So I would think (and hope!) safety precautions would be taken to ensure that as the hydrogen is pulled out of the resevoir that it would be replaced by some other element, to ensure that pure magnesium did not remain in the system in significant quantities.
Exactly the point - we'd think that'd be a natural thing to do. My guess (and experience) is that it won't happen.
Pick your favorite hybrid car. Big batteries in them, eh? Nice fat voltage and current, as well. Now, smash that car up with someone in it. I'm going to take an O-cutter, which is a fun tool that'll crank out about 80,000 lbs of force when it cuts, and try to get the person out. Big question - where do I cut? More importantly, where do I
not cut because it'll kill me?
You'd think something as trivial as standardizing the routes of the power cables would have happened. It hasn't. It's vastly different between makes, and is even vastly different between model years. Can I disconnect the batteries? Is there at least a standard way to do that? Nope. They move those too, along with how they're connected. And remember, the car is crushed and/or upside down - access panels aren't exactly going to work very well.
Pick your favorite "recent" car, with the airbags I mentioned. Airbags are great during a crash - they use a stored-energy device to inflate a bag (with great force) and hold you in place, so that you'll decellerate with the car over a longer period. Great concept, and it works - they now put airbags everywhere - side curtain, rear curtain, side-seat, you name it. It's wonderful.
Now, imagine that every one of those bags has it's own little "stored energy" device, because each one does. You know how much energy they have in them, and you know that they're rigged such that they'll still function for some period of time if the electric system of the car has been crushed. They're very well thought-out in how they are designed, and are quite robust.
Except for one small issue - where they get placed. They put them *everywhere*. In the door, low in the B-post, middle of the B-post, top of the B-post, in the roof center, in the roof edge, under the center console, in the dash, in the seat, in the floor under the seat...
...and they move the dumb things *every model year*. Care to guess what happens when my trusty O-cutter "finds" one of these things? I become "one" with the tool, and any EMS guys inside the car by the corresponding airbag will probably get his ribs or head crushed. You can cut the battery and pretend to disarm the system, but... nope, they've got nice little caps in them to store charge. We're supposed to wait 20 minutes before cutting, "to be safe" - and meanwhile, our poor victim is bleeding out. Even missing the charges is often irrelevent - when cutting things, all wires in that area get cut as well. And of course, they get shorted together (and grounded) in god-knows-what-order. One of those wires is the trigger for that charge. I'm going to short it to a whole pile of random potentials, aren't I? You'd
think they'd have considered this fact in the design, right? At least, I would... or let me rephrase that. At least, I
used to think so. As was demonstrated last month, I was mistaken.
Please note that this *is* the general case with modern cars. We're working on tactics to reliably mitigate these explosives, but it takes time - and we get no help from manufacturers or industry. All they want to do is make us carry a 1 billion page book that changes every week, detailing every make, model, and variation of car with where the hazards are *expected* to be. It doesn't work.
So, I'm cynical. You can wash, rinse, repeat this cynicism with modern housing construction. Nice windows that are 99.99999% heat reflective, and basically unbreakable. Great when things are fine, not a good thing when (1) the room is on fire and approaching 2000 degrees and (2) you're in the room next to it, trying to make exit through such a window. Add some nice kevlar bug-screens into the equation, and not even a K12 saw is going to help.
On the good side, such houses typically use the latest "state of the art" construction materials and techniques. They use engineered wood products in very complex truss assemblies. Gussested Wooden I-Beams are the norm, and J-Beams (an I-Beam with a glued strip of kevlar replacing the lower channel) are beginning to replace them on first floors where the lower basement is unfinished. They work real well, except that the glue that binds the wood-chips is "rated" to fail at 400 degrees in 4 minutes. We all know how forgiving a truss is to componant failure. 400 degrees, 4 minutes. Typical living room with a modern (petroleum based) couch in it will hit 1600 degrees in seven minutes. We should be able to receive the alarm, get there, and gain entry into the structure for a life-search just in time to wear the bulk of the structure on our heads.
An old fire-dog named Frank Brannigan has done enormous work regarding such structures, and with that industry. He actually met with the engineers who developed the "wooden I-beam", and asked them "So, how long is this stuff going to last in a fire?" The answer: "Don't expose it to fire." "No," he insisted, "We didn't start this fire. The house is burning, there's people in there, and we have to get them out.
How long is
this stuff going to last
in a fire?" The answer, amidst stares of "horror and disbelief" (Frank's words) by those engineers? "Uhm, don't expose it to fire."
The J-beams, which are even stronger due to the kevlar strip replacing the lower wood-piece, perform even worse under fire. The glue holding the strip fails in about 3 minutes at 400 degrees. Compression along the upper edge is still maintained by the wood-channel up there, but there's now nothing to resist the tension along that lower edge as the kevlar delaminates. Please note that a waste-basket fire will give you about 400 degrees directly above it, rather quickly... and also note that the glue in the rest of the "J-beam" is still approaching failure at the same time.
And recall that water weighs 8 pounds per gallon. An average fire truck will pump out 1200 to 1500 gallons per minute during a rocking fire. Call it 1200... at 8 pounds, let's see... carry the 2, add the 1... that's 10,000 pounds per minute that we're adding the the load in the structure. Think the engineers considered that fact? Nope, they didn't. Even at a "bread-n-butter" fire, where we'd use half the flow-rate - that's 5,000 pounds per minute. And you think landlords bitch about waterbeds weighing too much...
Attic spaces still use wood studs in their trussing, though - but they use gusset plates to hold them. Such trusses are good for an average of 7 minutes at 600 degrees before the wood has pyrolized around the teeth of the plates, at which point the plates simply "fall off" or "pop out" due to the wood's contraction away from those teeth. The result is typically a spectacular, instantaneous failure of the entire roof assembly. Or as we call it, "the attic situation has just converted itself into a basement situation." Hopefully noone is on that roof, or in the structure when this happens.
Still more craziness can be found to the North, in Canada. There, some putz created a *really* great idea. Take some very thin steel U-channels that only offer strength under compression (exactly under compression, and nothing else), and call them "studs". Sandwich them between two sheets of styrofoam, and call that a "load bearing" wall. Stick four of these walls together to form a box, stick a platform on it, and there's your first floor. Repeat to form the 2nd floor, etc. And believe me, the result *is strong as hell*.
Until it burns. Then the entire structure simply "screws down", almost immediately. You'd think that technique will never make it into the U.S.
So, I do hope people consider "eventualities" when they design and implement all of this (otherwise) great stuff. So far, however, they've completely ignored it. Various fire districts are reacting in different ways as a result. In one specific county in Virginia who's population is exploding, for example, they run a 10 minute timer from when the alarm goes off. If they show up and no fire is visible, they'll perform an aggressive attack until that 10 minutes is up. If any fire is visible when they show up, it's assumed that the 10 minutes already expired before the fire was called in. Either way, the structure is considered lost once the 10 minutes is "done", along with any people trapped inside of it. Attempts to rescue may still take place from the outside, but noone is allowed in. Ever. They do this, because in their tests of these new structures, the typical floor above the fire would come down within 12 minutes, and some in as little as 7. Kind of scarey, when you consider that it takes an average of 4 to 7 minutes for a fire company to even get on-scene, and that's assuming the alarm was raised the *instant* the fire started. You'd think the people building this stuff would have considered it. And you'd think the people buying these houses would know about it. They haven't, and they don't.
The above is a little off-topic from this thread, but it's very much on-topic with the "I'd think safety precautions" concept. My guess is that such precautions won't be considered, simply because they never have been.
Comments?
(and
thanks for listening to me vent on this topic. Obviously, it's a little stressful

)