Additional Software Problem Detected in Boeing 737 Max Flight Control System, Officials Say

Pilot error and a damaged sensor....sensors go bad all the time, there are safety measures in place to counteract the loss of a sensor...The plane was simply going too fast to recover...If the pilot had cut back on the throttle, they could have recovered...

https://www.reuters.com/article/us-ethiopia-airplane-reconstruction-insi/how-excess-speed-hasty-commands-and-flawed-software-doomed-an-ethiopian-airlines-737-max-idUSKCN1RH0FJ

.If the pilot had cut back on the throttle, they could have recovered...

No, I don't think so. The throttle makes the plane go up or down, pitch makes it go fast or slow. When the nose is pointed down, it's like driving down a mountain grade with no brakes. And I suspect since the engines are mounted low, less throttle might result in more pitch down, the last thing they needed.

Remind me never to get in a plane with you as pilot.... You'll be trying to take off with the trim control...

No. it's the throttle that gets the plane going and keeps it in the air.

OK, I have to admit my statement was too simplistic. If you're too low, you add throttle and if you're too slow you push the stick forward, but they all interact. The throttle adds energy and the pitch control swaps altitude for airspeed.

I guess my point was that the excess airspeed, IMHO, was more due to the pitch down attitude than the throttle setting. My thought was that the engines are below the wing and body of the airplane. The thrust and drag couple would exert a pitch up torque, and reduction of thrust would cause even more pitch-down.

"If you're too low, you add throttle," is only true if you are close to stall speed. If you're flying along at V_C just pull back on the stick. You trade airspeed for altitude that way.

The point is that the pilot never throttled back the thrust after takeoff...If you notice when you take off in an airliner the pilot always throttles back after getting airborne...In this case that never happened, the pilot left the thrust control on full, and when the plane went gear up and the MCAS system kicked the nose down, the plane was then over accelerating, the pilot then could have used the trim adjust on the yoke, it overrides the MCAS control....It looks like he attempted this, but didn't push it long enough to correct the trim fully, he should have used that control and throttled back the thrust slowing the plane and regaining control...Instead they shut the trim auto control off but then when they didn't have enough strength to manually adjust the trim because they were going so fast, they reengaged the autopilot , and that was his last mistake because that gave control back to the MCAS system further pushing the nose down, at that point he probably still could have used the override....This is , of course, only my opinion with the known facts, more information could come out...

I am compelled to disagree with you.

Air speed (throttle) alone does NOT make the plane go up or down. Yes the relation of the thrust to wing can make some difference.

It's not quite that simple. It's the four forces acting on a typical aircraft; lift vrs weight(gravity) and thrust vrs drag. The effect of the forces can be resolved with vector math.

In straight and level flight all are balanced.

Increasing thrust increases airspeed and tends to increase lift more than drag, and an aircraft will tend to gain altitude. Increasing airspeed without gaining altitude generally requires pitching the nose down simultaneously.

Decreasing thrust decreases airspeed and lift and without pitch adjustments will tend to make the aircraft lose altitude.

So in a strict sense the throttle does make the aircraft go up and down. But it's a rare case where the pilot relies on that alone. Then you thrown in stuff like flaps, trim tabs, lift killers, etc, and life gets much more interesting.

Hooker

My background is in missile systems where the flight is controlled by a 'Space Gyro'.

Why don't modern aircraft use a Gyro system as a back-up giving 'true' indication of the aircraft flight attitude or is this too expensive a fix?

Tom

They do, it's called an "Artificial Horizon"

Admin here.

The last Boeing 737 Max thread devolved into petty arguments about moral high grounds and nationalism. We closed that thread.

Don't let that happen here. Please stay on topic.

Have you admonished the offenders?

..

Sorry.. so sick of it myself.. but I do miss a certain hotjead.

TCM?

He's probably run amok....

Aw shuck Schuster, you knew. Shipcup, great rings of rubber fire.

I believe we really have to wait for all facts on this issue. We can be sure that the FAA and NTSB will be looking at all this very closely. It may take more than a year for investigators to thoroughly review all the information. A mob is not a good form of justice.

Lyn, I assume in all your years of engineering experience you worked on either avionics or airframe components. If you worked on the certification end, you would find the FAA and all other certification authorities in the world, do NOT have the technical expertise to approve or reject compliance in all designs.

In the case of the FAA, they have a method where DER's (Designated Engineering Representative) act in this authority. These can either be long term employees of the company producing system designs, or for hire contract DER's. Either way, they act for the FAA, and are trained in this end. I did this for a very short time before retiring from the same company (subcontractor) that produced the MCAS per Boeing design. I did not work on, or have any knowledge of this work.

There is no COZY relationship.

I do believe in this case that Boeing has a major design defect in that this MCAS (Maneuvering Characteristics Augmentation System) is a flight augmentation system to prevent stall. I do not know personally what the failure rate is classified to, but for general autopilots, trim runaway had a 10e-9 (1 per billion hours of operation) failure rate that results in a catastrophic event. This system is effectively an automatic pitch trim control. I believe this same rate should apply to MCAS. And there is no way to get there with a single sensor driving the system events. I do not understand what the heck happened at Boeing that produced this system design. All my previous 29 years of watching Boeing, was them doing overkill systems. And they always had the philosophy of letting the pilot fly the aircraft, and not hiding override into the system design.

Over all my years in aviation I have worked with over a dozen FAA DER's. As with most federal gov't programs this program is deeply flawed. I've worked with mechanical, electrical/electronic, structural, systems, engines, and flight test DER's. They have ranged from honest and competent, to incompetent, to not giving a damn and also "on the take to the highest bidder".

Many DER's are directly employed by the airplane manufacturers. These DER's would be hard pressed to delay a program by blowing the whistle on "minor" design flaws, particularly in software systems which were developed by outside contractors who are generally tasked with debugging their own products.

I don't know that this is the case with the 737 Max MCAS system but I would hardly be surprised. I will also be surprised if we ever learn the unvarnished truth of the problem(s) if FAA oversight is found to be lacking.

Hooker

I agree about DER's and their competence. The FAA is no better, as you CAN'T hire the talent needed. So the regulatory agencies have to trust the manufacture's DERs. As well schedule is the driver in this new non technical CEO driven world.

All a DER is doing is showing compliance to the regulations. It is nothing but a document check off list.

I quit my DER position, as I was asked to approve of a design that was not flightworthy. I quit my job over seeing the CEO giving away 80 years of intellectual property to India and China. I wanted no part in transferring IP that was given to me by a long line of mentors.

I do not understand how Boeing could have designed such a system with single point failures.

On the subject of competence, at the end of my career, only 1 in 20 new hires had engineering aptitude. The whole supply chain from grade schools through colleges, career counseling, to human resources, all believed that a degree'd engineer has the talent to practice the art of engineering. A high grade point average is the only litmus test. I have to assume this is the talent that created the MCAS system.

It looks like Hooker might disagree with you. I do too.

My stuff went into Earth orbit and at that time there was no way to repair mistakes after launch.

I hope that you will agree that someone (probably many in the chain) screwed up and killed people.

The system fails to observe the most basic of safety measures.
There are 2 sensors, the software should read both, if they disagree it should look for sudden changes in output from one sensor and if present, ignore that one. If it can't determine which is correct it should fail safe e.g do nothing, or, if stall is considered to be more dangerous than doing nothing, sound alarms and take minimal action. The results of the action should be cross checked against the sensor information to try and determine which if any is providing feedback that correlates to the corrections which were instigated. If there is no correlation stop making corrections.
If pilots over-ride the system it should only re-apply the correction a limited number of times.
To apply extreme corrections (greater than stated in the specification) based on a single sensor input is simply negligent.
To fit a bigger engine in a position that jeopardises the stability of the aircraft is dubious in the first place.
For the FAA to delegate testing to the company making the plane is bad practice, foolish, negligent and possibly illegal.
Some of the comments on this thread beggar belief... sensors fail, yes, so don't write software that just relies on one!
The pilots were not at fault they should not have been put in this situation.
Compare and contrast with the A320 landed on the Hudson when both engines failed... the plane "allowed" the pilot to fly it.... it didn't make him crash.
Also compare with the Challenger disaster, decisions made in spite of the engineers warnings to save money, meet schedules etc.
Rest assured, lessons will be ignored, heads will not roll, those responsible will maybe get pushed sideways or promoted.
People at the top have forgotten the basics... simple as that.
Del

"pilots must be able to hand fly airplanes out of danger."

Those days are long gone. The push to total automation is on. Damn the torpedoes, full speed ahead.

Hooker

Unfortunately too true!!!

I have flown on this aircraft several times. FA flies from Ottawa to Iqaluit and other destinations as well.

Computers don't care if we die. I can't imagine the horror of a blue screen lockup at 30,000 ft. I want a pilot that wants to live to see another day, and has the ability to override any and all computer programs if necessary.

It happens...

https://www.boston25news.com/news/trending-now/flight-diverted-after-planes-cockpit-screens-lose-power/937474712

It shouldn't. When we loose the ability to opt out of AI.....we're toast. It's probably already too late.

It happened yesterday.

CLICK HERE

I have always heard there are multiple backup systems, but here one sensor was used. Now they will use two, but what if they both fail? To me, three is the bare minimum, and four or more is preferred. And where's the backup for the software? Good engineering was not used.

The nose must always go down to compensate for the round water. Duh I hate it when stupid people say that water finds it's own level. It's round! !!!!!

Proof positive...

So you're agreeing with kramarat then?

Are you going to argue with him?

I think I'll leave that for Lyn.

Logically, I have to agree with you here. Nothing can be level on a sphere shaped object. Good point.