AOPA position statement: Proposed Mitsubishi MU-2B AD

December 16, 1998

Position statement for proposed Mitsubishi Heavy Industries, Ltd. MU-2B icing airworthiness directive

FAA Public Meeting, Kansas City, MO
December 8, 1998

The Aircraft Owners and Pilots Association (AOPA), representing the aviation interests of more than 340,000 pilots and aircraft owners, presents the following position statement regarding the proposed airworthiness directive (AD) applying to Mitsubishi Heavy Industries, Ltd. MU-2B aircraft ( text | PDF). The proposed AD would require several modifications to the aircraft and engine operating systems for the stated purpose of “preventing departure from controlled flight due to undetected ice accumulation when flying in icing conditions that exceed the airplane’s ice protection capability.” AOPA objects not only to the specific provisions of this proposed AD, but also to its stated purpose.

AOPA is not aware of evidence indicating that the MU-2B is prone to undetectable ice accumulation. The fact of the matter is that ice detection by visual cues is significantly easier and more reliable on the MU-2 than most other aircraft in the general aviation fleet. Even if it were possible to develop visually undetectable ice accretions (which has never been demonstrated that we know of), no airplane should be flown in icing conditions that exceed its ice protection capability. For that matter, no aircraft should be flown outside its operating limitations or capabilities of any kind.

While there are numerous aircraft that are certificated for flight into known icing conditions, to our knowledge no airplane has ever been certified for continuous flight in severe icing conditions. The certification basis for aircraft approved for flight into known icing conditions clearly defines the continuous and intermittent atmospheric icing conditions against which the airplane must be protected. The distance over which these specific conditions must be tolerated is also expressed as a function of time. These conditions are based on air temperature, atmospheric liquid water content, and water drop diameter. In all cases, the recent Fact Finding Focused Special Certification Review (FFFSCR) conducted by the Federal Aviation Administration (FAA) reaffirmed that the MU-2B meets or exceeds all published certification requirements for flight into known icing.

However, the FAA does not appear to be satisfied that the current regulatory certification basis for flight into known icing is sufficient. As a result, an examination of the effect of ice accumulation in a super-cooled liquid droplet (SLD) environment was conducted during the FFFSCR. SLD icing performance is not part of the current regulatory requirements for known icing certification. However, these tests further demonstrated that the MU-2B was capable of remaining under positive control even under SLD icing conditions, allowing the pilot time to safely exit these conditions should they be inadvertently encountered. Further, there was no evidence of either ice contaminated tail stall or uncommanded roll response even with artificial SLD ice shapes attached to the control surfaces. Under even these extreme conditions, the FFFSCR revealed that controllability was maintained throughout approaches to landing with flaps extended. Nothing we can find in the FFFSCR report indicates that the Mitsubishi MU-2B has any ice detection, handling, or controllability faults that would warrant the series of modifications currently proposed by the FAA.

Many aircraft pilots and owners have inquired why the FAA appears to be singling out the MU-2 for its icing characteristics. This is occurring because the MU-2 icing performance characteristics appear to be equal, if not superior, to those of other comparable turboprop aircraft. An objective view leads many to believe that the conduct of the FFFSCR was a reaction to political pressure exerted by former U.S. Representative Tim Johnson of South Dakota, who called for an FAA review of the MU-2 after two completely unrelated MU-2 accidents. The first of the two accidents occurred as a result of a propeller hub separation that killed the governor of South Dakota. While tragic, this accident was in no way related to the icing characteristics of the MU-2 aircraft. The second accident occurred near Malad City, Idaho, in reported icing conditions. It was apparently considered irrelevant that this MU-2 was reportedly operating in icing conditions without functioning deice boots and numerous other deferred maintenance items. It was also apparently not considered important that the windshield (which had previously been subjected to hail damage) was never located at the very compact impact site. Despite all the confounding evidence, an icing SCR was conducted anyway, and to no one’s surprise, no significant faults were uncovered.

With the completion of the recent FFFSCR reaffirming the results of a previous SCR conducted in 1983, the MU-2 has been extensively shown on two occasions to be in full compliance with the regulations, policies, and procedures for icing certification at the time of the original type certification. Although the FFFSCR referenced 13 fatal accidents where icing conditions were either present or forecasted, the accident summaries themselves indicated no correlation between the accidents and a design flaw or failure to meet icing certification standards as a causal factor. Rather, the accidents appeared to result from pilot inexperience, lack of proper maintenance, operation beyond the design limitations of the aircraft, or a general lack of good judgment by the flight crew. The proposed icing AD addresses few, if any, of the causal factors identified in the cited 13 accidents.

AOPA maintains that there is insufficient justification for the proposed AD based upon the successful completion of the second icing SCR, and the apparent irrelevance of the accidents cited by the FAA to any specific problems encountered during icing conditions. The economic impact of this AD has been estimated to be well in excess of $27,000 for each aircraft in the U.S. fleet, and few of the specific proposed requirements have been linked to the prevention of any previous accident. The AD proposals take a broad brush approach to “improving” icing safety and are no more critical or relevant to the MU-2 than any other turboprop aircraft certificated for flight into known icing. Based upon this, AOPA strongly urges the FAA to significantly modify the proposed AD, limiting the actions to only those that have specific justification or merit. If not, the proposal should be withdrawn entirely. The following is a response to each of the six individual modification proposals contained in the AD:

Ice detector installation

This proposal would require the installation of a Rosemount ice detector probe on the upper right-hand nose of the aircraft. The system is designed to activate a cockpit enunciator light in the presence of ice. As an electromechanical device, it is prone to false indications and unannounced/undetected failure. The traditional detection of ice by normal visual cues is exceptionally good in MU-2 aircraft. In fact, ice is more readily visually detectable on this aircraft than virtually any other in the fleet. Ice is immediately visible on the windshield wiper post, propeller spinner dome, and tip tanks upon entering icing conditions. These indications are very difficult to miss by the pilot, and the detection process is consistent and not prone to failure unlike a mechanical device. To any pilot who has flown a Mitsubishi MU-2 in icing conditions, the addition of an ice detector probe is absolutely superfluous. The FAA has not established a significant link between an inability to detect the presence of ice in a timely manner with any accident. With a compliance cost of approximately $10,000 per aircraft, there is no evidence to justify the installation of this extraneous piece of equipment. AOPA maintains that this modification provision should be removed entirely from the proposed AD.

Trim motion alarm

This proposal would require the installation of an audible alarm that would sound in cruise flight when uncommanded trim motion is detected. This appears to be in response to the discovery of a nose-up trim setting in several of the accident aircraft. Earlier AD action has been taken to address what was previously believed by the FAA to be a runaway servo failure mode in the autopilot system. Again, to any pilot who has flown a Mitsubishi MU-2, this modification seems redundant because the MU-2 has a particularly large and obtrusive trim wheel that is adjacent to the pilot’s right knee. Any uncommanded trim change would be felt by the pilot as it is occurring and is virtually impossible to miss. For this reason, AOPA does not support this proposed modification and maintains that it should be dropped from the proposed AD.

Automatic autopilot disconnect

This proposal would require the addition of a system that would prevent the connection of the autopilot or would automatically disengage the autopilot below a given speed (approximately 140 knots for a long-body MU-2 and 135 knots for the short-body version) when the flaps are retracted. AOPA maintains that this is an unnecessary and potentially very dangerous proposal. Previous AD action requires that an airspeed in excess of 180 knots be maintained when in icing conditions. Since the inception of this requirement, AOPA is unaware of any accidents attributable to icing or uncommanded pitch-up. Further, we are deeply concerned with potential safety ramifications of an automatic autopilot disconnect. In our view, such a device raises the potential for a whole new set of human factors-related accident scenarios.

Under the proposal, the autopilot would automatically disconnect or could not be engaged during many of the highest workload phases of flight such as climb-out, approach, terminal airspace operations, etc. This is of particular concern since the MU-2 is most frequently flown with a single pilot. In some instances, the proposed disconnect would occur at an airspeed greater than best rate of climb, requiring the pilot to hand-fly the aircraft while copying clearances, following SIDs, programming flight management systems, and other high workload activities.

In our view, the FAA should study the human factors implications of an autopilot disconnect in single-pilot operations during high workload phases of flight. These effects must be fully understood before any operator is required to spend in excess of $10,000 to have such a device installed on his or her aircraft. AOPA is unequivocally opposed to this specific proposal.

Improved tail boot inflation verification

This proposal would require a modification to the existing system that verifies sufficient pressure is present for the inflation of both the main wing and tail deice boots. Currently, an indication is only provided for main boot pressure. AOPA agrees with the FAA that such a system could have potentially prevented a previous accident by warning the flight crew that there was a malfunction of the tail deice boots. While we do not feel that this modification warrants AD action and could be simply handled under the existing manufacturer service bulletins, monitoring tail boot inflation pressure does close the loop on a potential undetectable malfunction. For this reason, AOPA does not specifically oppose this provision of the AD. However, we do not believe that this alone is sufficient justification for the issuance of any AD against the MU-2 aircraft. This is especially true since the same undetectable condition could exist on numerous other aircraft types. We support the manufacturer service bulletins and encourage owners to comply but do not believe that this should be the subject of an airworthiness directive.

Auto relite/continuous duty ignition system

This proposed modification would require the installation of continuous duty time ignition boxes to replace the existing ignition systems that are generally limited to an operating period of five minutes. In addition, the AD proposes to require the installation of an auto relite system. AOPA can find no specific justification for requiring the installation of a continuous duty ignition system or auto relite system other than to meet the latest certification standards. We do however, accept the potential safety enhancement of having one or the other of these two systems. However, there is no evidence that this should be the subject of an airworthiness directive. In any event, the requirement to have both an auto relite system and continuous duty ignition boxes is redundant and unnecessarily costly. If the FAA insists on issuing an AD against the MU-2 ignition system, only one system or the other should be required.

Icing placard

It has always been the position of AOPA that placards should be saved for only the most safety critical information. They should not be used to state the obvious or outline standard procedures. By doing so, the importance of placards is watered down to the point that they go unread or unheeded. The information being proposed for the placard called out in this AD is already contained in the flight manual. There are many processes and procedures that a pilot should follow prior to each flight, and most of them are contained in flight manuals and checklists. Preflight procedures should remain part of these documents, not part of the instrument panel. In addition, unnecessary placards such as this take up the limited instrument panel space needed for real safety-enhancing items such as GPS, TCAS, and the soon to be mandated GPWS.


In summary, the Aircraft Owners and Pilots Association does not believe that the 1983 SCR, the 1997 FFFSCR, or the accident history of the MU-2B aircraft supports the issuance of an icing AD against the Mitsubishi Heavy Industries, Ltd. MU-2B. However, should the FAA insist on issuing a final AD in opposition to the weight of public comment, we feel that the AD must be scaled back to the few specific acceptable modifications outlined above.