Hacker News Reader

Comment by trhway | Hacker News Reader

Now that everybody, even non-pilots on HN like me, know how to turn MCAS off, it doesn't really make sense to keep them grounded.

View on HN

While pilots now know how to counteract this specific problem, it also highlighted how flawed the safety analysis of this plane was. It makes sense to re-evaluate to make sure there isn't a second problem that just hasn't killed anyone yet.

Plus, how much testing has the airplane gone with MCAS in an off state? Certainly less time than in the on state.

And at the next level up the causality chain, the safety analysis was flawed because of the regulatory capture of the FAA by Boeing.

Should we therefore ground and review all the planes?

The plane isn't airworthy in an MCAS upset condition resulting from faulty AOA sensor. That's because numerous contraindications are happening in the cockpit as a result, including the failed side stick shaker activating. In the Lion Air 34 and 610 flights, the stick shaker was active almost continuously. That's not airworthy.

Also, the autopilot uses stabilizer trim. If you turn stabilizer trim to cutoff, you have no autopilot. Also, the emergency airworthiness directive says it's possible a runaway continues even after stabilizer trim is set to cutoff: If runaway continues, hold the stabilizer trim wheel against rotation and trim the airplane manually.

Once there was a second crash, with stabilizer jackscrew in the full nose down position, it was untenable to allow the planes to be considered airworthy, for the very central reason that we don't completely understand what happened and how to avoid it.

The proper response in aviation safety is not to assume pilots are doing the wrong thing upon receipt of an emergency airworthiness directive. It's to assume they are following the directive. You don't just pretend they're morons and keep on subjecting the general public to risk, while you guess at what the problem is.

It's there to solve a problem though. You can cut power to the recalled airbags in your car too.

The whole point of MCAS was to avoid pilot retraining on the different handling characteristics. To say that you can just not have that system is also to say the pilots need to be retrained for the Max planes. Something that Boeing felt necessary to avoid and thus implementing this MCAS. Now that 2 planes have crashed you can't just say "Oh Okay just turn MCAS off duhhhhhhhhh".

The misfeatures that are suspected to have contributed to the recent accidents exist in part to keep the plane similar enough to previous 737s that lengthy pilot retraining is not necessary. Now that that fiction has been dispelled, additional training seems to be called for. (probably a bit more is needed than passing on whatever knowledge 'non-pilots on HN' have picked up)

If turning it off was a fix then it wouldn't have been there in the first place.

I am sure the feature is there to help keep the plane stable. It can be flown without the system being on if the sensors don't work. You can drive your car without cruise control, but it makes it more convenient with it on.

The feature is there so pilots don't have to be retrained on the different handling of the plane. It was entirely cost avoidance on Boeings part. To say that they could just turn it off is also an admission that the plane is not the same and pilots would need to be retrained. Boeing can't have it both ways.

> the feature is there to help keep the plane stable

And we don't want that?

> You can drive your car without cruise control

Sure, because cruise control is there to allow me to be lazy, not to keep my car on the road.

The aircraft has handling deficiencies at high angles of attack due to the engines being mounted differently. The MCAS is there to correct this. It's this or redesign the plane. Only 2 things that could have helped: an angle of attack disagree light to warn the pilots of the issue, and to to retrain the pilots. The first was sold just as an optional package, the second was considered unneeded since Boeing treated the new plane as a perfect equivalent of its predecessor. Disabling the MCAS is not a fix and without it the plane is not certified to fly.

It works when the sensor works. When it doesn't you need to fly it manually.

What I'm saying is that the MCAS is there so the plane can get the certification. You can fly the plane without it but it's not considered normal (or safe) operation. Which is why it must be there and it must work.

You can fly a plane without an engine if needed but just try certifying a plane with one engine under one wing.

Given that there was a second crash , similar to the first, I don't think this argument holds.

At the time of the second crash it wasn't as clear what the likely culprit was. If it was believed to be a defect with the aircraft at that time then most other countries would have already grounded them by then. I think that's a bad argument to make as well, but not for those reasons.

Everyone who earned a type rating to fly 737s also knew how to turn it off, by following the non-normal (aka "emergency") checklist memory items for stabilizer runaway.

It is shocking to me that the second crash happened, if they both turn out to be MCAS and stab trim sourced. How could you step into the cockpit of a 737-Max that day and not know about the MCAS issue and be primed to respond according to the checklist?

I have to add that I'd be happy to put myself or my family onto a 737-Max flown by a US-flag carrier (Southwest, American, United [Delta doesn't fly the -Max]), as I believe there's a significant variance in crew training worldwide.

Bad sensor data was represented as accurate sensor data. It is easy to solve some problem you know is occurring, it is hard to solve a problem where something weird happens and the data in front of you is telling you that what you think is happening is not what is happening. ("I think we're stalling." "Nah, the sensor that directly measures that says we're not." "Hmm, what else could it be? Get me the checklist" crash)

It shocks me that the redundant angle of attack sensor and the "disagree" light is not standard equipment. Who is buying multi-million dollar airplanes and cares about the cost of an angle of attack sensor and a lamp? Not people that care about safety, that's for sure. I know the airline industry loves nickel-and-diming their customers, but safety equipment is a little more important than priority boarding and a checked bag. It is shameful that Boeing adopted that business model.

More on angle of attack sensors: https://www.flightliteracy.com/angle-of-attack-indicators/. If you fly GA, you should get one. Having a readout of the flight energy you have available is one of the greatest safety upgrades you can make.

Agreed on the value of AoA; it’s a toss-up whether we’ll be upgrading in the next couple years. Not much sense in adding it to a plane we might be selling shortly as it doesn’t bring much increase in resale. Next airplane is likely to have it as OE.

There's circumstantial evidence this particular variety of MCAS upset doesn't look like runaway trim. The manifestation is transient, seems capricious, results in surprise.

U.S. based pilots have said they didn't know anything about MCAS or MCAS upset until the FAA issued an emergency airworthiness directive, AD#: 2018-23-51, expressed surprise and were critical to discover this after the fact. Years after the fact. And it took a crash.

A second crash, despite the issuance of the AD suggests the AD is somehow sufficient. Maybe we don't know the full range of possible MCAS upset behavior. Maybe we don't know if there are additional factors that contribute to surprise and confusion.

>It is shocking to me that the second crash happened, if they both turn out to be MCAS and stab trim sourced. How could you step into the cockpit of a 737-Max that day and not know about the MCAS issue and be primed to respond according to the checklist?

the MCAS caused almost crash the day before the first crash of the same plane - it was only that off-duty pilot who knew about MCAS and thus saved the plane. Yet the next day that plane flew again with an MCAS unaware crew. Looks like some systemic issue. Grounding of the planes would allow to correct the specific tech issue of MCAS. Whether the systemic issue is going to be corrected is yet to be seen.

Agree, seems like waste of resources to keep them grounded.

I’d prefer to waste the resources of an airline or Boeing, rather than the kind of “waste of resources” in losing whole plane loads of human beings. Turning off MCAS solves one problem and trades it in for another, it is not a waste of resources to solve all of the relevant problems.

What is a waste is pretending that wasted resources is the worst thing that can happen here when we have two fatal crashes being investigated.

What an irresponsible comment. One can't just hand-wave away safety protocol, even if the failure appears to be obvious.

The issue is more of incompetent pilot. As indicated that the day before the lion air crash, the pilot know what to do to handle the issue. At least for lion air, the airline has been locally famous for being careless about maintenance and pilot training.

"Stop Trying to Fix the User"

https://www.schneier.com/blog/archives/2016/10/security_desi...

So the plane's not unsafe, you're just statistically vastly less safe as a passenger ON the plane, because the pilots are much more likely to crash it, killing everyone aboard.

Well, that put me my place. I'll hop right on another one tomorrow!

You could also say that Max flights now require 3 pilots to safely manage.

No, what required is a properly trained pilot.

Oh my plane is pointing at the ground. Let me just stare at the trim indicator to see some intermittent change that only occurs every 10 minutes.

Exactly, but instead of requiring the retraining of pilots for a new airplane that flies significantly differently then the previous model, Boeing put in an automated system and charged extra for the instrumentation needed to determine if that system is failing.

Reply on news.ycombinator.com