Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] real-world physics

In the grocery store this morning, I saw a handicapped person
zooming down the aisle. She was driving an electric scooter
with her right hand, and dragging along a shopping cart using
her left hand. The two conveyances were moving as an ensemble,
side by side.

In the middle of the aisle, there was a pillar. At the appropriate
point, she let go of the cart. The cart passed to the left of
the pillar while the scooter passed to the right. She re-grabbed
the cart and kept going without missing a beat. Newton's first
law works!

Of course this requires more than Newton's laws; it also requires
cart-wheels that are reasonably well lubricated and undamaged.
Still, under favorable conditions, it works.

It also requires multitasking, to be able to steer two things at
once ... but that's a whole nother topic.

====================

As a more complicated example of real-world physics, here's a riddle
for you:

Once upon a time, I was serving as a poll worker, one of the guys
who sits behind the desk and hands out the ballots.

We were keeping the door to the room wide open. Part of the reason
was the symbolism: We wanted the place to look wide-open and welcoming.
Another reason was handicap access: It was a wide, heavy, opaque,
spring-loaded door, and it would be unacceptable to have a handicapped
person stuck outside struggling with the door.

Everything was going fine until about 3:30 in the afternoon, when
it started getting unpleasantly warm in the room. It was bad and
rapidly getting worse.

a) The other poll workers suggested that I turn down the thermostat
a couple of notches. I declined, since the AC was already running
flat-out. Telling it we /want/ it to work harder would not actually
make it work harder. To say the same thing in more technical terms,
the feedback loop had long since unlocked. Changing the set-point
would only make it more unlocked, with no effect on any of the
variables we cared about.

b) Then they suggested I turn on the Casablanca-style overhead fans.
I declined, because I figured the fans would just push the air
around, pushing the remaining semi-cool air out the door.

So the riddle is, what *should* be done in such a situation?

Here is a diagram with some additional information:

_____________________________________
(north) | |
| |
| |
| rather |
[snip] large [snip]
| building |
| |
| |
| |
|_____________________________________|
| |
| |
door | |
folded | | |
open | | table "B" |
(other stuff) |
|
| |
| table "A" |
| |
(south) |_____________________________________|


Huge hint: The tablecloth on table "A" was fluttering. This was
a clue that it was windy outside ... even though it had not been
windy earlier in the day.

Huger hint: The tablecloth on table "B" was not fluttering.


Answer: I wedged the door open /at a shallow angle/ as indicated
in the diagram below. I immediately observed that the tablecloth
stopped fluttering.

Result: The temperature plummeted.


_____________________________________
(north) | |
| |
| |
| rather |
[snip] large [snip]
| building |
| |
| |
| |
|_____________________________________|
| |
| |
| |
| |
| table "B" |
door / (other stuff) |
/ |
| |
| table "A" |
| |
(south) |_____________________________________|


The physics works like this: The wind was coming pretty much straight
out of the west. However, streamlines cannot flow through the building.
When the air hits the building, it splits. Half of it flows around
the north side and half around the south side.
http://ars.els-cdn.com/content/image/1-s2.0-S0196890411000070-gr3.jpg
Since we were near the south side of a rather large building, I assume
there was pretty much simple laminar flow along the exterior wall.

In the original configuration, with the door open 180 degrees, there
will be friction between the moving air just outside the doorway and
the stationary air just inside the doorway. This will cause the
boundary layer to extend into the room. When the moving air hits the
door jamb on the downwind side of the opening, wackiness ensues. Lots
of mixing.

In the improved configuration, with the door open only about 50 degrees,
the door serves as a ramp. It helps the streamlines jump over the
opening. If you calculated the optimal configuration, to get the best
aerodynamic effect, you would use a ramp with a lot less chord than
the full width of the opening ... but using the door itself was more
convenient, and it worked.

If you wanted to actually predict and/or optimize what happens in this
situation, it would take days of work and hours of computer time. I did
not actually know in advance that this trick would work ... but I knew
it /might/ work, and the experiment was super-easy to do, so I gave it
a try. There was immediate confirmation, because the tablecloth stopped
fluttering.

Also note that this door configuration cut down on the amount of sunlight
entering the room, which lowered the heat load by thousands of watts. So
this was a win/win situation.

The door was sufficiently wide so that even in the not-fully-open state,
people could come and go without any inconvenience, without any temptation
to open it wider.

Pedagogical note: It is worth examining the thought process that led
to this result. The short answer is that I thought of it in terms of
stream lines. I have enough experience with fluid dynamics that I could
sorta guess what the streamlines would do. This experience comes from
many hours of doing computer simulations and thinking about the results
... and also from attaching tufts of yarn to airplane wings, sailboat
sails, and various other things.

I emphasize that the streamline picture was the only thing I was
/consciously/ aware of at the time.

However ... when I was preparing to send this note, I went looking for
pictures of streamlines (as in the link given above). I stumbled across
articles about the SOFIA project, which is a telescope that looks out
through an aperture in the side of a 747.
http://www.flightglobal.com/news/articles/truth-or-consequences-testing-the-door-on-nasas-new-window-on-the-319307/
As you might imagine, the NASA guys had to install some clever fairings
to keep the turbulence from ruining the view and/or ripping the tail
off the aircraft.

Similar considerations apply to the aperture in the dome of a terrestrial
observatory, but the problem is less severe because the air speeds are
less.

Similarly, on my car, when the moon roof is closed, it lies flush with
the top of the car ... but when it is open, a fairing pops up to persuade
the air to jump over the opening. If you reach up and fold down the
fairing, tremendous buffeting is observed.

In retrospect I realize that I knew about SOFIA and about moon roofs
et cetera ... but I was not consciously aware of any of that on election
day. I just thought in terms of stream lines.

This illustrates a point that I've been talking about for a few days:
Most of the memory/recall process is subconscious. When people cite
the literature to "prove" that thought is serial and single-threaded,
it makes me roll on the floor laughing. There's a huge difference
between conscious, attentive thought and the other 99.999% of thought.
Just because you can't consciously attend to it doesn't mean it isn't
happening.

This also illustrates another point: Learning starts from examples,
but it doesn't end there. The examples can be (and should be) used
to help construct a systematic model, i.e. a concept. In this case,
I don't need to remember the 747 example or the moon-roof example,
because I have long since understood those examples in terms of
streamlines and other fluid-dynamics concepts.

Tangential remark: This also tells us that /plagiarism/ is a tricky
thing. It is quite possible to know an idea, yet have not the
slightest recollection of where the idea came from. It may "feel"
completely original, even if it's not. OTOH this is *not* an
excuse for plagiarism. For this reason, before you give yourself
credit for an idea, you absolutely must search the literature.
You may find that the idea is already out there. It may even be
out there in references that you must have read, even if you don't
remember seeing the idea there.


On 06/05/2012 07:17 PM, Kirk Bailey wrote:
there is physics in damn near everything, so making a real
world problem without physics would be much more daunting.

Agreed.

The trick is to strike a balance: Real enough to be real, but not
so real as to be intractable.

The other poll workers would not have been able to figure out the
door trick. Indeed, even after they saw the door trick and saw
that it worked, they had no idea /how/ it worked. It might as well
have been magic. I didn't explain it. We were all too busy for
explanations ... and fluid dynamics is not an easy thing to explain
to physicists, let alone non-physicists.

People spend their lives surrounded by fluids, and they "think" they
understand fluids, but mostly they don't. They grossly underestimate
how complicated fluid dynamics is. Most people know more about
unicorns than they know about fluid dynamics.


On 06/16/2012 08:53 PM, Hugh Haskell wrote:
It really does no good to use questions that the students will not be able to answer.

That's important. Let's talk about that. It's 100% true as stated,
but it's not the whole story.

First of all, I tell my students we are interested in /important/ problems.
In particular, the best thing is to find *easy* ways to solve important
problems.

Let's be clear: I look for problems that are important ... not
problems that are hard. I don't like hard problems. I don't like
doing them, and I don't like assigning them.

The reason we study physics (or anything else) is to learn easy
ways of solving problems that _would have been hard_ if you didn't
know the tricks. So here's the big secret: First learn the
techniques, then use the techniques to solve important problems
that _would have been hard_ hard but are now easy.

There are standard techniques for teaching topics in the "would-have-
been-hard" category. For starters, I'm a big fan of the building
block approach. That is, you take the task apart into elements of
manageable size. You learn each of the elements, and then you put
the elements together in larger and larger chunks, until the whole
task is accomplished.

My criticism of the tests and the textbooks used for high-school
physics is that they leave off the last step. They say "here's
a Lego and here's a Lego and here's another Lego" ... but they
never actually /build/ anything important with the Legos.

No wonder there are motivation and retention problems. If you
give a kid a bushel of Legos but don't allow him to build anything,
he will think Legos are not very interesting.

We all agree that it makes no sense to give students problems they
can't handle and expect them to get up-to-speed all in one step.
For the first N problems, for some very large N, you will need to
take the problem apart into its elements and lead them through it
step by step. I don't want to hear about inquiry-based or
discovery-based learning in the context of real-world problems i.e.
complicated problems. The students will virtually never figure out
the building-block strategy on their own ... much less figure out
what the appropriate building blocks are for any given problem.
Forsooth, usually it takes all the smarts I can muster before *I*
can take the problem apart properly ... and I'm starting from a
much better place than the students are.

That leads to another point: There is an important distinction
between complicated problems and hard problems. There are tricks
for dealing with complexity, so that it doesn't drive you crazy.
A classic example is balancing a chemical equation with, say,
six variables. If you try to do it the way they teach you in
chemistry class, which is basically glorified guessing, you will
never solve it. OTOH if you do it systematically, using Gaussian
elimination, it will require several dozen steps ... but each step
is very very easy. It's as easy as walking, once you know how to
walk. Once you get started, you can walk a very long ways.

At a more advanced level, the building-block idea becomes a topic
for discussion. It becomes a metacognition topic. The idea is
that if they want to do well in real life (and double-especially
if they want to be parents or teachers) they need to be able to
take a would-have-been-hard problem and pull it apart into
pedagogical-sized elements ... without relying on you to do it
for them. This is an important goal, but it is not the starting
point. They will need to see many examples of the building-block
process in action, before they have any hope of understanding the
process as a whole, as a concept unto itself. Again, learning
starts with examples, but it should not end there.