Thread Tools
Mar 18, 2019, 08:21 AM
Registered User
Thread OP

MAX-8 possible cause of sensor failure

When I lived in Seattle years ago I found something "silly" re: then-crashing earlier model 737 which Boeing, FAA and APA actually investigated with independent computer simulations as an emergency method to increase chance to regain control. (see below)

So the following might alao be not-so-silly and I leave it to some highly experienced pilots here to evaluate. That is, beyond the being-corrected anti-stall system per se, and the fact that the sensor got knocked by ground crew prior to the Lion Air crash -- which may turn out to be a Red Herring if the below surmise is correct -- there is the following (mere) possibility re: sensor malfunction, a separate but linked issue:

When Boeing does extensive flight testing of a new model, they of course fly it in different environments worldwide. But so fas as I know they do not LEAVE the being-tested aircraft in each environment for months at a time. Further, I assume the testing crew is especially careful to make sure that all systems are properly maintained, including the sensors.

Among other things the two MAX-8 crashes have in common is months-long use in a similar environment which is:

1. moist

2. hot

3. dusty/dirty

4. insect laden

Further it is possible that the cleaning method and solvent/cleaning fluid allowed 1. - 4. to build up certain problems that would not show in the Boeing flight tests.

As an admittedly perhaps totally unreasonable scenario I first imagined trapped condensed water or a tiny piece of ice sliding back and forth in the pitot tube as the AS forced the plane into extreme phugoid, or an accumulation of whatever to diminish free movement of vane sensors.



posted in another thread

.... 737s were also crashing decades ago when the rudder would go hard over and lock to one side

Originally Posted by xlcrlee
I found out at Boeing that their airliners will in fact sustain inversion .... but the oil feed is not pressurized and the engines will soon starve for lack of oil.

.... that came when some of the old model 737s were getting uncontrolled hard-over rudder deflection, causing deadly crashes. I told some top aero engineers at Boeing that since the fairly stiff wings on those 737s still flexed some, that holding it inverted should reduce the rudder-effect by flattening the dihedral when inverted. Then, surprising enough, actual studies were then made following my "stupid idea" (as an emergency tactic till they found a real solution) were made by Boeing, ALPA (Airline Pilot's Union) I was told, and FAA/NASA. And I was told that despite spilled coffee and dying engines the plane could stay inverted. OK, then a short time afterward I am waiting for a flight from SeaTac to SFO and see the pilot of the 737 I'm waiting to board. Thinking how much I'd truly hate myself while sardonically laughing at Fate before being splattered if this plane encountered the rudder issue, I calmly went over to the pilot and relayed I'd figured out, and the ongoing studies, etc. THEN it turned out this guy had been a test-pilot and I think in a simulator flown a 737 inverted! He smiled while quitely "bragging" of his experience and expertise ... and once again my idiocy proved to be less idiotic than might otherwise be expected.

757 coverstock model in that series I produced for Boeing,
same size but different scale as 737 model that I used to show pilot

NOTE: though what we discussed was well out of earshot, the other passengers were NOT amused as the captain and I were sort of seriously using the coverstock model and rolling it inverted .
Last edited by xlcrlee; Mar 19, 2019 at 09:14 AM.
Sign up now
to remove ads between posts
Mar 18, 2019, 09:00 AM
An itch?. Scratch build.
eflightray's Avatar
Extensive testing is mandatory, ..... but it may help if you tell end users ....

Two days before the crash, a sensor was replaced and tested but pilots continued to experience issues with “erroneous airspeed data” and a feature named “automatic nose down trim” unexpectedly kicking in, which pilots were forced to counter manually.

The problem is, most of the pilots operating the new aircraft hadn’t been trained in how to override MCAS. Worse still, most of them weren’t even made aware of its existence.
Mar 18, 2019, 09:14 AM
Closed Account
If you're going to use AoA sensors for flight envelope protection, it has to be assumed that the sensors can fail.

The question is, what do you do about it?

Systems like this have been used on military aircraft for over 40 years. The safety record isn't perfect, but is generally good.
Mar 18, 2019, 01:20 PM
Red Merle SJ VIII
Curtis Suter's Avatar
Originally Posted by eflightray

The problem is, most of the pilots operating the new aircraft hadn’t been trained in how to override MCAS. Worse still, most of them weren’t even made aware of its existence.
Hand fly, MCAS only works on A/P. Hit manual trim and it's gone!
Mar 18, 2019, 01:21 PM
Red Merle SJ VIII
Curtis Suter's Avatar
Originally Posted by NC14310
If you're going to use AoA sensors for flight envelope protection, it has to be assumed that the sensors can fail.

The question is, what do you do about it?

Systems like this have been used on military aircraft for over 40 years. The safety record isn't perfect, but is generally good.
Hand fly the plane. And use airspeed/altitude instead of AoA along with your attitude indicator.
It'll be interesting to see how long the A/P was engage prior to crash. That's what I haven't found an answer too in the two crashes is if/when the A/P was in use. It makes a lot of difference as what system(s) were being relied upon.
Mar 18, 2019, 01:23 PM
Registered User
In my view, MCAS should automatically disengage whenever getting conflicting signals from the AoA sensors - simultaneously warning the pilots about that, so that the pilots could take over the controls, manually preventing the stall condition during take-off .
What makes this issue particularly dangerous is because it occurs during climb near the ground, giving little time for the pilots to react and take the right actions.

However, if MCAS is a "safety system" that must be disabled in order to allow the pilot to fly the plane safely... that sounds like a really unsafe system in my book...

Mar 18, 2019, 01:56 PM
Registered User
richard hanson's Avatar
Guesswork is just that
Years back at an Orly (I think) show for the French Airbus, they stuffed one
The French said we have designed the plane to be flown by th computer and the pilot can intercede
T Boeing said We designed the plane for the pilots to fly with help from the computer
Maybe someone one decided to modify this approach.
Anyway -- let the witchhunts begin-----
Mar 18, 2019, 02:49 PM
Registered User
MCAS system is just a band aid for a pitch instability that comes with the new bigger engines on the B 737 MAX.

Since the 737 has a rather low ground clearance, the new bigger engine diameter required moving the engines forward and higher, this resulting in those engines causing the plane to pitch up at high AoA, at full or near full throttle, as when during initial climbing with flaps up.

Now you have got an airliner that is pitch unstable at high AoA at near full throttle...
And how did Boeing solve this?
Well, they just put in a software trim system where the computer takes control of the airplane.
Maneuvering Characteristics Augmentation System (MCAS).

Unlike the autopilot, the MCAS never disengages itself, even when getting conflicting signals from the sensors and at manual setting, keeping forcing the plane's nose down, without warning the pilots.
It should be obvious that this is a lousy way to 'fix' a pitch stability problem... what could possibly go wrong?

Last edited by funfly2; Mar 18, 2019 at 03:08 PM.
Mar 18, 2019, 03:30 PM
Registered User
richard hanson's Avatar
Without any knowledge as to the real causes the 737 disasters, I have some background on models with thrust lines way off the center of the airframe drag

A model of the 737 with lots of power , would try to loop under full power on take off
But then the same basic setup is used on most large passenger jets
Obvioiusly the full scale setup isn't a full throttle hands off take off approach
The flip side of the coin was an amphibian with the thrust line way above the center of drag.
Years back, one here went full tilt down the runway - but would not unstick .so the pilot chopped the throttle -- then it came unstuck and the inevitable result ensued.
You don't need a Masters in aerodynamics to see what is obvious.
Mar 18, 2019, 04:09 PM
Registered User
The earlier B 737 versions don’t have this problem because their engines are less powerful and are also placed at the “right” location.

The problem arose when Boeing fitted the B 737 with bigger, more powerful and efficient engines, which became known as B 737 MAX 8 or 9.
Instead of rising up the landing gear to accommodate the new bigger engines, they moved the engines forward and higher.

Then they put in a software patch to solve the pitch up problem they’ve created.
Since the software patch relies on info from the AoA sensors, one can figure out what happens when those sensors give wrong values…

Mar 18, 2019, 04:19 PM
Registered User
Thread OP
Right. Why the OP questions only possible environmental cause of the sensor problems and not the supposedly being-fixed AS (MCAS). Note: cleaning method and solvent relates to as well as also being in a way "environmental".
Mar 18, 2019, 04:45 PM
Registered User
Yes, sensor failures might have been due to the environmental conditions, but such failures are expected to occur sooner or later with any airplane, that’s why airliners are fitted with redundant AoA and airspeed sensors.

The autopilot disengages itself and warns the pilots whenever it gets conflicting signals from the redundant sensors, whereas the MCAS does not.

Mar 18, 2019, 04:52 PM
Closed Account
Originally Posted by funfly2
Now you have got an airliner that is pitch unstable at high AoA at near full throttle...

According to this article:

the system's function is to trim the horizontal stabilizer nose down to compensate for the increased pitching moment due to the new engines. This is not "instability" - it's a trim change dependent on flight conditions. If the aircraft were truly "pitch unstable" then pitch rate feedback would be required - not AoA.

Having said that it's a mistake to pass judgement on a system without access to known good information. The press usually gets something wrong, and frequently gets everything wrong in aviation issues. Just listening to the commentator this morning equating angle of attack with pitch attitude made me cringe.

AoA sensor failures can and have caused aircraft losses in the past. It may not be possible to determine if contamination was an issue - after all the sensors are likely to be contaminated after the crash.
Mar 18, 2019, 05:20 PM
Registered User
That's about stability definition. In this case Static Stability.

When during a certain flight condition the plane’s AoA starts drifting upwards, so that the pilot has difficulties to keep it on path, means that it is pitch unstable, which may be trimmed and later re-trimmed back at a different flight condition.

The earlier versions of 737 are pitch stable, since they don’t need such a system to help the pilot keeping the plane’s on right path during a climb.

Last edited by funfly2; Mar 18, 2019 at 05:34 PM.
Mar 18, 2019, 05:59 PM
If it flies, I can crash it.
rocketsled666's Avatar
It doesn't matter how or why the sensors failed, if they did fail and it's not a software bug instead. Sensors will always fail eventually. Software can find ways to fail no matter how well it's tested, too.

What matters is that the autopilot system does not appear to have included consideration for sensor failure. So when the sensors reported bogus data, the system reacted as if they were reporting valid data.

Common sense would seem to dictate that a system that can override the pilot should give up and let the pilot have his or her way if they spend more than a few seconds trying to counteract what the system is doing.

But it appears that Boeing implemented the "Buzz Lightyear" of autopilots, it never gives up and never surrenders.

Quick Reply

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Question What can cause a motor failure on my hex ? Axefly Multirotor Drone Talk 1 Mar 29, 2016 10:44 AM
Discussion What caused motor failure? JanneHir Multirotor Drone Power Systems 5 Mar 08, 2016 09:15 AM
Discussion Radio Failure-Is this a possible cause? trident58 Radios 5 May 24, 2009 08:12 AM
Discussion servo failure caused receiver failure Jetskeeter Radios 1 Jul 03, 2008 12:15 PM