Inside the Effort to Fix the Troubled Boeing 737 MAX
Via The Wall Street Journal
By Scott McCartney
After takeoff, the Boeing 737 suddenly warns pilots that the plane is about to lose lift and stall, an erroneous signal from a bad sensor. The control column shakes, loudly. Pilot Roddy Guthrie diagnoses the problem-and then the plane’s nose suddenly pitches down, on its own. Emergency No. 2.
He pulls back on the control column to keep climbing and gets the airplane back to the proper orientation, nose up. But it happens again, with more force. And then a third time, with even more force, so that he’s looking almost straight down at the ground-the most terrifying sight for any pilot.
The episode, a repeat of the system failure suspected in two Boeing 737 MAX crashes, takes place in a Boeing full-motion flight simulator Wednesday morning. A few minutes later, Capt. Guthrie and another pilot try again, this time with Boeing’s proposed software fixes installed-software that’s critical to Boeing, airlines and travelers world-wide.
Attorney Nomaan K. Husain of Husain Law + Associates has filed a lawsuit against Boeing for the Ethiopian Airlines Flight 302 Boeing 737 Max crash. If you have been affected by this crash, don’t wait a moment longer. Contact us today so we can review your case! Call 713-987-7126 or click here to send us a message!
Capt. Guthrie, fleet captain for the 737 at American Airlines, faces the same problem with the faulty sensor. This time the stick shaker gives its same loud warning, the result of the faulty sensor, but the 737’s computer never points the airplane down. A yellow caution light comes on and he easily maintains control.
“You don’t see compounding emergencies going on,” Capt. Guthrie says. “The second emergency never occurs. You’ve eliminated a major distractor.”
He and other pilots at American and other airlines have been voluntarily advising Boeing on fixing its troubled model of the 737, grounded in March after crashes in Indonesia and Ethiopia killed a total of 346 people. Wednesday morning’s session was the first time he checked out the changes in a full-motion simulator. (He had tested it previously in a stationary cockpit simulator in Seattle.) Boeing is running actual flight tests of the fix, with regulators world-wide scrutinizing the results.
The future of the 737 rests on the fix. Boeing drew loud criticism for failing to inform pilots and the Federal Aviation Administration about all aspects of its now-suspect system when it launched in 2017 and its reticence to act after the crashes. The company must win back the trust of regulators, then airlines, pilots, flight attendants and the traveling public.
Many hurdles lie ahead. Though the proposed fix has performed well in tests like Wednesday’s simulator session, testers could discover new problems. It’s too early to declare the MAX fully repaired.
Acting FAA administrator Dan Ewell said in May that October might be too soon for the plane to return to service. In addition to the FAA’s review, an outside agency established by the FAA will also review the proposed fixes, and regulators in Europe and Canada have said they planned to conduct separate reviews.
Sessions like this have convinced American it needs to put its pilots through additional training related to maintaining control of airplanes when systems fail, whether the FAA requires this or not. (The FAA says it hasn’t decided yet on training requirements.) Once the FAA signs off on changes, it will take 45 days for American to train all 4,200 of its 737 pilots. The airline figures that pilots will be the key to restoring faith in the airplane. They need to be convinced it’s safe before passengers will be willing to board the 737 MAX again.
American requested Boeing to allow me to ride along with Capt. Guthrie and Capt. Alan Johnson, the airline’s 737 fleet training and standards manager, as they put the new version of the maneuvering characteristics augmentation system, or MCAS, through its paces.
MCAS was added to the flight control computer to help the MAX, which has new engines, handle like previous models of the 737, particularly at slow speeds and when the nose is pitched high. When functioning normally, MCAS helps prevent the airplane from getting so nose-high that the wings stop producing lift and stall. When an airplane stalls, it becomes a rock.
I’m a private pilot with about 1,000 hours in my logbooks, including time in planes this size and larger. I flew the MAX simulator with the prior MCAS software, known as version 11.1, and the proposed fix, version 12.1.
I found it to be a night-and-day difference. Even though I knew what was coming and had seen two veteran pilots demonstrate proper recovery procedures, I struggled to maintain control with the old system. MCAS pushes the nose down with such cumulative force if you don’t use the exact prescribed procedures that as hard as you pull back on the control column, you can’t get the nose pointing up. “It’s not a good feeling,” Capt. Guthrie says.
With the new version, the problem with the faulty sensor in this test appears to become, essentially, a routine emergency-something pilots can handle while getting the airplane to the nearest suitable runway. MCAS doesn’t fire at all, leaving you with just the initial problem of the single faulty input.
“You just shut the faulty system down and the airplane still operates normally. It’s very manageable,” Capt. Johnson says of the MAX with version 12.1. “It’s like taking away your cruise control in your car.”
MCAS takes information from a single vane near the nose of the airplane. The 737 has two angle-of-attack vanes, but version 11.1 of MCAS only used one at a time. If that vane were broken, the system could activate erroneously and angle the plane’s nose down, thinking it was preventing a nose-high stall.
Fix No. 1 to MCAS is that in version 12.1, it will take input from both sensors. If they disagree by 5.5 degrees or more, MCAS will shut down and not fire at all.
With the troubled version of MCAS, pilots would instinctively pull back on their control column-or yoke-to keep the nose up when MCAS is pushing it down. And they’d use a switch on the control column to trim the airplane to a desired orientation. The trim switch can counteract the inputs from MCAS.
But once you release the trim switch, MCAS 11.1 fires again after five seconds because the erroneous data from the broken sensor remains. And it will keep firing five seconds after you stop moving the trim-the effect can be cumulative, so each time the nose points down more sharply.
Fix No. 2 to MCAS is that the system will only fire once.
Under the original version, Boeing assumed that pilots would diagnose the repeated downward pitch movements as “runaway trim,” something they used to train for more frequently. Runaway trim in modern planes is rare-American had one incident in its last 750,000 737 flights. Even Capt. Guthrie and Capt. Johnson say they had to brush up on speed-trim system functions after the Lion Air crash into the Java Sea in October.
The longstanding prescribed procedure for runaway trim is fairly simple: stabilize the aircraft, then disconnect the system by flipping two toggle switches called the Stab Trim cutout, short for stabilizer trim, that are on the pedestal between pilots just below the throttles. Then you use a wheel beside each pilot’s knees to crank in necessary trim-to move the stabilizer manually as needed.
The Ethiopian pilots did cut the stabilizer trim off, investigators have said. But with multiple warning alarms sounding and multiple emergencies at hand, they didn’t control the speed of the airplane. The plane can get going so fast that a pilot can’t move the wheel-think of the force of the wind when you stick your hand out a car at 35 miles an hour, then think of what it might be like at 350 mph.
Testing this in the simulator, Capt. Guthrie found he couldn’t move the wheel unless I let go of the control column I was pulling back to try to keep the nose up. Relaxing the column momentarily pitched the nose down even more, but removed enough pressure to crank in some trim.
That procedure of relaxing pressure to add trim repeatedly-sometimes called a roller coaster maneuver or a fishing maneuver, may be added to standard procedures for runaway trim, the American pilots say.
With the fixes in place, the faulty sensor still triggers the stick-shaker stall warning. But unless both sensors break exactly the same way within nanoseconds of each other, MCAS never fires because the angles reported by each sensor will differ. Boeing will also include a message on pilot displays when the readings differ as a standard feature.
And if MCAS should somehow erroneously fire, it will only pitch the plane down once, never moving it so sharply that it exceeds the ability of pilots to pull back on the control column and get the nose up.
Capt. Guthrie says his testing finds version 12.1 eliminating the problems of version 11.1. “We are, at this point, quietly confident,” he says.
Attorney Nomaan K. Husain of Husain Law + Associates has filed a lawsuit against Boeing for the Ethiopian Airlines Flight 302 Boeing 737 Max crash. If you have been affected by this crash, don’t wait a moment longer. Contact us today so we can review your case! Call 713-987-7126 or click here to send us a message!
oudly. Pilot Roddy Guthrie diagnoses the problem-and then the plane’s nose suddenly pitches down, on its own. Emergency No. 2.
He pulls back on the control column to keep climbing and gets the airplane back to the proper orientation, nose up. But it happens again, with more force. And then a third time, with even more force, so that he’s looking almost straight down at the ground-the most terrifying sight for any pilot.
The episode, a repeat of the system failure suspected in two Boeing 737 MAX crashes, takes place in a Boeing full-motion flight simulator Wednesday morning. A few minutes later, Capt. Guthrie and another pilot try again, this time with Boeing’s proposed software fixes installed-software that’s critical to Boeing, airlines and travelers world-wide.
Capt. Guthrie, fleet captain for the 737 at American Airlines, faces the same problem with the faulty sensor. This time the stick shaker gives its same loud warning, the result of the faulty sensor, but the 737’s computer never points the airplane down. A yellow caution light comes on and he easily maintains control.
“You don’t see compounding emergencies going on,” Capt. Guthrie says. “The second emergency never occurs. You’ve eliminated a major distractor.”
He and other pilots at American and other airlines have been voluntarily advising Boeing on fixing its troubled model of the 737, grounded in March after crashes in Indonesia and Ethiopia killed a total of 346 people. Wednesday morning’s session was the first time he checked out the changes in a full-motion simulator. (He had tested it previously in a stationary cockpit simulator in Seattle.) Boeing is running actual flight tests of the fix, with regulators world-wide scrutinizing the results.
The future of the 737 rests on the fix. Boeing drew loud criticism for failing to inform pilots and the Federal Aviation Administration about all aspects of its now-suspect system when it launched in 2017 and its reticence to act after the crashes. The company must win back the trust of regulators, then airlines, pilots, flight attendants and the traveling public.
Many hurdles lie ahead. Though the proposed fix has performed well in tests like Wednesday’s simulator session, testers could discover new problems. It’s too early to declare the MAX fully repaired.
Acting FAA administrator Dan Ewell said in May that October might be too soon for the plane to return to service. In addition to the FAA’s review, an outside agency established by the FAA will also review the proposed fixes, and regulators in Europe and Canada have said they planned to conduct separate reviews.
Sessions like this have convinced American it needs to put its pilots through additional training related to maintaining control of airplanes when systems fail, whether the FAA requires this or not. (The FAA says it hasn’t decided yet on training requirements.) Once the FAA signs off on changes, it will take 45 days for American to train all 4,200 of its 737 pilots. The airline figures that pilots will be the key to restoring faith in the airplane. They need to be convinced it’s safe before passengers will be willing to board the 737 MAX again.
American requested Boeing to allow me to ride along with Capt. Guthrie and Capt. Alan Johnson, the airline’s 737 fleet training and standards manager, as they put the new version of the maneuvering characteristics augmentation system, or MCAS, through its paces.
MCAS was added to the flight control computer to help the MAX, which has new engines, handle like previous models of the 737, particularly at slow speeds and when the nose is pitched high. When functioning normally, MCAS helps prevent the airplane from getting so nose-high that the wings stop producing lift and stall. When an airplane stalls, it becomes a rock.
I’m a private pilot with about 1,000 hours in my logbooks, including time in planes this size and larger. I flew the MAX simulator with the prior MCAS software, known as version 11.1, and the proposed fix, version 12.1.
I found it to be a night-and-day difference. Even though I knew what was coming and had seen two veteran pilots demonstrate proper recovery procedures, I struggled to maintain control with the old system. MCAS pushes the nose down with such cumulative force if you don’t use the exact prescribed procedures that as hard as you pull back on the control column, you can’t get the nose pointing up. “It’s not a good feeling,” Capt. Guthrie says.
With the new version, the problem with the faulty sensor in this test appears to become, essentially, a routine emergency-something pilots can handle while getting the airplane to the nearest suitable runway. MCAS doesn’t fire at all, leaving you with just the initial problem of the single faulty input.
“You just shut the faulty system down and the airplane still operates normally. It’s very manageable,” Capt. Johnson says of the MAX with version 12.1. “It’s like taking away your cruise control in your car.”
MCAS takes information from a single vane near the nose of the airplane. The 737 has two angle-of-attack vanes, but version 11.1 of MCAS only used one at a time. If that vane were broken, the system could activate erroneously and angle the plane’s nose down, thinking it was preventing a nose-high stall.
Fix No. 1 to MCAS is that in version 12.1, it will take input from both sensors. If they disagree by 5.5 degrees or more, MCAS will shut down and not fire at all.
With the troubled version of MCAS, pilots would instinctively pull back on their control column-or yoke-to keep the nose up when MCAS is pushing it down. And they’d use a switch on the control column to trim the airplane to a desired orientation. The trim switch can counteract the inputs from MCAS.
But once you release the trim switch, MCAS 11.1 fires again after five seconds because the erroneous data from the broken sensor remains. And it will keep firing five seconds after you stop moving the trim-the effect can be cumulative, so each time the nose points down more sharply.
Fix No. 2 to MCAS is that the system will only fire once.
Under the original version, Boeing assumed that pilots would diagnose the repeated downward pitch movements as “runaway trim,” something they used to train for more frequently. Runaway trim in modern planes is rare-American had one incident in its last 750,000 737 flights. Even Capt. Guthrie and Capt. Johnson say they had to brush up on speed-trim system functions after the Lion Air crash into the Java Sea in October.
The longstanding prescribed procedure for runaway trim is fairly simple: stabilize the aircraft, then disconnect the system by flipping two toggle switches called the Stab Trim cutout, short for stabilizer trim, that are on the pedestal between pilots just below the throttles. Then you use a wheel beside each pilot’s knees to crank in necessary trim-to move the stabilizer manually as needed.
The Ethiopian pilots did cut the stabilizer trim off, investigators have said. But with multiple warning alarms sounding and multiple emergencies at hand, they didn’t control the speed of the airplane. The plane can get going so fast that a pilot can’t move the wheel-think of the force of the wind when you stick your hand out a car at 35 miles an hour, then think of what it might be like at 350 mph.
Testing this in the simulator, Capt. Guthrie found he couldn’t move the wheel unless I let go of the control column I was pulling back to try to keep the nose up. Relaxing the column momentarily pitched the nose down even more, but removed enough pressure to crank in some trim.
That procedure of relaxing pressure to add trim repeatedly-sometimes called a roller coaster maneuver or a fishing maneuver, may be added to standard procedures for runaway trim, the American pilots say.
With the fixes in place, the faulty sensor still triggers the stick-shaker stall warning. But unless both sensors break exactly the same way within nanoseconds of each other, MCAS never fires because the angles reported by each sensor will differ. Boeing will also include a message on pilot displays when the readings differ as a standard feature.
And if MCAS should somehow erroneously fire, it will only pitch the plane down once, never moving it so sharply that it exceeds the ability of pilots to pull back on the control column and get the nose up.
Capt. Guthrie says his testing finds version 12.1 eliminating the problems of version 11.1. “We are, at this point, quietly confident,” he says.
Attorney Nomaan K. Husain of Husain Law + Associates has filed a lawsuit against Boeing for the Ethiopian Airlines Flight 302 Boeing 737 Max crash. If you have been affected by this crash, don’t wait a moment longer. Contact us today so we can review your case! Call (713) 804-8149 or click here to send us a message!