Boeing hopes to have the 737 MAX back in the air by December. But a list of five major requirements issued by the European aviation regulator EASA lets one doubt that the time frame can be kept.
EASA’s checklist includes a number of issues that have been disclosed: the potential difficulty pilots have in turning the jet’s manual trim wheel, the unreliability of the Max’s angle of attack sensors, inadequate training procedures, and a software issue flagged just last week by the FAA pertaining to a lagging microprocessor. But the agency also listed a previously unreported concern: the autopilot failing to disengage in certain emergencies.
We will discuss the five issues below.
It is not clear if EASA will insist on all the points to be fixed:
“Any of these could significantly affect the return to service, but we don’t know if they are actually going to become requirements or are they just items for discussion," said John Cox, a former 737 pilot who is president of the aviation consulting company Safety Operating Systems.
As usual the regulators will not tell Boeing how to fix the problems. Whatever solution Boeing offers for those items simply has to comply with the general demands the regulations make.
Some of the listed items seem to require hardware changes that will have to be applied to all 737 MAX and maybe even to the older 737 NGs.
Manual trim
We discussed the trim wheel issue back in May:
The 737 MAX incident also revealed a problem with older generations of the 737 type of plane that is only now coming into light. Simulator experiments (video) showed that the recovery procedure Boeing provided for the case of a severe mistrim of the plane is not sufficient to bring the plane back under control. The root cause of that inconvenient fact does not lie with the 737 MAX but with its predecessor, the Boeing 737 Next Generation or NG.
…
- The smaller manual trim wheels on the 737 NG make it more difficult to trim a runaway stabilizer back into a regular position.
- The larger stabilizer surface makes it more difficult to counter a runaway stabilizer by using the elevator which was kept at the same size.
- 737 NG pilots no longer learn the rollercoaster maneuver that is now the only way to recover from a severe mistrim.
EASA listing the trim wheel issue is the first official recognition of this problem.
The manual trim via the trim wheels is a necessary backup for the electrical trim system which relies on only one motor. If the manual trim can not be used in certain parts of the allowed flight envelope, Boeing has a severe issue at hand.
A 2015 EASA safety finding, previously discussed here, accepted the 737 MAX only because Boeing said that the manual trim wheel was operational even at higher speeds and when the electric trim cuts out. It also promised that its training material would cover the issue.
It is now known that the manual trim, especially at higher speeds, may require more force than an average pilot can apply. The general issue and the difficulty is still not mentioned in the current Boeing training material.
The trim wheel problem seems to be an item where the U.S. regulator FAA and the European EASA disagree:
The FAA has also previously denied that the trim wheel — which is used to lift or lower a plane’s nose during an emergency — would cause delays.
It is hard to see how a manual trim system that, as Boeing told EASA, should be used where the electric system comes at its limits, can be acceptable without change, when it can not be moved in especially those cases where it should be used.
Angle of Attack Sensors
The second item on the EASA list is, as Bloomberg describes it, "the unreliability of the Max’s angle of attack sensors".
Angle of attack sensor
The two angle of attack sensors on the MAX are not inherently unreliably. They are external sensors that are prone to get damaged. The original Maneuvering Characteristics Augmentation System (MCAS) that led to the crash of two planes, relied on only one of the two sensors. When that sensor got damaged, likely by a bird strike, MCAS trimmed the plane nose down into the ground. Boeing will now use both sensors to control the MCAS system. This may however not be enough.
During start and landing planes can collide with a passing flock of birds. In such cases both AoA sensors could easily get damaged. Many will remember that US Airways Flight 1549 landed in the Hudson river because bird strikes disabled both of its engines.
More modern planes than the Boeing 737 have 3 or 4 angle of attack sensors. Some other Boeing plane types have systems similar to MCAS. In addition to the AoA sensors they use an inertial system, acceleration and absolute position sensor, to determine if their MCAS like system should react.
It is quite possible that the regulators will now require a third sensor to be used to control the MCAS on the 737 MAX. If that is the case Boeing will likely prefer to add an additional internal sensor box to the plane instead of a third external sensor.
Training
Boeing only provided a few electronic pages of training material for pilots switching from the older 737 NG to the 737 MAX. It omitted any mentioning of MCAS. The American Airlines pilot union voiced concern about the new training material and advice Boeing plans to provide:
“However, at APA we remained concerned about whether the new training protocol, materials and method of instruction suggested by Boeing are adequate to ensure that pilots across the globe flying the MAX fleet can do so in absolute complete safety,” [the president of Allied Pilots Association] said in his statement.
In a Congress hearing Captain 'Sully' Sullenberger, who saved flight 1549, demanded additional simulator training for new 737 MAX pilots:
"They need to develop a 'muscle memory' of their experiences so it will be immediately available to them in the future when they face such a crisis," Sullenberger said.
The British Civil Aviation Authority also called for more training. It issued a Safety Notice (pdf) on Flight Crew Training that relates to the MCAS incidents that brought the two 737 MAX planes down:
Over the last five years, there have been number of large commercial air transport aircraft accidents and incidents which were attributed to lack of awareness of the aircraft’s trim condition. Factors which contributed to loss of control in-flight were inappropriate trim inputs or mishandled automatic trim malfunctions, especially during a high energy state or at low altitude, which resulted in excessive elevator or stabiliser load forces.
A number of situations that pilots should be additionally trained for is listed:
- Automatic trim malfunctions, associated crew actions and implications of manual intervention and lack of awareness of the aircraft’s trim state. This should include strategies to recover from an out-of-trim condition after an automated system failure and various energy states at different altitudes
- The difficulty of manual trim intervention at high aerodynamic loads with applicable commercial air transport aircraft, particularly at lower altitudes and with consideration of crew coordination difficulties/techniques
The British Safety Notice is an indirect but strong hint to the FAA that it should demand extra simulator training for flying the 737 MAX.
Boeing however has signed 737 MAX sales agreements with Southwest Airlines and possibly also other customers that require it to pay back $1,000,000 per plane should new MAX pilots require additional simulator training. Southwest ordered a total of 292 of the 737 MAX type. 31 have been delivered so far.
Flight Control Computer
We discussed the "lagging microprocessor" problem in detail some ten days ago.
Each of the two Flight Control Computers (FCC) has two microprocessors. If one processor fails the other is supposed to take over. But the two processors already share some work. When FAA test pilots disabled one of them, the other one had too much to do and was too slow in reaction to the pilot's input. The simulator test flight ended "catastrophic". Boeing said it can make software changes to prevent processor overload. It is doubtful that a software fix is a solid solution without additional side effects.
Autopilot Disengage
The last item on the EASA list is "the autopilot failing to disengage in certain emergencies". This problem was not completely unknown. It is likely related to the lagging microprocessor.
In general the autopilot, which consists of a number of programs running on the Flight Control Computer, must disengage immediately when the pilot turns it off to fly the plane himself. When the pilot uses the electric trim switches on his steering column, because of a "runaway stabilizer" or some other unsavory trim condition, the pilot's signal must have priority over all other processes. But that does not work when the FCC is too busy.
The FCC delayed the tuning off of the autopilot and the manual electric trim signal the pilot gave did not go through.
A functional diagram of the electric trim system shows that the Main Trim Interlock, which is another program running on the FCC, must close the Main Trim Relay to allow electricity from the pilot's column switch to the stabilizer motor to flow through. The computer can effectively block the pilot input even in emergency cases.
source – bigger
When the FCC is overloaded it fails to stop the autopilot program immediately and it does not trigger the Main Trim Interlock process. The Main Trim Relay stays open and the electric circuit from the pilots column switch to the stabilizer motor never closes.
This works, like MCAS, "as designed" but is far from an optimal solution. It depends on the FCC to be fully functional without any delays. In a plane as old the 737 type, which is not constructed from the ground up for fly-by-wire, the pilot's input should have priority over all automatism. The electric circuitry of the horizontal stabilizer trim control system should be designed in a way that gives electrical priority to the pilot input without any need for an intervention by the FCC's digital magic.
The crash of the two 737 MAX planes revealed several issues with the plane and with Boeing as a company that were essentially caused by greed. Boeing did not want to construct a new plane to counter the announced Airbus 320 NEO. It remodeled the 737 NG and used the old type certification of the 737 to avoid greater costs. MCAS was a band-aid for a problem that required a solid aerodynamic solution. Engineers and test pilots were pressed to sign off on some shoddy management decisions. The FAA was not informed or asleep at the wheel. Developing and launching the MAX cost Boeing only some $2 billion. To build a new plane in the 737 class would likely have cost some $10 billion. The additional time needed would probably have cost Boeing some market share but with a modern platform it would have had a good chance to again catch up.
It is doubtful that all the above issues can be solved by the end of the years. Some 340 737 MAX were grounded in March. Boeing has since each month build 42 more of them. It will probably have to again reduce the build rate for lack of storage capacity. This will hurt not only Boeing, but also all its suppliers. Those hundreds of planes standing on the ground cost a lot of money. The typical lease rate for such a plane is up to $10,000 per day.
The total cost for Boeing of the 737 MAX accidents and their grounding due to the MCAS band-aid are now estimated at more than $10 billion. Boeing's outstanding shares value dropped from a peak of $242 to $200 billion. A lot of bad press is still to come as are the many lawsuits and investigations. Boeing's market share in the 737/320 segment will likely decline as at least some passengers will avoid that plane.
To revamp the 737 NG into the 737 MAX was the wrong decision. It would have been cheaper to develop a new plane.
As Capt. Sullenburger said in a recent interview: "Nothing is as costly as an accident."
—
Previous Moon of Alabama posts on Boeing 737 issues:
- Boeing, The FAA, And Why Two 737 MAX Planes Crashed – March 12 2019
- Flawed Safety Analysis, Failed Oversight – Why Two 737 MAX Planes Crashed – March 17 2019
- Regulators Knew Of 737 MAX Trim Problems – Certification Demanded Training That Boeing Failed To Deliver – March 29 2019
- Ethiopian Airline Crash – Boeing Advice To 737 MAX Pilots Was Flawed – April 9 2019
- Boeing 737 MAX Crash Reveals Severe Problem With Older Boeing 737 NGs – May 25 2019
- Boeing's Software Fix For The 737 MAX Problem Overwhelms The Plane's Computer – June 27 2019