

Page 34 PREV PAGE TOP OF DOC

AIRCRAFT ICING RESEARCH

Mr. PASTOR. What time frame do you have to get from the research and to the practical application?

Mr. LOEB. It depends, obviously, on the nature of the research. But when it involves wind tunnel testing or ice tunnel testing or ice tests behind a tanker, those projects take a long time to plan, develop and implement; and so we have been talking about this problem now for years. I hope that within the next year or two there are going to be some significant changes introduced by the FAA based on some of the research.

There are rubber boots on the leading edges of wings and, in fact, tail surfaces of a number of planes, so that you can actually expand them and they will pop the ice that forms on them.

The belief has been that you have to wait until you get a quarter to half inch buildup on them before you can pop them. Because if you don't it will build a bridge and you will never be able to pop them and so forth.

There is now work under way to determine whether, in fact, that is true or not. It was just a commonly accepted belief, but no one had ever verified it, no one had ever gone into a tunnel.

What we have learned is that when you do pop the boots, in between the boot cycling, you can get very, very small amounts of roughness from the icing that can create problems that had not generally been recognized and that many of the pilots, even those flying in the Midwest, just are not aware of.


Page 35 PREV PAGE TOP OF DOC

This is a piece of information learned from the current research. I think, as we finish the research that is going on, lots of good things will happen.

"I agree with you regarding the fear of change prior
to public release of the findings by the CTSB. I too am troubled over that."

It is extremely troubling. Speaking personally, no matter what the final report looks like I will be very skeptical that I am seeing all that I should that was in the original draft. I can't imagine why anyone would think this was a good idea. I wish I knew if there was some way for family members to file some sort of grievance but I'm pretty sure there is not. I don't think we can successfully protest Canadian law. If there was a way, believe me I would do it.

Cecil, thanks for your explanation regarding petitions. Do family members in the US get to see a draft prior to the release of the final report?

Barbara

Barbara.... "Do family members in the US get to see a draft prior to the release of the final report?"

I am not aware of any draft type reports issued by the NTSB. They do publish updates and progress reports, along with their public hearings, but these are available for public participation and distribution. There of course could always be confidential reports and fact finding statements that do not make their way into the public domain, but I believe that any probable cause explanation is always in "final" report form, and available to the public.

>I am not aware of any draft type reports issued by the NTSB. They do publish updates and progress reports, along with their public hearings, but these are available for public participation and distribution. There of course could always be confidential reports and fact finding statements that do not make their way into the public domain, but I believe that any probable cause explanation is always in "final" report form, and available to the public. <

Thanks for explaining that Cecil. From following a few of these tragedies since sr111, that's pretty much what I thought they did. There seem to be more updates issued here in the US throughout the investigation and the information seems to be more open and to the point. I have found it very frustrating getting any real information out of the TSB. As I've said before they are very pleasant to family members if you do call them but you end up frustrated that you don't know a darn thing more after speaking with them. Over time this has made me feel some resentment towards their system of secrecy. The last straw was this draft report which as I write is in the hands of those responsible but once again, I am left to guess and turn to other sources to obtain information. Who fits the description of 'affected party' more than the people that have lost loved ones in a tragic crash?

Barbara

In this week's Air Safety Week, there is a very good explanation for what occurred recently in the Roselawn NTSB post investigation:

Icing, Not Crew Incompetence, Caused Roselawn Crash, U.S. Board Says

French petitions did not produce substantive changes to original report
Basic conclusions will not be changed about the fatal 1994 crash of an ATR twin-turboprop in icing conditions while flying over Roselawn, Indiana.
The case is yet another where U.S. accident investigators find themselves at odds with foreign officials in an investigation involving multinational parties. The case also raises significant issues regarding the process by which foreign aircraft are approved (certificated) for operation in the U.S.
The National Transportation Safety Board (NTSB) released last week its response to French officials who three years ago had petitioned for a revision of the investigators' determination of the cause of the Roselawn accident (see ASW, Oct. 18, 1999 and Oct. 25, 1999). All 68 passengers and crew were killed when the aircraft, operating as American Eagle Flight 4184, plummeted from 10,000 feet into a field.
After an exhaustive investigation, U.S. investigators concluded that the accident airplane suffered an aileron reversal as the result of ice buildup on the wings. Further, the NTSB found that similar upsets had occurred before, giving warning for corrective actions that were not taken. The NTSB concluded that French aircraft manufacturer Avions de Transport Regional (ATR) was insufficiently aggressive in dealing with the problem, and that the French regulatory authority, the Directeur G�n�rale de l'Aviation Civile (DGAC), exercised "inadequate oversight" of the airplane's development and certification. The DGAC is France's equivalent to the U.S. Federal Aviation Administration (FAA). The NTSB report scored the FAA for similarly lax oversight.
Stung by the NTSB's blunt conclusions, the DGAC petitioned the NTSB to reconsider its conclusions. The FAA weighed in supporting the DGAC arguments. ATR followed with a petition of its own, using unusually blunt language that was tantamount to accusing the NTSB of incompetence. For added spice to the controversial brew, the Air Line Pilots Association (ALPA) added its voice, supporting the NTSB's conduct of the investigation and its conclusions.
Shortly after the ATR petition was submitted, the NTSB coolly advised that a response was "in staffing." That was in October 1999. Now, nearly three years later, the NTSB has issued its response. In what appears to be a nod to French sensibilities, the NTSB shifted its assessment of ATR and DGAC shortcomings to contributing factors in the tragedy. At the same time, the NTSB placed greater prominence on the short-comings in certification standards. Airplanes are routinely dispatched into icing conditions involving supercooled liquid droplets that they may not be able to handle. The certification standards do not cover this type of icing. On rare occasions, this condition can kill - the Roselawn accident is a prominent example.
Larger issues
In these regards, the NTSB has taken the occasion of its response to reiterate its concerns about a number of issues that transcend the particulars of this accident:
� The need to upgrade and expand certification standards for flying in conditions of supercooled liquid droplets.
� The need for improved sharing of information about precursor incidents and accidents which, if left uncorrected, can lead to needless loss of life.
� The need for greater rigor on the part of the FAA when it comes to approving foreign aircraft for operations in the U.S. The ATR 42 and the stretched version of this aircraft, the ATR 72, are examples. More recently, some board members have quietly expressed their concerns about FAA certification of Canadian and Brazilian-built regional jets. The first models of these aircraft featured wings with hard leading edges, that is, wings without leading edge slats. These wings can be vulnerable to in-flight icing, as was sadly revealed with the DC-9, the first model of which also featured a hard wing.
The NTSB's concern about arms-length FAA certification of foreign aircraft may have implications for the coming Airbus A380 double-deck very large transport aircraft (VLTA). This airplane will contain numerous advanced features, including a wing box built of advanced composite materials, increased use of electro-mechanical flight control systems, a 5,000-psi. hydraulic system, and numerous other features heretofore not seen in transport-category aircraft. Moreover, the A380 may pose unique challenges for emergency evacuation, for which the airplane also must be certified. Thus, NTSB concerns about the FAA's role in certifying the ATR 42 and 72 have implications when the A380 comes up for certification.
The Roselawn accident also illustrates the difficulty harmonizing opinions about crash causes when multinational parties are involved. This is the first of three cases in which the NTSB has taken exception to foreign views as to the cause of a crash. Here, the NTSB differs with the French in declaring that the circumstances of the Roselawn crash were prefigured by at least five earlier incidents. In the 1997 crash of a SilkAir B737, the NTSB disagreed with Singapore officials, maintaining that the pilot's actions most probably caused the crash. And, again, in the 1999 crash of an EgyptAir B767, the NTSB sharply disagreed with Egyptian officials. The Egyptians pointed to a mechanical problem with the horizontal stabilizers - the NTSB found no evidence of such and concluded that the relief pilots' actions put the airplane into its death-dive.
Controversy, not comity, has surrounded these three crash investigations. On the other hand, the NTSB has agreed wholeheartedly with the findings of the Transportation Safety Board (TSB) of Canada regarding the 1998 crash of a Swissair MD-11. The Canadians believe the plane was lost due to an in-flight fire, and NTSB officials have expressed complete agreement with this interpretation. They have passed the TSB's recommendations to the FAA, endorsing them fully.
In the Roselawn crash controversy, the NTSB's response matched in detail the lengthy French petition for reconsideration. While acknowledging the merits of certain points raised by the French, the NTSB stuck to the main themes of its original concerns.
The main issues
Item: Development and certification of the ATR 42 and 72
Petitioner's arguments: "No hinge moment activity comparable to what occurred at Roselawn was ever encountered during ATR 42 or 72 development or certification flight tests."
NTSB response: "Information conveyed to Safety Board investigators by ATR engineers ... clearly indicates an awareness early in the history of the ATR 42 and 72 airplanes that they were subject to hinge moment shifts [that] could be introduced by asymmetrical ice buildups in front of the ailerons
"In response to questioning about the SPS (stall protection system) thresholds, (ATR's chief engineer) indicated that the setting of those thresholds was aimed at eliminating both basic wing stalls and what he referred to as ... 'separate phenomena' ... It does not follow, however, that these activities indicate an awareness of the potential for hinge moment shifts ... below stall thresholds like the event at Roselawn. The pertinent point is only that the general phenomenon was understood and the special sensitivity of the ATR 42 and ATR 72 flight control arrangements is part of the backdrop against which subsequent consideration of ATR icing events occurred."
Item: Previous ATR 42 icing/roll upsets
Petitioner's arguments: "The petitioners assert that the previous ATR 42 icing/roll incidents discussed in the Safety Board's report were not precursors to the Roselawn accident and the DGAC responded properly to each of them."
NTSB response: "To the extent that the petitioners are asserting that one or more of the prior incidents did not involve uncommanded aileron deflections, they are incorrect. Although, in each of the prior ATR 42 incidents cited in the Safety Board's report, ice accretion led to asymmetric lift loss, it is also a fact that in each incident, ice-induced flow separation also caused aileron hinge moment shifts that resulted in uncommanded rudder deflections. These ... deflections are evident from the FDR (flight data recorder) plots ... each of which shows aileron deflections in a direction opposite (that) needed to promote recovery from the roll ... Further, the FDR data shows that the ailerons deflected at a high rate, in some cases exceeding the deflection rate capability of a pilot or any airplane system.
"The continued occurrence of roll departures aggravated by uncommanded aileron deflections ... during icing encounters ... should have indicated to ATR and its regulators that the corrective action taken after earlier incidents ... were inadequate."
Item: Initial certification
Petitioner's arguments: "The DGAC and ATR disagree with the Safety Board ... which characterizes FAA's reliance on the DGAC during initial certification of the ATR 42 and 72 as 'excessive' and as 'not in the best interest of safety.' "
NTSB response: "The Board appreciates the FAA's clarification regarding the purely administrative purpose of the meetings at which the FAA was represented by only one employee and its assurance that the agency's technical review of certification data occurred prior to those meetings.
"Nonetheless, the Safety Board remains concerned about the apparently limited nature of FAA's involvement. The Board questions whether a technical review that takes place so far from the manufacturer's facility, engineers, and design experts would be as thorough and probing as one that takes place with those experts present or readily available to answer questions and provide explanations."
Item: Continuing airworthiness
Petitioner's arguments: "The DGAC claims that, contrary to the Safety Board's (original])probable cause statement, the DGAC did not 'fail to provide the FAA with ... information developed from previous ATR incidents and accidents in icing conditions.' "
NTSB response: "The lack of a requirement for the exchange of information ... raises concerns about the scope and effectiveness of the BAA (bilateral airworthiness agreement).
"In connection with its investigation of the Roselawn accident, Safety Board investigators requested all ATR 42 and 72 incident and accident reports from a variety of sources, including ATR, FAA, AMR [American] Eagle, ALPA (Air Line Pilots Association) and NASA (National Aeronautics and Space Administration). The Board found that the FAA had little of the pertinent information in ATR's possession that was eventually instrumental in identifying and developing an interim solution for the problem of the airplane's aileron instability in certain icing conditions."
Item: Role of crew performance
Petitioner's arguments: "Petitioners assert that the Safety Board did not adequately address the role that crew performance played in the accident ... In particular that ... the crew was unaware of the severe icing conditions they were operating in and was unprepared to respond effectively to the upset when it occurred. The DGAC believes the captain should not have left the cockpit."
NTSB response: "Flight 4184 remained in a holding pattern for 39 minutes, as instructed by air traffic control. During that time ...the pilots recognized that they were experiencing some icing ... 'I'm showing some ice now' (about 9 minutes before the initial upset) and 'We still got ice' (about 2 minutes before the initial upset).
"The crew cannot be faulted for holding with flaps 15�. Neither the ATR 72 flight manual, which was approved by the DGAC, nor the FAA addressed the use of flaps in icing conditions. Holding with flaps 15� would provide a more desirable stall margin than the flaps 0� configuration. Similarly, ATR's 1992 All Weather Operations brochure did not recommend against flap extension in icing conditions.
"Further, the Safety Board continues to believe that ... it was not inappropriate for the captain to choose to leave the cockpit temporarily ... to use the lavatory. The flight was in a holding pattern ... The upset occurred several minutes after the captain's return.
"Because pilots at the time of the Roselawn accident were not informed about the possibility of ice-induced aileron hinge moment shifts and the effects of the resulting uncommanded aileron deflections, the Safety Board continues to believe that the upset would not have been recoverable by the average line pilot."
John Dow, a retired FAA expert on in-flight icing who participated in the investigation, the testing and the special certification review surrounding the Roselawn accident, placed a greater emphasis on the crew's shortcomings. "The digital flight data recorder shows the crew pulling on the control wheel without any coordination. The unedited version of the cockpit voice recorder should be mandatory listening in crew resource management training of what not to do," he said. With respect to the airplane's vulnerability to icing, Dow said manufacturer ATR "did a great deal of work in freezing rain-sized drops."
"I was astonished that the airplane flew with what they put on it in terms of ice shapes," he said, adding, "ATR did run into freezing drizzle during certification. I have photos."
"The difference in the outcome? The flaps were not extended for starters," he declared. Dow does not agree with the emphasis the NTSB places on the aileron hinge moment reversal aspect of the case. It caused the roll of the ATR 72 and was not the normal consequence of wing stall as in the ATR 42 prior events, he believes.
"When I teach this accident, I say the airplane rolled 77� right wing down because a sharp-edged ridge of ice probably only forward of the ailerons accreted in icing conditions beyond the certification envelope because the crew held with the flaps down," he said. "The airplane started correcting itself and got to 55� right wing down. What happened form there to the ground is a crew issue."
Dow's views may give comfort to the French, but after three years "in staffing," the NTSB response clearly takes a different view - average pilots could not have regained control of the aircraft, and prior incidents foreshadowed the disaster. The revised probable cause is more of an explanation of what happened, not why it happened, and it lowers the French and FAA role in the causal chain. But to the NTSB, the facts of the case remain unchanged. (ASW note: full text of the NTSB response is available at http://www.ntsb.gov/publictn/1996/AAR9601.htm) Dow, e-mail jdowsr@aol.com
'J'Accuse!' The French Charges
� "Ice-induced lift loss vs. aileron hinge moment shift ... are entirely different phenomena ... The NTSB's blurring of these phenomena is inexcusable."
� "It is also stunning that the NTSB also ignores its own earlier investigations of prior incidents, which did not find the Roselawn aerodynamic phenomenon."
� "The NTSB ignored critical information concerning the SCR (special certification review) team findings [which directly refute NTSB] erroneous conclusions regarding the development and certification of the ATR 42 and ATR 72 series aircraft." Source: ATR
Hinge Moment Reversal Explained
"The ATR control system uses standard cables and manual force to move the control surfaces. Unlike jetliners, ATR models have no hydraulically boosted primary flight controls. To remove some of the control wheel forces from the pilot, designers included surfaces that extend forward of the control-surface hinge line called aerodynamic balance horns (ASW note: servo tabs and offset hinges also are part of the design).
"Should the air pressure on the control surface be removed or reduced for any reason, such as a detached or disrupted airflow caused by ice, the effect of the horns would require pilots to place force opposite to what would be expected to move the controls. This phenomenon is known as hinge moment reversal.
"When hinge moment reversal occurs, the pilot feels the force turning the wheel without his or her input and has to hold back the wheel against the pressure by applying strength in the opposite direction ... When that happens, it tends to occur with a rapid onset." (ASW note: Some experts familiar with ATR aircraft maintain that the control wheel forces during hinge moment reversal are manageable)
From: Unheeded Warning, The Inside Story of American Eagle Flight 4184, by Stephen Fredrick, McGraw Hill, 1996, p. 111 (Note: Frederick is a former ATR pilot for American Eagle) Fredrick, tel. 920/876-3553
Evolution of Probable Cause
Extracts of the six-year controversy over the fatal 1994 crash at Roselawn, Ind., of an American Eagle ATR 72 twin turboprop
The NTSB's original probable cause (issued July 9, 1996):
"The probable causes of this accident were the loss of control, attributed to a sudden and unexpected aileron hinge moment reversal that occurred after a ridge of ice accreted beyond the deice boots because: (1) ATR failed to completely disclose ... information concerning previously known effects of freezing precipitation on ... stability and control characteristics ... when the ATR 72 was operated in such conditions, (2) the [DGAC's] inadequate oversight ... [and, contributing to the accident] the FAA's inadequate oversight of the ATR 42 and 72 to ensure continued airworthiness in icing conditions."
What French manufacturer ATR sought (Sept. 22, 1999):
"The probable cause of this accident was the lack of recovery of the aircraft from its loss of control, attributed to a sudden and unexpected aileron hinge moment shift that occurred after a ridge of ice accreted beyond the de-icing boots following a prolonged operation of the aircraft in hazardous icing conditions well outside the certification envelope for which no airplane is certified."
The NTSB's revised probable cause (Sept. 13, 2002):
"The probable cause [all changes to the original in italics] of this accident was the loss of control, attributed to a sudden and unexpected hinge moment reversal, that occurred after a ridge of ice accreted beyond the deice boots while the airplane was in a holding pattern during which it intermittently encountered supercooled cloud and drizzle/rain drops, the size and water content of which exceeded those described in the icing certification envelope. The airplane was susceptible to this loss of control and the crew was unable to recover.
"Contributing to the accident were (1) The French Directorate General for Civil Aviation's (DGAC's) inadequate oversight of the ATR 42 and 72, and its failure to take the necessary corrective action to ensure continued airworthiness in icing conditions; (2) DGAC's failure to provide the FAA with ... information developed from previous ATR incidents ... (3) The FAA's failure to ensure that aircraft icing certification requirements ... adequately accounted for the hazards that can result from flight in freezing rain; (4) FAA's inadequate oversight ... and (5) ATR's inadequate response to the continued occurrence of ... icing/roll upsets."
Sources: NTSB, ATR