Go to Making Light's front page.
Forward to next post: The Feeste of Kalamazoo
Subscribe (via RSS) to this post's comment thread. (What does this mean? Here's a quick introduction.)
Most of us will have heard that because glass is a supercooled liquid,* or an extremely viscous liquid, or an amorphous solid, or anyway a material with some very odd properties, old panes of window glass will gradually, over time, flow downward and become thicker at the bottom. Have we also heard that this is in fact untrue?
There are several ways to prove it, ranging from hm&te scientific demonstrations, to the simple observation that whereas old glass windowpanes do tend to be thicker at the bottom and thinner at the top, no one has ever spotted an instance where the supposedly flowy glass has actually flooped out over the edge of the framing material.
In my opinion, this one’s simple. Old manufacturing methods produced glass of irregular thickness. If you were a glazier, almost every pane you installed would have a thicker end and a thinner end. The lower portion of the pane bears more weight, so you put the thick end at the bottom. Light normally comes from above, so you put the thinner and more transparent end at the top. The arrangement is no more evidence that glass flows, than tapered wooden shingles are evidence that wood flows.
For the fraction of a percent of readers here who never poked their heads into the usenet group alt.folklore.urban, the glass flow trope worked as a bit of a shibboleth there for some years.
Oddly enough, I had never heard that particularly scientific legend, at least, if I had, I hadn't remembered it. It doesn't actually make any sense when you think about it. Solids generally don't become liquids or liquid-like unless conditions change so as to make them unstable. But glass is appealing, magical stuff, especially in mirrors: it stimulates the imagination delightfully, particularly if one has seen it made. I work with glass in the form of glaze (as in pottery & scultpure). Part of the fun of the process is not knowing exactly what will happen when a piece is glazed.
Lizzy, I've done the same. Glass is charmingly perverse stuff.
Now I want to make a window pane out of silly putty.
Is the link at the asterisk supposed to be the same as the link at "tapered"?
I first read this in a Heinlein book, I think. You are shattering my cherished illusion that RAH was always scientificly rigorous. (Although the truth is fascinating. Who knew glass was such interesting stuff?)
While I was recently cured of this particular fable, I still believed that glass does flow, but much too slowly to be noticeable even over hundreds of years; thus flow occurs but does not account for the thick-at-the-bottom panes in those old churches. (And not only does the glass never overflow the paning at the bottom, it never gets "melt-holes" in the top, either, no matter how old it gets.)
But the subtle second-order phase transition means that's wrong too.
One of the coolest things I've heard about recently is the creation of metallic glasses. Stronger for their weight than crystalline metal. Of course, I immediately started wondering about their musical properties...if they differ at all from crystalline metal.
I do my best to shoot down this myth wherever I find it.
Another is the use of Bernoulli's principle to explain airplane lift.
So, you're, ummmm, saying wood doesn't flow?
Um...Mark? Explain how Bernoulli's principle doesn't explain airplane lift? I still believe(d) that one. How does it work then? I just saw a Nova program (or something) a few months back that explained it in Bernoulli terms.
I HATE it when I believe nonsense, especially when I've been telling it to other people!
As somebody who gets to tell people cool stuff about materials for a living, my thanks go out to Our Gracious Hostess for helping to dispel this myth. Ashby and Jones point out that it would take on the order of 10 000 years for glass to flow appreciably at ambient temperatures; we can come back and check out our cathedrals (or shopping malls, or junked cars) in a few millennia.
They also mention that there is one important ceramic, with a relatively low melting point, for which flow under load is really important - ice, in the form of glaciers.
I don't know... While I'm willing to concede that it probably is a myth, I don't think you've provided evidence to that end.
The manner in which small quantities of window glass (or for that matter, sheet glass of any kind) was made in the pre-industrial/early-industrial period was by getting a blob of molten glass onto your punty (the blowing rod), and blowing out a bubble which was then flattened into a plate. This plate was then heated and spun, using the blowpipe as an axle, forcing the plate out wider & wider. When you were finished, you snipped the plate (up to 1.25m across, or larger) from the punty, annealed it to remove the stresses, and then scored it into sheets. (The punty mark at the middle of the plate is what made 'bullseye' glass.)
Early industrial-era glass was made by floating molten glass on a bed of molten tin. You then ran a heated steel roller over the molten glass, forcing it down into a more-or-less even layer. This is called 'float' or 'water' glass, and the process results in the wonderfully ripply, vaguely watery appearance of pretty much all glass produced until the early- to mid-20th century, when techniques for manufacturing plate glass as we know it today came into widespread use (our house dates to 1914, and has some original float glass windows; one of them has a central pane which is almost 1.5m across and probably almost 1m tall).
Neither the hand-blown technique nor the float glass method (the float glass method even less so) result in (many) panes of glass that are shaped as you describe. Glass wants to achieve the same level when heated - it really, really, really wants to be no more than 5mm thick or so, and will spread out on its own to a uniform level with astonishingly little persuasion. You have to coax it to get thinner than 2-3mm, but not all that much...
Glass doesn't melt like other solids - so many other substances suddenly give way from one state to another. Glass gradually phases, more like sugar candy softening under gentle heat than, say, an ice cube. I could see how observing glass under heat could lead one to believe that it is more of a funky liquid than a 'true' solid.
Glass is way cool stuff. I dabble in soft glass kiln-forming and beadmaking - but I've never blown glass.
I first read this in a Heinlein book, I think. You are shattering my cherished illusion that RAH was always scientificly rigorous.I don't recall Heinlein mentioning flowing glass. The only vaguely related thing I can recall from him is a mention of glass exposed to strong sunlight turning blue over time, which I think is in "Waldo." Whether glass does turn blue or not, I have no idea.
How Airplanes Fly: "The second description we will call the Popular Explanation which is based on the Bernoulli principle. The primary advantage of this description is that it is easy to understand and has been taught for many years. Because of its simplicity, it is used to describe lift in most flight training manuals. The major disadvantage is that it relies on the 'principle of equal transit times' which is wrong. This description focuses on the shape of the wing and prevents one from understanding such important phenomena as inverted flight, power, ground effect, and the dependence of lift on the angle of attack of the wing."
D'oh - lost a couple of sentences in formulating my comment - so, what I meant to include was something to the effect that any irregularities are due to the manufacturing processes used; any instances where the glass is thicker on the bottom than on the top is probably because in those cases, the glass was noticeably thicker on one end than the other. If you were to perform a comprehensive survey of old glass panes, I'm guessing that you'd find that many of them are thicker on one end than the other - but that it takes a certain 'obviousness' threshold for the thicker end to wind up on the bottom. This would also contribute to the myth: the thicker-bottomed panes are, by definition, somewhat more remarkable.
Thus ends our first SWAG of the day.
Neither the hand-blown technique nor the float glass method (the float glass method even less so) result in (many) panes of glass that are shaped as you describe.According to several of the references Teresa linked to, the punty method does produce a noticeable number of panes which vary in thickness, which from the physics of it makes sense (spinning a disk of viscous material will tend to make the edges thicker).
But... but... but I was told that myth by a glass-blower! *sobs* In Grants Pass! On a First Friday Art Walk when the open up the forge to spectators! *wails*
Re: Bernoulli's principal - [puts on pilot's headset] I'm pretty sure I remember being taught that that's what causes the wind flowing over the top of a mountain to be really really bad for the structural integrity of a small plane. And also what causes carburator icing.
Protected static:
Two things: one, all of the pieces of glass cut from the 1.25m-diameter spun glass, except the bullseye, would be thicker on the side closer to the axis (the blowpipe). That's why older glass has a thick side and a thin side.
Second, I think you've conflated 'hot-rolled' glass (made by flattening a high-viscosity sheet of glass between rollers) and float glass, made by floating low-viscosity glass on tin. The latter is the way that sheet glass (large panes with parallel sides and excellent optical clarity) is currently made.
Oh, and Xopher - I have a sample of amorphous metal in my lab - I'll have to go whack on it when I head back. My guess is that it should have markedly different musical properties than the traditionally-used metals since it has an unusual composition and is particularly hard. But I know that assessing the acoustic properties of materials for instruments is insanely complicated, and I couldn't presume to address that side of things.
debcha: Yeah, the hand-blown method would produce pieces with an integral slope, but probably not quite as many as you'd think. It would, however, produce enough obviously sloped pieces to contribute to the myth.
As to the 2 processes, it is quite possible that I have conflated them to a degree, though my understanding (derived largely from reading the semi-technical materials from glass suppliers - although the materials are written for non-technical audiences, and there's no guarantee that they don't contain myths or radical oversimplifications as well...) was that the float method could also produce rippled and wavy glass.
Protected Static:
The punty is a solid rod, as distinguished from the hollow blowpipe.
The method I learned involved blowing the "light bulb", then attaching a punty to the other side and cutting the pipe loose. Spinning/heating makes first a bowl shape as the hole from the pipe expands, ultimately a flat disc with the bullseye from the punty in the middle. There is frequently more thickness at the outer edge of the disc, though it depends on how often you reheat the blob during the bubble phase, and how evenly you blow out the bubble. Elongated bowls with deliberately thicker edges can be made by holding the punty vertical and letting the shape sag towards the floor.
Protected static, as I understand it, glass panes were cut down from blown disks and cylinders until the 1870s. After that, glass was poured flat, rolled, and then ground and polished. Plate glass, with its genuinely regular surface, was a premium product. Less premium products were less regular. The technology for drawing continuous sheets of glass was developed in the Teens, so your 1913 house would most probably have poured-and-rolled glass in its windows. True molten-metal float glass didn't come in until 1959.
Unless, of course, glassmakers use more than one sense of the word "float."
Dan, crown glass -- that's the round stuff -- is thickest in the middle and thins further out.
Wood floors tend to be perfectly flat, which is exactly what you'd expect from the liquid wood theory.
See, that's what I like about this site - I almost always leave feeling smarter than when I started (after I get over the feeling dumber part, that is...). It has a wonderful way of highlighting the gaps.
Mind the gap ;-)
I see places where liquid wood has flowed around chain-link fences all the time.
TNH: To specifically respond to your question about more than one sense of 'float', I dunno. I've seen it applied to modern plate glass as well as stuff that was decidely un-modern looking. Could be miscomprehension on my part, could be sloppiness on their part. Could be a modern process that deliberately undoes some of uniformity.
I'll take miscomprehension for $200, Alex.
Carol K: Thanks for the clarification - I was unaware of the distinction.
Falsback to highscool science class and the demo for glass sheets without a barrier between them sticking together over time. Fine at the start of the semester but by final exam, pain in the ass to pry apart.
Float glass was introduced by Pilkington in the late fifties; it's a very interesting process.
Sun-blued (or even violetted) glass is a result of electrons getting kicked out of iron impurities in the glass by high-energy photons, and then (because glass is a very viscous liquid) not being able to migrate back to the original atom site. Houseowners on Beacon Hill in Boston are proud of this effect because it attests to the antiquity of the property. Newer glass doesn't do this, though; additives such as manganese cancel out the iron.
There's an accurate explanation of how flight works in _Nature's Flyers : Birds, Insects, and the Biomechanics of Flight_. However, I find it a little too involved to summarize here, especially since the book is at home and I am not. (If I had a better technical understanding I'd probably be able to explain it more simply. However, I didn't take fluid dynamics in college. The closest I came was electromagnetic theory.) Also, I, coincidentally, just finished that chapter this weekend so I should probably read it again before even attempting an explanation.
However, if you're interested, the book, among other things, debunks the use of Bernoulli's principle to explain lift generated by wings and gives a scientifically sound explanation for the phenomenon.
This is one of my favorite bits of scientific misinformation, and a great example for pointing out the need to be skeptical about authoritative-sounding claims. In my experience, there's one surefire way disprove it in a sentence or so. Just observe that, if glass flowed quick enough to distort by millimeters over the course of centuries, those who make glass telescope lenses, which must be smooth and accurate to the scale of tens of nanometers, would find their work sagging into uselessness almost before it could be installed.
My understanding is that glass is amorphous - the atoms don't line up in regular patterns like they do in crystals. This makes it like a liquid, but doesn't make it a liquid. Or we'd be having very different experiences with glass jars holding instant tea.
I suspect that the glass-panes-sticking-together has as much to do with moisture and microscopic grunge sticking the pieces together as anything else.
Glass discoloring over time got mentioned many years ago in, IIRC, National Geographic: they were talking about 'desert glass', where it's been out in the sun for years, and has changed color.
I don't know how aeroplanes fly, although I have had it explained to me repeatedly. This is largely because I have also had it explained to me repeatedly (by my Dad, who worked for 35 years as an aeronautical engineer) that the explanation I can understand - based on Bernoulli - is wrong.
I don't know if this leaves me any further along than when I started. Mind you, I had been happily accepting the flowing-glass factoid without ever thinking too hard about it, so I'm glad I came by.
if the windows found in early Colonial American homes were thicker at the bottom than the top because of "flow" then the glass found in Egyptian Tombs should be a puddle.
Heh, great myth killer there.
Also, I suddenly have a desire for a "puddle of glass"...
I recall some short story from a year or two back about a guy who carved marble and his wife died and he blames the mountain for it (she fell? can't remember.) anyway, the mountain responds by having his marble slowly puddle. The story starts with a statue he just finished carving, and bits and pieces slowly "melt" and the guy is trying to figure out what the heck is going on, and by the end, there is a puddle of marble on his floor.
Helicopters don't fly.
They beat the air into submission.
Dan, crown glass -- that's the round stuff -- is thickest in the middle and thins further out.Hrm. John Baez' site (the "amorphous solid" link) says crown glass is thickest at the edges. Is there any chance the same term is being used for two different manufacturing techniques?
I always am surprised by why the Bernouilli-based explanation is regarded as easy to understand. It's not intuitive at all for me that faster-moving air should have lower pressure -- if anything, it looks to me like the curved upper wing surface might squeeze the air above it and drive the wing *down*. That's not what happens, but it seems more intuitively plausible than "fast-moving air sucks the plane up".
The cheap and almost-correct explanation based on angle-of-attack seems far more natural to me. Wings are basically ducts, canted downwards. When I put my hand out a car window at an angle to the ground striking the bottom of my hand and pushing my arm up. That's basically what wings do, except (this is the "almost-correct" part) that it turns out in practice that air bending over the top of the wing and down is the more important part of the flow. But the point is that wings drive the air down, and the plane stays up.
I just wrote "I always am surprised by why...", but meant "I always am surprised that..."
Well, nuts.
Here I am standing in this house, and I was sure I'd be okay if I threw the stone very slowly.
I don't know... While I'm willing to concede that it probably is a myth, I don't think you've provided evidence to that end.
NASA has a great stance on the situation which seems to be "Yeah, well... *sigh* ghod, this is going to take forever. Okay, so people talking about Bernoulli aren't exactly right, but they aren't exactly wrong either. [sotto]why did I get myself into this...[/sotto] Let's just say it's way more complicated".
My grandfather was a potter by trade, and mixed his own glazes; I didn't learn until after he died that he was esteemed in some circles for the science of his glazemaking almost as much as for his artwork. This thread's making me wish I'd had the patience, years ago, to learn some of that strange alchemy from him.
Also, thanks to Xopher, I now have a new toy on my wishlist: a transparasteel lute.
Dan...I hate to break it to you, but while metallic glasses have many properties associated with glass, transparency isn't one of them. Silica glass is transparent because silica is (or can be). The metallic glass they showed on the TV show I saw was all shiny but quite opaque.
Party pooper. Next you'll be telling me I can't have a flying car, either.
Arg. Air traffic controllers are still taught that Bernoulli explains flight. I was deeply offended to learn, quite recently, that this is incorrect. I cannot recall the exact true mechanism; it is complicated enough to not be a strong competitor with Bernoulli.
Xopher: You mean I can't have that transparent-aluminum aquarium, either?
Xopher, transparent aluminate glass was created in 2004:
The article has a good explanation for why it's so easy to make glass from silica and so difficult to make from other materials.
I thought it was manganese or magnesium, one of those M elements, that turned glass purple. But I'm remembering bits from the Seattle Underground tour I took years ago.
On glass flow, I was told earlier this year, while I learned to break knives for microtome sectioning (so much fun!), that the knives will dull themselves over time. They're as sharp as they'll ever be right as you break them. I assumed it was a weirdness of glass, but didn't connect it to the glass-flow bit. I have heard that knives will do much the same thing, returning to a duller or sharper shape, but I haven't observed it myself.
It's an important trope in Walking on Glass by Iain Banks (or is it Iain M. Banks; always get confused between the two). One of the protagonists, trapped in a castle, discovers that the glass in the windows has begun to bulge out at the bottom of the frames, demonstrating how long the castle has been in existence or something.
IIRC the quick way to discredit the Bernoulli effect on the carefully shaped airfoil as the only thing holding a plane in the sky is to fly upside down. A quick way to show that the Bernoulli effect is somewhat involved is to try and takeoff with a wee bit of frost on the airfoil, spoiling the laminar flow of air over the wing.
If you look at the panes of really old glass that were thicker at the bottom, the edge should still be fairly sharp (but I don't know if the glass would have been annealed after cutting back then-- which might soften the edge a little bit).
Congealed tar can be liquid enough to flow over long periods of time as this IgNoble winning experiment shows .
Bah. Of course glass doesn't flow anymore, it's now made by boring people. It was much more fun when you could only get glass from mediterranean or arabic people... then it was flowing like water, creating shapes of exhilarating irregularity, and each time the artisans blowed into their pipes, you could get a cursed wine glass to poison your enemy, a fragile bottle to capture foolish gods, or a decorated window to see the future through.
(Then you could buy these wonders and they'd magically become worthless antiques, and very opaque in several areas, but of course that was not the point. Jeez, who'd want glass that doesn't magically change?)
Giacomo--
Yep. See A.S Byatt's "The Djinn in the Nightingale's Eye."
glass is a supercooled liquid,* or an extremely viscous liquid
Why does the asterisk point to a website about wooden shingles? I couldn't find a reference to supercooled liquid anywhere on that page.
Dan Layman-Kennedy: Flying car? I was hoping for a flying carpet.
transparent aluminate glass was created in 2004:
Oh no! That means that when Kirk, Spock, and the crew comes back to get some humpback whales, Scotty won't be able to barter a secret formula for transparent aluminum for six inch stock of plexiglass. Maybe we should start a fund, have a bake sale.
Giacomo, can I come visit your universe? Sounds like my kind of place exactly.
I'd just like to express my admiration for the word "flooped."
Thank you.
Cassie:
I believe that microtome blades made of broken glass get dull because of chemical attack from the atmosphere - the sharp edge would be highly susceptible because it's almost atomically sharp (IIRC) and those atoms are hanging out there naked - they'd really rather be huddled inside with their neighbours or joined up with one of the O2 or other molecules floating around. In addition, the freshly-broken surface will have some cracks, which will grow, and will certainly accumulate cracks when you brush against them, set them down, etc. so that will take off the edge.
Actually, come to think of it, freshly-broken microtome blades, in use, may be one of the few cases where the glass flows to a practically-important degree under load. Glass under load can either break due to cracks or it can (in the absence of cracks, and under high enough load) deform, the same way that metals do. Obviously, in most situations, the breakage due to cracks wins handily over the deformation. But in this case, since the edge of the glass blade is so thin, it's possible that the stress (force divided by area) is high enough that it actually flows. I'd have to get some numbers to check that as a possibility, though.
This is an awesome book that talks about how cool freshly-drawn or broken glass is; it's one of the things that made me grow up to be a materials scientist. Giacomo, apologies for being one of the boring people.
Oh, and d'oh! I forgot to check out the amorphous metal when I was in the lab. Sorry, Xopher - I'll try to remember to check it out later.
Oh no! That means that when Kirk, Spock, and the crew comes back to get some humpback whales, Scotty won't be able to barter a secret formula for transparent aluminum for six inch stock of plexiglass.
Egads! I never thought of that! This discovery could change the course of history! Or the future. Or something like that.
Flying car? I was hoping for a flying carpet.
Well, if you want to keep a bird in your car, that's up to you. But it's going to make driving a little difficult.
Dolloch says: NASA has a great stance on the situation which seems to be "Yeah, well..."
Well, okay. The "Bernouilli" and "Newtonian" views are complementary, like different frames of reference, and you can get to a correct answer either way. It's true that if you look at the direct force balance on the airplane, the reason it stays up is that the pressure on the upper surface of the wings is lower than the pressure on the bottom surface. And the reason it's lower is because the air is flowing faster there, and Bernouilli's law applies. But the next step in that logic -- why is the air flowing faster on the upper surface of the wing? -- is incredibly non-obvious. The Coanda effect comes in, and maybe some other stuff -- fluid dynamics has never been intuitive to me.
The "Newtonian" wings-are-ducts view is also only sort-of-correct, but it seems to me it is both easier to visualize and gets closer to the truth without going over, as it were.
IIRC the quick way to discredit the Bernoulli effect on the carefully shaped airfoil as the only thing holding a plane in the sky is to fly upside down.Yeah, but, you first.
Argh, the "Bernoulli is bunk" thing. *crycry*
This one drives me crazy. I'm glad that Colin Roald already started with an explanation, because I might have gone crazy if I was starting from scratch.
1. Your average jumbo jet DOES NOT produce enough thrust to keep itself aloft purely by forcing air downwards. Those are big engines, but the airplane is also very heavy. If you point a jumbo jet upright, so those big engines are working directly against gravity, the plane will not fly. (Some airplanes do have engines strong enough to do this, most obviously the Harrier Jump-Jet, which doesn't even rely on tricky vortices (and/or whatever) for lift the way helicopters do (I actually don't know anything about helicopter flight, I've just heard that "vortices" are involved, so take that for what it's worth (nothing))).
2. If your explanation begins and ends with forcing air downwards, then with the information presented in 1, you lose, you are wrong. Because there is no free lunch. Using wings to force air downward is (Newtonian) mechanically equivalent (on the body as a whole) to using jet engines to force air downward, and would require the same amounts of thrust to remain aloft.
3. Why don't those complaints stop an airplane from flying with "airfoil explanations"? Because they are using ForceA to create a pressure differential which exerts ForceB (in this case, due to the design of the wing, B > A). Which is (Newtonian) mechanically different from using ForceA to create (or cancel) an acceleration.
4. The pisser is that it's not simple. Some things in the is world are simple, airfoil flight isn't one of them. Fluid dynamics are complex (especially to somebody like me that never got down to any sort of unifying principles) and I haven't studied them [adequately]. Like Colin Roald said, you use (complicated) fluid dynamics to create a pressure differential which has creates a simple Newtonian equivalence of forces (weight & airfoil lift).
Like the "bees shouldn't be able to fly" story, the moral is that when a simple model predicts something shouldn't work, but reality demonstrates that it does, the model is leaving something out. In the case of the "Bernoulli explanation" of airfoil flight, the thing that's being left out is the creation and preservation of streamlines.
I think I am too wicked and deserving punishment for my sinful thoughts. I actually read Xopher's last comment without ever thinking about small animals with wings.
The best paper airplane in the world
I was going to point out how it can fly despite having completely flat wings, but the heck with it. Paper airplane, wahoo!
Thus far debcha: This is an awesome book that talks about how cool freshly-drawn or broken glass is; it's one of the things that made me grow up to be a materials scientist.
Seconded! (And, in fact, I opened the link thinking, that's going to be him, isn't it...) I first encountered this one when I was 17, and it made sense of a lot of my A-level chemistry and physics, and then kept right on making sense of chemistry at university. (I still remember the look on my colloids lecturer's face when I finally "got it" - and blurted out "oh, like metallurgy, I see now!")
I ended up researching nanomaterials rather than the kind of things he writes about, but it's still the same basic set of principles, and I partly blame this book for my interest in history and archaeology as well - historical manufacturing techniques and the evolution of, well, Stuff.
Cassie wrote: I thought it was manganese or magnesium, one of those M elements, that turned glass purple. Yep - manganese gives purple/pink/brown, all those shades. You see it in solutions in the lab too, and I actually think of evening city skies as being "manganese-coloured" because of it.
Most glass naturally contains iron, so classically, glass in the State of Nature is green or greenish - manganese counteracts this, and more turns it mud-coloured. Cobalt turns it blue (surprisingly enough), gold turns it rosy pink or red, and if you use uranium, you get a bright vivid green. (This information brought to you by the Glass Gallery of the Victoria & Albert Museum, one of the most amazing parts of an utterly wonderful place.)
"IRC the quick way to discredit the Bernoulli effect on the carefully shaped airfoil as the only thing holding a plane in the sky is to fly upside down."
It only discredits Bernoulli if it's the only thing holding a plane in the sky. I always thought that if you just tilted the wings up, you could fly - but it would take a tremendous amount of power - its terribly inefficient. The Bernoulli effect gets you the efficiency to make it practical.
So, if flying a plane upside down is as efficient as one flying rightside up, then I'm all wet.
Well, I had fallen for that thinner-thicker and therefore fluid talk all these years, and am actually relieved to hear it debunked.
Mostly though, I'm just beaming with delight to have learned that viscosity is measured in units called poises.
Charming word. Call me silly, but I'm as gobsmacked as my poet-father was the first time he heard me use the word "ormoulu" casually in a sentence.
Poise. What poise.
How sticky is that glue, in poises, please?
Let me measure, en pointe.
-hee!
...if you use uranium, you get a bright vivid green...
I'd hate to be a collector of 'canary glass' in this day and age. Can you imagine travelling through an airport with your latest finds - only to start some TSA radiation detector a-clicking?
"No, it's not a dirty bomb, dammit - it's my collection!"
I've seen all sorts of science books that use the Bernoulli effect. It's been persistent ever since the early days of aviation when the military, in the best sense of "military intelligence," decided to use it to explain airplane flight in their manuals because they deemed it easier to understand than the real Newtoniaon explanation. The Wrights, Otto Lilienthal, Glenn Curtiss, and just about all pioneering aviation innovators knew better, but the explanation has persisted because of the military's use and because the diagram is so pig-simple to duplicate. It's become iconic.
It's not that the Bernoullli effect doesn't occur over a wing. It's that by itself it's insufficient to explain lift.
Bernoulli diagrams nearly all show an airplane wing that is curved on the top side and flat on the bottom. I have yet to see a wing on a real airplane (or on a bird) that is shaped that way.
Even if they were, if the Bernoulli effect were all that was holding a plane up in the air, how could stunt planes fly upside-down?
If you think of a helicopter blade as a moving wing (since that's what a propeller is, by its shape and its function of moving air), and stand under it while it's moving, you can feel with no uncertainty exactly what it is that's moving that helicopter into the air.
Oooh, thank you for the book recommendation. I learned a lot about slicing and breaking during my time at the microtome, and I like things being solid.
Bernoulli's one of my favorite effects for everything because it showed up in almost every biology class I ever took. Blood cells congregate in the center of a vessel? Tunicates, sponges, et cetera with their currents? All Bernoulli. That it helps keep planes in the air is incidental.
Oh, well, if we're upping the ante to flying carpet, then what I'd really like is an oud made of crystal and dragonbone, strung with frickin' moonbeams.
(And while we're at it, an enchanted plectrum that lets me play "Discipline" on it.)
Taking advantage of my materials science soapbox-for-a-day, I'd also like to recommend Mark Eberhart's lovely book, Why Things Break, a more modern and accessible book about, well, why things break. It's only a few years old, and is a worthy update to my beloved old Gordon. Eberhart talks about everything from Pyrex to how to design plutonium-based energy sources for spacecraft so that they can survive crashes to what the difference is between Boston snow and Colorado snow.
Okay, I really have to get back to putting my mat sci test together, so I can go home while it is still tonight and not tomorrow morning.
Xopher: Where I come from, having a bird in your car would definitely distract from driving; but not for those reasons.
Dan Layman-Kennedy: That is one hell of an oud!
returning to the glass for a second, it seems entirely sensible that the 'flow' argument is both i) an intuitive explanation for the layperson observing the thick/thin phenomenon on old windows, and ii) entirely wrong. the structural argument for orienting tapered panes as base-heavy is compelling; it seems to me, though, that the link to 'proof' offered by chris clarke (in this article) only explodes the myth as far as the windows go; the article also states that glass does in fact flow. in a timeframe comparable to (if not significantly larger than) the age of the universe, to be specific. the problem here is the timescale involved, but we already knew that; the age of cathedrals is way too short to give adequate glass flow, but this does not mean glass flow doesn't happen at all. is there a 'fluidity threshold' included in the definition of a liquid? if so, chances are it is a measure relative to the lifespan of the observer.
Odd. I could've sworn I remembered an Educational Program from long ago demonstrating "liquid glass" by putting a weight on a windowpane and showing how the glass bent.
I wonder what childhood memory I'm distorting?
Consider, if you will, the obsidian layers that come out of shield volcanoes; those crack, rather than flowing, despite having had millions of years to slump into strange shapes if they so desired.
Oh, and as I recall, float glass is floated on zinc, not tin, and that the critical difference -- aside from needing about a third the distance and it costing much less -- between float and plate glass is that with float glass, it's picked up on the rollers hot, and vertically; that way it sets fast and smoothly, rather than slowly and ripply and needing to be ground flat.
Kinda like why plagiaclases have big crystals.
(I actually don't know anything about helicopter flight, I've just heard that "vortices" are involved, so take that for what it's worth (nothing))).
vortices, rubber bands, paper clips, and a particle accelerator. Oh wait, that's how I became immortal. never mine.
4. The pisser is that it's not simple. Some things in the is world are simple, airfoil flight isn't one of them.
Well, some things really are simple, like actually flying an airplane. I like the fact that birds can fly and don't bother with the explanations. I also like that with all the complications to flying, one of the most important "instruments" is a bit yarn dangling outside in the middle of the windshield for your slip indicator. Sometimes its teh little things that count.
Graydon: Thx, that explains a lot.
It's been persistent ever since the early days of aviation when the military, in the best sense of "military intelligence," decided to use it to explain airplane flight in their manuals because they deemed it easier to understand than the real Newtoniaon explanation.
You're almost approaching truth here, be careful. The truth is that SIMPLIFIED EXPLANATIONS of the lift by pressure differential are presented in text-books. The NASA articles that Doloch (Thanks Doloch!) linked above are pretty clear on the topic. The "equal transit time" explanation doesn't reflect reality. Yes, equal transit time is in textbooks. Textbooks also say that the sun is composed of very hot gasses. Neither is true, but that doesn't mean that the sun isn't very hot, or that pressure differentials aren't the primary source of lift in airfoil flight, or that the Bernoulli effect isn't important to the creation of those pressure differentials.
If you think of a helicopter blade as a moving wing (since that's what a propeller is, by its shape and its function of moving air), and stand under it while it's moving, you can feel with no uncertainty exactly what it is that's moving that helicopter into the air.
That is an utterly inappropriate analogy. Amusingly, Helicopters don't produce enough Newtonian reaction upforce (by pushing air down) to keep themselves aloft.
Check this out, and pay attention to the part about autorotation... with air moving UPWARDS relative to the blades of the helicopter, they still provide lift. You will become a source of great amusement to me if you suggest that the upflowing air supports the helicopter in the manner of a parachute or kite. (just making sure that you are adequately warned)
I remain of the firm opinion that we simply have no real idea how the universe actually works, and that the universe likes to change how it works when we're not looking (kind of personifying quantum mechanics, so to speak). We make bold guesses, which are then "proven facts" until years, decades, even centuries later, someone comes along and decides that we're wrong, and it's actually x.
I'm perfectly content to let glass be its quirky self. It's not a fluid, but it's not a normal solid either. It doesn't "flow" per se, in most situations, but I've seen fractured glass heal itself completely before (while I didn't put it under a microscope, under a jeweler's loupe at least, it was entirely gone). I don't really care what we use to explain that, I just think it's neat. :)
nmrboy:
(this conversation was too fun to stay away from)
Glass flow happens, of course, just not at room temperature. And since glass doesn't have a sharp melting point (like ice), there has to be an arbitrary 'fluidity threshold.' But it's determined by how easy it is to work the glass (blow, pour, whatever), not by our lifespans (although I guess there is an indirect relationship between our lifespans and how long we are willing to wait for our glass to spread out).
Here's a graph showing the viscosity of glass as a function of temperature (curve (d) is regular soda-lime silica glass). Note that the graph is the logarithm of the viscosity (1 = 10, 2 = 100, 3 = 1000, etc.) and that the viscosity goes up as you go up the y-axis - low values of viscosity = runnier.
The melting point is defined as 10 Pa.s ('1' on this y-axis, but the curves don't go that far down). The working point is defined as 10^3 Pa.s, a viscosity at which glass is easily deformed. The minimum temperature at which the glass can be handled but still hold its shape is the softening point, corresponding to a viscosity of about 4 x 10^6 Pa.s (or about 6.5 on the y-axis); the temperatures between the softening point and the working point are the working range. For soda-lime glass, the melting point is around 1400 deg C, and the working range is between about 700 and 1000 deg C.
Given the viscosity and the load (ie stress) caused by the weight of the glass, you can estimate the time for a given flow. Note, however, that the flow of glass at room temperature and under low loads is small enough to be not of practical importance, which is why temperature-viscosity curves don't normally go down to twenty degrees Celsius (and presumably why it took until 1998 to debunk this particular myth).
(I should credit my mat sci text for these criteria and numbers)
And then there are planes like the F-117, which is essentially a wedge with large engines and a computer that makes it fly. Without that computer, it'd be about as nimble as a brick.
Sorry, I don't know why that link didn't work. Here's the graph.
I was enchanted with the "supercooled liquid" theory of glass, because it made it seem like glass existed on a timescale all its own.
I was taking a lot of glass art classes at the time, and the idea of working in a medium that inhabited its own universe lent an added note of mystique to what was already fascinating stuff. Until an instructor debunked the pleasant fantasy by telling us the points mentioned here: old glass is uneven because of the manufacturing methods of the time, and if glass really was a supercooled liquid, then all those glass artifacts from the Ancient World should be puddles.
Now that I think about it, it's probably just as well glass isn't a mystical magical element with its own timeline. Working with glass is fraught with enough complications. Along with differences in heating and cooling temperatures, expansion and contraction rates, different vulnerabilities to heatshock, and how you can't mix glass made by different manufacturers because they all use different formulas - glass artists would probably also have to know and remember different "ooze" rates, if glass really was a supercooled liquid :)
Melissa Mead said:
Odd. I could've sworn I remembered an Educational Program from long ago demonstrating "liquid glass" by putting a weight on a windowpane and showing how the glass bent.
I wonder what childhood memory I'm distorting?
Steel bends under weight. If you piled enough heavy stuff on top of a steel I-Beam, eventually you'd have a puddle of steel below a bunch of heavy stuff (Steel is malleable (though much less malleable than iron)). This is similar to what you're seeing when glass is bent under a weight.
The question presented in the post was, will glass deform under its own weight, and will congealed (physically indistinct) middle-section droplets deform (away from nearby upwards neighbors) under their own weight. I'm not a materials scientist, but I'll take their word for it when they say "no."
Greg London said:
I like the fact that birds can fly and don't bother with the explanations.
I spend (some) days wandering around in a stupor, too impressed with how much easier it is for animals to do things than to understand how they're done. The cognitive complexity of knowing when you're looking at a car, (much less a pencil, a computer monitor, a plate of spaghetti, or a towel crumpled up on the floor) still make me want to go back into academia and study Cognitive Science. Coordination for physical action is stranger and harder than most people imagine... by contrast, knowing how long you have until you're hit by a car is simpler than you would be taught to solve that problem in trigonometry class (but it's still harder to calculate on paper than it is to know when something will hit you).
Okay, one last thing before I head home and to bed, about broken glass fusing back together.
Nabil, I wrote above about how atoms like to be bonded with other atoms. When you break a piece of glass, all the atoms on the surface are still happily bonded to the atoms below them, but they have naked bonds waving in the wind where the other surface used to be. Since glass is so brittle - that is, it barely deforms at all before it breaks - the two broken pieces can fit back together with atomic-level closeness. So if you put the pieces back together again immediately, the bonds can join up again and fuse the glass. However, if you wait any length of time, atmospheric gases will bind to the atoms (remember, those dangling bonds really don't want to be out there, and they'll bind to anything within reach) and the glass will no longer fuse.
I've never actually tried this, and now I want to go over to my lab and give it a shot...
On the subject of groovy glassy things.
In the part of Texas where I grew up, we had a lot of lightning storms with no rain. The strikes would sometimes hit in sandy patches out in the pasture, and we'd get what we called "lightning glass."
Apparently they're actually called fulgurites, and can look something like a piece of coral if dug up carefully enough.
http://www.menzelphoto.com/gallery/big/lightning6.htm
So cool.
transparent aluminate glass was created in 2004
Oh no! That means that when Kirk, Spock, and the crew comes back to get some humpback whales, Scotty won't be able to barter a secret formula for transparent aluminum for six inch stock of plexiglass. Maybe we should start a fund, have a bake sale.
No, the time line is completely consistent:
A) Kirk, Spock, Scotty et al travel back in time to the 1980s.
B) A few years later, transparent aluminum is "discovered".
Wow. I didn't mean to run away after tossing a grenade.
The bogus lift explanation that I do my best to shoot down is the one NASA labels "Equal Transit." It uses Bernoulli's principle, and only that, to explain lift. I think it's perpetuated in classrooms and textbooks because there are so many simple, dramatic classroom demonstrations of Bernoulli's principle. As others have already said at length, lift isn't that simple.
Colin Roald linked to a great explanation of lift at FIU. The NASA flight pages also have a lot of great content. Take a wander and play with their Java applets.
my favorite SF glass idea was, i think, from a Gene Wolfe book where we are told that the sand is composed of the tiny bits and pieces of what was once the glass in windows and containers. it did a great job of showing how far in the future this civilization was. I've never forgotten that description, even if I have forgotten exactly where it's from.
Most of the physics here is beyond my feeble understanding (even though I enjoy bemusedly skimming the more learned posts), so what struck me most were the few teasing comments about liquifying wood. If in Norway wood has the consistency and behavior of glaciers, this would conjure up new images for the "so I lit a fire" line in "Norwegian Wood". (Whooee, all her dern furniture melted!)
IIRC, one of the earlier computer graphics gurus (Blinn? Sutherland?) said they'd have achieved success when they could simulate a wood scarf draped over a silk table. I don't think that's exactly the kind of flowing wood that was meant, but ...
It's an important trope in Walking on Glass by Iain Banks (or is it Iain M. Banks; always get confused between the two).
The one, actually. (I'm startled that I am the first person to respond to this.) The way I heard it, he was told at one point by a Publishing Person that he needed a different name for his SF stuff than his non-SF; the middle initial was his solution. Amazon tells me that the M. is in SF books; I will probably reforget this information immediately.
Scott writes: Amusingly, Helicopters don't produce enough Newtonian reaction upforce (by pushing air down) to keep themselves aloft.
Do you have a reference for this, or for how autorotation actually works? The idea that rotors would continue turning in the same direction after the flow of air reverses seems entirely counter to my experience with fans.
A) Kirk, Spock, Scotty et al travel back in time to the 1980s.
B) A few years later, transparent aluminum is "discovered".
Why would anyone who is sitting on a discovery worth billions of dollars stick it in a drawer for a couple of years? I have operated under the assumption that as long as transparent aluminum is not available, that means Kirk and crew haven't come back in time to get the whales yet. Once it becomes available and is "discovered", then I'll know that Kirk was just here (or in time-travel speak, Kirk was just now.)
Why would anyone who is sitting on a discovery worth billions of dollars stick it in a drawer for a couple of years?
Not sticking in a drawer - going through the patent process. And probably developing the large-scale manufacturing processes, too. At which point the Monterey aquarium replaces the current version of Scotty's windowpane with a sheet of transparent aluminum and we find out how good it really is.
Do you have a reference for this, or for how autorotation actually works? The idea that rotors would continue turning in the same direction after the flow of air reverses seems entirely counter to my experience with fans.
autogyros are aircraft with propellors on the front and a helicopter rotor on top. The rotor is only powered while the aircraft is sitting on the ground, and that is simply to get the thing spinning. Once spinning, the clutch is released, and the rotor remains unpowered for the rest of teh flight. This can be seen to be intuitively true because if the rotor were powered, there would have to be an anti-torque tailrotor to keep an autogyro from spinning while in flight.
The main rotor gets energy to spin by the airflow that goes up through the rotor, causing it to windmill. This also creates lift, which keeps you from converting into a small grease spot on the ground.
The propeller on the front of an autogyro allows you to add energy that is lost by drag, inefficiencies, and what not. It also allows you to add extra energy so that when you climb, you can maintain the same airflow up through the rotor system and keep the amount of lift.
When a helicopter loses its engine, you pull back on the cyclic like reining in a horse and drop the collective. Pulling back forces the air to go up through the rotors and dropping collective reduces drag so you don't lose rotor RPM. In a second, you've gone from powered flight with air going down through the rotors to unpowered flight with the air going up through the rotors. How the hell this works from an aerodynamic point of view, I have no clue. I am like the bird who can fly but doesn't know the math.
In a helicopter, you then trade off altitude-energy to offset your rotor-drag energy that you're losing. This allows you to autorotate unpowered safely to the ground from high altitudes. As you autorotate, you must maintain forward velocity to keep the airflow going the right way. up and through the rotors, which keeps them spinning.
If you lose an engine while you are hovering, I seem to recall that the procedure is to go cyclic forward because you have to achieve forward velocity or you're dead, and drop collective as before. I'm a little rusty on that because it's just stupid to hover at altitude.
Then when it comes to actually landing, you are basically heading towards the ground, nose down, gliding in, and around 30 feet from the ground, you pull back on the cyclic which causes the helicopter to point up, the air is suddenly pushing on the blades like a windmill, which spins them a bit faster and stops your forward velocity. The faster spinning will actually cause you to gain a bit of altitude, and when you've bled off the last of your forward velocity, you cycle forward, and land straight down, cranking the cyclic just before you crash, er, touchdown.
Take a spoon, put the scoop end on the table, and hold the handle end in the air at a 30 or 40 degree angle. turn it so the scoop end is upside down. And that approximates your ideal flight path when you lose an engine in a helicopter.
Not sticking in a drawer - going through the patent process.
Hm, that could be. OK, so as long as transparent alumninum is not available on teh market, I can assume that Kirk and company didn't come back to get the whales over two years ago or more. He may have come back in the last two years. Or, he may not yet have come back.
The important thing, though, is that the Star Trek universe remain true.
when you've bled off the last of your forward velocity, you cycle forward, and land straight down, cranking the cyclic just before you crash, er, touchdown.
ack! that should say "cranking the collective" not cyclic. Good grief, I hope no one had an engine out after reading that. I'm a bit rusty.
Greg London writes: When a helicopter loses its engine, you pull back on the cyclic like reining in a horse and drop the collective.
"Dropping the collective" means changing the angle of attack of the rotor blades?
Meanwhile, I remain unconvinced that this from Scott can possibly be true: "Amusingly, Helicopters don't produce enough Newtonian reaction upforce (by pushing air down) to keep themselves aloft. Check this out, and pay attention to the part about autorotation... with air moving UPWARDS relative to the blades of the helicopter, they still provide lift." Greg's comment that "you have to achieve forward velocity or you're dead" makes me question whether the airflow is actually upwards relative to the moving rotor blades, but even if the air has relative upward velocity before encountering the blade, it may still be redirected to a downward velocity in the course of the encounter. Fundamentally, if you draw a surface of integration at some reasonable distance around the chopper, *something* must balance the tonne of excess mass contained within that surface. Either somehow the chopper maintains a substantial pressure gradient that can integrate out to counter a tonne of force over an arbitrary enclosing surface, which I think is impossible, or else (1 tonne)x(10 m/s/s) of air is somehow being accelerated downward through the surface. Fluids are weird, but you still don't get a pass on Newton's laws.
"Dropping the collective" means changing the angle of attack of the rotor blades?
Yep. both cyclic and collective change angle of attack.
The cyclic changes it so there is more angle on one side than another, causing the helicopter to tilt away from the big angle (more lift) side.
The collective changes the angle on all blades simultaneously, so the helicopter goes up or down.
Rotor RPM is constant, so to go up/down you change the collective pitch of the blades. and to go forward/backward/sideways, you move the cyclic.
makes me question whether the airflow is actually upwards relative to the moving rotor blades,
Hm, there's a diagram here.
When you're flying, you generally don't think in terms of integrals and gradients. You're thinking "airspeed indicator, altitude indicator, rotor RPM, engine rpm. Big sky. repeat." Meanwhile, your hands are constantly trying to balance a thousand pound spinning plate on a yardstick.
I don't know the math of autorotation, but the thinking involved in actually performing one is the function that graphs the line from "oh f**k!" at the origin and "shiiiiiiiiiiiiit" at the end point.
ah, bigger drawing here.
note the coordinate axis is labeled "plane of rotation" and "axis of rotation". The x axis is the rotor disc, the y axis is the shaft that turns it. The arrow marked "relative wind" goes from lower left to upper right. It goes up through the rotor disc.
also, the little diagram on the right shows the aircraft, the rotor disc tilted back, and the airflow going up through it.
My brother =taught= autorotation when he was a helicopter flight instructor for the Army -- so imagine not only deliberately turning off your engine to practice it, but being the "driving teacher" sitting in the copilot's seat while some greenhorn pilot tries it for the first time....
(Note to self, remember to admire brother the next time I see him.)
Early industrial-era glass was made by floating molten glass on a bed of molten tin.
later someone mentions zinc, not tin.
No one ever bothers to explain why the heck the tin is molten or why you wouldn't pour the glass on something that isn't molten but really smooth.
Ah, so the incident angle of the wind is shallowly up through the disc. But I don't think that figure is drawing streamlines of the airflow, just the angle of incidence. I would wager the flow is bent substantially downward after contact with the blades.
And I'll speculate that some game played with the cyclic is what lets you use a mostly-forward-incident-wind to keep the rotor spinning even when unpowered. Would that be the effect of "pulling back on the cyclic," which Greg says is the first thing you do when thrown into autorotation? I think it might be.
Colin Roald:
Either somehow the chopper maintains a substantial pressure gradient that can integrate out to counter a tonne of force over an arbitrary enclosing surface, which I think is impossible, or else (1 tonne)x(10 m/s/s) of air is somehow being accelerated downward through the surface. Fluids are weird, but you still don't get a pass on Newton's laws.
I wish I knew enough about helicopter flight to answer your question. And I'm very glad that Greg London stepped in with answers I don't have even vague access to.
The autogyro article that Greg London linked shows the rotor blades as airfoils, so that maybe helps? I really don't know though, sorry. I'm just a big pile of second, third, and aribtrarily half-remembered information about flight.
I'll speculate that some game played with the cyclic is what lets you use a mostly-forward-incident-wind to keep the rotor spinning even when unpowered. Would that be the effect of "pulling back on the cyclic,"
Trees pretty. Fire bad.
(someone on this board should recognize that reference and to what end it was made.)
Uhm, I think the deal is that when you're moving under power in a helicopter, you're nose down and the wind is going down through the rotor disc. When you lose power, you need to get the air going up through the disc and you pull back on the cyclic to rock the fuselage back.
A quick google of both tin and zinc gives tin with a melting point of 231.93°C and zinc at 419.53°C. Cooler than molten glass, at least. (I want to say 'way cooler' but it doesn't seem to fit. Glass is way cool, though. I've seen some Roman stuff; having been buried, it hadn't changed color much.
I am perfectly willing to believe glass flows down despite the facts. I realize this belief is a slippery slope that ends in enormous oil profits, an increased federal deficit and lots of dead soldiers, but, at least in this case, our children will think it was cute.
Oh, and planes don't really fly. Big metal things in the air? That's just crazy. It's magic.
No one ever bothers to explain why the heck the tin is molten or why you wouldn't pour the glass on something that isn't molten but really smooth.
I'm speculating, but could it be because nothing that isn't molten is as smooth as molten metal?
Greg:
As far as I know, it's normally molten tin, not zinc, that's used to make float glass (sources are the mat sci textbooks I've mentioned above, as well as Wikipedia).
Early sheet glass was, in fact, made by pouring glass on very flat sheets of metal (like iron). And the reason why you use molten metal is because a layer of liquid on top of another liquid allow
Comments on Glass flow: