(upbeat music) - When you Google the word convection, you get images like this one,
a 2D diagram of what happens in a sauce pan when you
put it on the stove, and it seems to fit with experience. Like sometimes I lift the
lid on a pan of spaghetti and the spaghetti is
gathered together into a kind of donut shape, and I
can imagine the strands spinning around to form that shape, but is it actually what's going on? Well, it turns out the truth
is a lot more interesting, and I found that out by
building an actual 2D version. You know, I like building 2D versions of hydrodynamic mechanisms. If you're into that sort of thing, consider hitting subscribe and clicking the notification bell. My first thought was
that it should be made of a single piece of glass, but it needs to be
resistant to thermal shock. So borosilicate, but
that's hard to work with and it would need to be
fashioned from sheets of glass, not blown glass, which can end up cloudy. You can see that here
in the first iteration of my glass pop pop boat
that had a flat boiler. So in the end we went with polycarbonate, which you don't wanna get too
hot, but it should be okay. My first thought was that I should heat it from the middle only. That way there'll be a very
clear route for the liquid to take in its little
cycle around the container. So only this small
section here is aluminum, like if the whole bottom part was aluminum and I heat it from the middle, eventually that heat would spread out because aluminum is a
good conductor of heat. I made the container quite tall because I wanted to see what
kind of movement you got for different heights of
liquid in the container. There's also the question
of how do you visualize how the water is moving? And I had a few ideas for that. The first was to make
the liquid rheoscopic, and that's what you're seeing here. That's really easy, you just
add mica powder to water. The tiny flakes of mica tend to line up with each other locally, and
as the water around them twists and changes direction, these lined-up flakes reflect
light in different ways according to the
orientation of the flakes. And so that's how the
flow becomes visible. I haven't yet found an explanation for why the mica flakes align
with each other locally, which is annoying, and here it is. It takes a little time to get going, but I love the way it looks like the smoke from a candle flame. If I speed it up, the resemblance
is even more striking. There's some interesting detail here that isn't really conveyed
in the traditional diagrams that you get shown at school. You get fast, turbulent
flow in a narrow channel in the center and a slow descent in wide channels either side. That might be down to the fact that I'm heating it from the middle only. So here's a new container where the whole of the bottom is an aluminum strip. The flow still seems really chaotic. Sometimes it seems as though you can see different convection cycles,
but then they break up. At one point, the whole container
is just one giant cycle, but again, it doesn't last very long. Rheoscopic liquid gives
you a sense of which parts of the liquid are in motion, but it doesn't necessarily
give you a great sense of like the velocity of the liquid. It does to a degree, but I wanted to see if I could do better. My first thought was to use some kind of neutrally buoyant particle. In other words, find some
material that has the same density as water, which is one
gram per cubic centimeter. That was the original
definition of a gram, by the way, one cubic centimeter of water. So a lot of plastics have a density close to one gram per cubic centimeter,
but no plastic nails it. And then I thought, well, hold on. I only need to find a plastic
that has a density somewhere between the density of water and the density of salt saturated water, which is about 1.2 grams
per cubic centimeter. ABS fits the bill nicely,
and just by trial and error, I found the right amount of
salt to get those ABS particles pretty much neutrally buoyant. The problem is they stuck
together and that's annoying, and they seem to stick to the sides a bit, so it doesn't really help to visualize the convection currents. You can buy special particles
that are a precise mixture of different plastics
that give you 1.00 grams per cubic centimeter of density. The problem with those is
they're really expensive. So I tried something else, water beads. Water beads are made
of sodium polyacrylate, which has a density of about
1.2, which isn't great, except because they
absorb hundreds of times their own weight in water, once they're saturated in this way, their density is extremely
close to one, and importantly, they don't stick to each other or the sides of the container. The question then is would
the convection current created in this 2D convection current
viewer be strong enough to lift the beads off the bottom? And it turns out they are. That's quite nice, isn't it? But I really wanted to see
a good flow of particles. So you know what? I bought the expensive beads, right? I bought them. It's about a third the
cost of gold by the gram, which might be the most
expensive substance I've ever bought. Actually, maybe not. Actually, what's nice about these beads is they're fluorescent,
so they really stand out if I shine this UV lamp on them. And I'm sure this is
something you already know, but this convection flow we
are seeing here is driven by changes in buoyancy. When the water at the bottom is heated, it becomes less dense,
it becomes more buoyant, and so it rises to the top. At the top when it's further
away from the heat source and closer to the cool air,
it cools back down again. It condenses, becomes less
buoyant, and sinks back down. When it reaches the
bottom, it heats up again and the cycle continues. And because when it reaches the top, it doesn't have anywhere to
go except left and right, you end up with these two cycles. Another way to view the convection current actually without using a marker at all, is by looking at something
akin to heat haze. As the water heats up,
it becomes less dense, bends the light differently, and that distorts this
checkerboard that I put behind it. There's a convection
phenomenon that I really wanted to try and reproduce. It happens on the surface
of the sun, for example. You get these convection cells,
so if you have a thin layer of convecting fluid, it breaks
up into these smaller units. I should be able to get that
by just making the water in my convection viewer shallow and heating the whole of the bottom. I think it's possible to discern a few different convection cells here, but they're not fully
independent of the whole body of liquid, like there seems
to be a general movement outwards across the top and
inwards across the bottom. And when I do it again
with rheoscopic fluid, it seems really chaotic actually. Even just in a Petri dish, it
doesn't seem to work either, but I was aware of this working
with oil instead of water. So look, here's that mica
in just cooking oil now, and look at those lovely convection cells. I'm just heating the
base of the Petri dish with hot water from the kettle. That's amazing, isn't it? Quite fast spinning
convection currents there. And then you've got this slow movement as the cells change shape
and merge and move around. This is called Rayleigh-Bénard
convection, by the way, when you have these stable cells. Apparently if you get the
conditions just right, those convection cells
will spontaneously form into irregular hexagonal lattice. I thought the trick would
be to make the layer of oil really thin, but I can
never get hexagons myself. So with oil, it seems
that you can get those multiple convection cells
set up when you have just a shallow layer of convecting fluid. So let's try the oil in the 2D container. And there you go. It's cool, isn't it? Look at those individual convecting cells and there's hardly any
turbulent flow at all. And for completeness, here
is a full container of oil heated evenly from the bottom, and you can see that the
flow is a lot less turbulent. That's because oil has a higher viscosity, which means for the same
flow rate, less turbulence. And it seems like the convection
cells are a bit more stable with oil, though it is
still quite chaotic. It's not just on the sun that you get these
convection cells, of course. Now an important part of the
weather system here on earth, you get Hadley cells that
take warm air from the equator and deposit it further north and additional convection
cells at higher latitudes. We typically think of
convection being driven by differences in temperature,
but it's not the only way. If you take a dark colored
liqueur like Tia Maria and you pour cream very carefully on top, you also get convection, but
this is solutal convection. It's driven by changes in concentration. In this case, the Tia Maria
that's close to the surface loses alcohol by evaporation,
becomes more dense and then falls down to
the bottom to be replaced by Tia Maria from lower down
that has more alcohol in it. That alcohol evaporates
and the cycle continues. Again, this is driven
by changes in buoyancy, but buoyancy is changing
for a different reason. It's not obvious why
the cream is important, but it does seem to be. Convection can also be driven
by changes in surface tension and the cream might be influencing
the system in that way, but I don't believe anyone knows for sure. I'll link to one useful
paper that I found though. Something really weird happens when I search through my
comments for the word Henson. Basically lots and lots
of people banging on about how good the Henson AL13 safety razor is. I mean, it's not that weird
when you consider the fact that all of those comments are on a video that was sponsored by Henson Shaving, but it is really unusual to get just that much positive
feedback about a sponsor. And actually it made me realize that I made a mistake in
the last sponsorship read. Like I thought I was being
really clever talking about the dubious business model
of cartridge razor brands. It's literally called the
razor and blades model, and it's a massive false economy, but that's not what the
commenters care about. They just seem to really
enjoy the experience of using the Henson AL13. So let's look at why that might be. Well, it comes down to
two things basically, precision and blade support. The solid aluminum body
gives incredible support to the blade, which you just
don't get on cartridge razors, almost by design actually. But with all the support in the world, if the blade is in the wrong position, you're not gonna have a good shave. Well, Henson is a family
run aerospace machine shop that pivoted to making their own products. So these razors are
made using CNC machines to aerospace standards. Like once the blade is installed, it protrudes just 33 microns
from the shave plane. There's not much room for error there. I mean, this is all good
information to know, but when it comes down to it, like, I just wouldn't
shave with anything else now that I've tried the Henson
AL13 and neither would any of these people in the comment section. And just for comparison, I
shaved this side of my face with the leading cartridge brand and you can really see the difference. And I will just make this point again, like you could buy a cheap razor handle and get stiffed on the cartridges
for the rest of your life, or you could pay a reasonable price for this precision
engineered all-metal handle that's gonna last a lifetime and then get the blades
for literal pennies. If you're interested
the promo on this one, it's really good. If you go to hensonshaving.com/stevemould and use promo code
Steve Mould at checkout, you'll get a hundred free blades with your Henson AL13 safety razor. Just make sure both of those
items are in the basket when you apply the code. That's equivalent to three
or four years of shaving. The link is also in the description, so check out Henson Shaving today. I hope you enjoyed this video. If you did, don't forget to hit subscribe, and the algorithm thinks
you'll enjoy this video next. (upbeat music) (upbeat music continues)
