Hacker News Reader

Hacker News

Comment by calamari4065 | Hacker News Reader

Parent Full thread

calamari4065

I can come up with several methods to solve this problem at the hardware level.

Slap a temperature sensor on/in/near the heater

Monitor current through the heater for temperature coefficient response

Capacitive sensing of ice on the heater

A timer that shuts off the heater after some time and a buzzer to alert the crew

You could even have a thermal fuse with a manual time-limited override

Honestly there's so many easy ways to prevent thermal runaway that creating a situation where THE ENGINE FALLS OFF is inexcusably negligent.

My space heater has no less than three independent, redundant safeties. As does my clothes iron, my coffee pot, my slow cooker. It really is not that complicated.

View on HN

Remember that Boeing was too stupid to put an odd number of pitot tubes on their planes, so the computer could have determined which airspeed indication was crazy. They put TWO.

Matt Stoller wrote a compelling article asserting that the breakdown at Boeing resulted from the breakdown of the separation between their military and civilian divisions: https://www.thebignewsletter.com/p/the-coming-boeing-bailout

If you think about it, civilian standards must be way higher than military ones; you're talking about millions of people per year who can't bail out of a crashing plane.

The breakdown came from trying to crush their engineering organization for costing too much/being unionized — and particularly firing the mid-career engineers, who would by now be the missing senior/principal engineers.

MCAS is a reflection of their systems integration engineers not being the caliber and experience they need: it was obviously a defective design.

Not to mention that it was a fraud to begin with, trying to skirt pilot-training rules by having (monumentally incompetent) software tweak the controls of a perverted airframe design to make it imitate previous 737s.

If they had done what they needed to do and designed a new plane from scratch, those people wouldn't be dead.

OR if they'd simply acknowledged the changed characteristics of the plane, struck deals with purchasers to subsidize training expenses, and let pilots fly it themselves... those people wouldn't be dead.

Boeing's recent behavior, as reported by this article, warrants further disgust and ought to be featured in the mainstream news.

If they had designed a new plane from scratch, those people wouldn't have died because the new airframe wouldn't have been flying yet. Check back again in 2030.

Still a win.

Tell me, what is the exact number of dead children you're willing to accept in order to get new planes flying?

Just like the optimal amount of fraud is nonzero (https://www.bitsaboutmoney.com/archive/optimal-amount-of-fra...), the optimal number of dead children is too. A child doesn't have a meaningfully greater chance of dying on an airplane than an adult, so what this question is really asking (while crudely attempting to tug on the heartstrings) is whether we're willing to tolerate people dying in order to keep planes flying. The answer is obviously yes. We don't cancel all flights globally when a plane crashes and everyone dies, because we suddenly can't tolerate the risk anymore.

It’s not an electric heater, it blows engine core air from the compressor into the inlet lip.

And sensing icing on an inlet isn’t as simple as you make it out to be.

So humans are supposed to be able to do it (within five minutes) with the sensor data already available to them, but it's not easy to make a sensor do it automatically?

Maybe this is overly optimistic, but I'm confident they could figure out something if they wanted to.

Great, that doesn't change the fact that there are many ways of ensuring a structure doesn't overhead and explode.

And one or more of those is probably going to be the long-term fix. However any change like this also needs FAA testing and certification, and enough design and review to be sure the changes don't introduce other new problems, all of which will take some time.

yes, good, time for the FAA to force Boeing to think ahead

Reply on news.ycombinator.com