Latest / Mechanical Engineering Made Simple / (#23) How Engineers Manipulate Flow, Micro fins, and EHD to Maximize Condensation Heat Transfer

Transcript

(#23) How Engineers Manipulate Flow, Micro fins, and EHD to Maximize Condensation Heat Transfer · Mechanical Engineering Made Simple

• 0:00
  Welcome back, design jackies to Mechanical Engineering Made
• 0:02
  Simple. If you're looking for the
• 0:04
  ultimate shortcut into the thermal design of heat
• 0:06
  exchangers, well, this deep dive is definitely for you.
• 0:09
  We are pulling back the curtain today on the core mechanisms of
• 0:12
  heat transfer optimization. Specifically, we're focusing on
• 0:16
  how engineers manipulate fluid flow, geometry, and surface
• 0:19
  forces to really maximize efficiency during condensation.
• 0:23
  And that's, you know, a nonnegotiable process and
• 0:25
  everything from your air conditioning unit right up to
• 0:27
  utility scale power cycles. Right.
• 0:29
  And our mission today really is to uncover the specific physical
• 0:33
  levers that high performance engineers are actually pulling
• 0:36
  to boost thermal performance. We've synthesized quite a stack
• 0:39
  of sources covering the whole gamut, the fundamentals of
• 0:42
  condensation physics, detailed flow regimes inside tubes, you
• 0:46
  know, advanced passive techniques like micro fins at
• 0:49
  cutting edge, active enhancement using something called Electro
• 0:52
  hydrodynamics or EHD, and even the the crucial role of surface
• 0:56
  material science itself. OK, let's unpack this then.
• 0:59
  Maybe we should start right inside the tube where the
• 1:01
  condensation actually begins. When I talk about two phase flow
• 1:03
  and horizontal tubes, I understand there's this constant
• 1:06
  kind of fundamental conflict playing out.
• 1:08
  It's a battle between gravity pulling the liquid down and the
• 1:12
  vapor shear trying to strip it away.
• 1:14
  How does that conflict? How does it define how well the
• 1:17
  heat exchanger actually works? Well, it pretty much defines
• 1:20
  everything because the distribution of that liquid
• 1:24
  condensate, that's your single biggest enemy for heat transfer,
• 1:27
  what we call 2 phase flow, it gets dramatically reorganized
• 1:31
  just by changing the speed of the vapor core.
• 1:34
  So think about it, when the vapor velocity is low, gravity
• 1:37
  just winds easily. The condensate pools at the
• 1:40
  bottom of the tube and that creates a thick insulating
• 1:43
  layer. We call that stratified flow.
• 1:46
  Is exactly what you don't want. It's the enemy of heat transfer.
• 1:49
  Absolutely. It's a barrier, but OK.
• 1:51
  Once you start increasing that vapor velocity, the shear
• 1:55
  forces, the drag essentially exerted by the fast moving
• 1:58
  vapor, start to overcome gravity, at least locally.
• 2:01
  This pushes the liquid around the tube wall and you see this
• 2:04
  transition sequentially. It goes into WAVY flow, then
• 2:08
  WAVY annular flow, and eventually you hit the prized
• 2:12
  regime annular flow. Right, annular flow, that's the
• 2:15
  goal because the liquid coats the entire wall.
• 2:17
  But as a thin film, right? Minimizing that insulating
• 2:20
  layer? Precisely.
• 2:21
  A thin, evenly distributed film is what you want, and the flow
• 2:25
  regime directly determines the efficiency.
• 2:27
  Our sources highlight something quite interesting here, your
• 2:30
  heat transfer coefficient dollars.
• 2:32
  It actually becomes pretty insensitive to the flow speed or
• 2:34
  mass flux when the flux is low. Let's say around $75 text
• 2:38
  kilograms per second per meter square.
• 2:40
  Insensitive. Why is that?
• 2:41
  Well, it's because gravity is just dominating how thick that
• 2:44
  film is at the bottom. Almost regardless of how much
• 2:46
  vapor is actually flowing through the core.
• 2:49
  The geometry of the pool is set by gravity.
• 2:51
  OK. That makes sense.
• 2:52
  But once we cross some threshold, maybe get that vapor
• 2:54
  moving faster, and we're driving this fast annular flow, then the
• 2:58
  heat transfer suddenly cares a lot about speed.
• 3:01
  It becomes highly dependent on mass flux and quality.
• 3:05
  Yes, exactly. For example, the sources point
• 3:08
  to mass flux is like $300 text kilograms per second per meter
• 3:11
  squared and above. At that point you're firmly
• 3:13
  enforced convection territory. And here the heat transfer
• 3:16
  coefficient different dollars is shown to be proportional to the
• 3:19
  square root of the pressure drop gradient.
• 3:21
  That's the square root of minus delta P divided by delta Z or
• 3:25
  delta P delta Z. OK, O.
• 3:29
  That mathematical relationship tells us something immediately.
• 3:31
  Imortant doesn't it? The heat transfer gain is
• 3:34
  directly. Proportional to the.
• 3:35
  Resistance and friction. We're generating it sounds like
• 3:37
  the harder the flow has to push, the more effectively the vapor
• 3:40
  shears that liquid film, thinning it out, and the better
• 3:43
  the performance gets. That's exactly the mechanism.
• 3:46
  The friction is the enhancement in a way.
• 3:48
  But you know, to really appreciate how much enhancement
• 3:51
  we're talking about with these methods, we need a baseline.
• 3:54
  What happens if we just let gravity do its thing?
• 3:57
  And that's where the classic Nussel analysis for gravity
• 4:00
  driven film condensation comes in.
• 4:02
  It sets that sort of foundational limit.
• 4:04
  Right, the Nusselt equation. It looks pretty complex when you
• 4:06
  read it out at one AG .7299 times the 4th root of a whole
• 4:11
  bunch of terms. Can you maybe unpack what that
• 4:13
  famous equation tells the listener conceptually,
• 4:15
  especially about the levers we can't easily change?
• 4:18
  Sure, let's breakdown. It's 10.7299 times the 4th root
• 4:21
  of the quantity rho sub L * G times rho sub L minus rho sub G
• 4:25
  * K sub L cubed times Lambda, all divided by MU sub L * T sat
• 4:30
  minus T sub W times D sub O Phew.
• 4:34
  OK, what does that mean? It's all about the natural
• 4:36
  physics of the situation. That equation shows that the
• 4:38
  theoretical limit of gravity driven performance is entirely
• 4:40
  set by physical properties and conditions.
• 4:43
  Things like liquid density Musev L, gravity G, liquid dermal
• 4:45
  conductivity, case of L, the latent heat of vaporization
• 4:48
  Lambda and the liquid discosity Musev L, plus of course the
• 4:52
  temperature difference between saturation and the wall tsat
• 4:55
  minus T sub W and the tube diameter D sub O.
• 4:58
  These are mostly inherent properties of the fluid and the
• 5:00
  setup. So the point is, if we want to
• 5:02
  exceed this inherent baseline performance dictated by gravity,
• 5:05
  we'd have to stop relying solely on gravity.
• 5:07
  We need to actively interfere with the physics.
• 5:09
  OK, so interfering. If we can't always just crank up
• 5:13
  the vapor speed to force that efficient annular flow, maybe
• 5:17
  due to pressure drop constraints or something, we need to look
• 5:19
  for other physics we can interfere with, perhaps
• 5:21
  passively. And that brings us to this kind
• 5:24
  of counterintuitive world of tiny fins.
• 5:28
  We're talking really small fins here, less than 1.5 text
• 5:31
  millimeters high. Why are we deliberately keeping
• 5:34
  the fins low? Isn't the.
• 5:36
  Standard heat transfer rule. Just add more surface area.
• 5:39
  Bigger fins better. That's the classic thinking.
• 5:41
  More area is always better. But for condensation, the goal
• 5:44
  with these low fins is radically different.
• 5:47
  It's quite clever actually. We aren't just adding area.
• 5:49
  I mean, there is some area addition, but that's not the
• 5:51
  main point. We are specifically engineering
• 5:53
  microscopic pressure gradients and we're using surface tension
• 5:57
  drainage forces to actively thin that insulating liquid film.
• 6:01
  Surface tension drainage forces. OK, so it sounds like we're
• 6:03
  turning the liquid into a sort of self squeegeeing system.
• 6:06
  How exactly does surface tension achieve this drainage?
• 6:09
  That seems subtle. It boils down to curvature.
• 6:11
  The physics is described by Laplace's equation.
• 6:14
  Essentially, any curved liquid vapor interface creates a
• 6:17
  pressure difference across that interface.
• 6:19
  The equation states that the pressure difference sub L - π
• 6:22
  dollar G is balanced by the surface tension Sigma and the
• 6:26
  radii of curvature of the interface.
• 6:27
  Let's call them core one in order to so Paulo PG O ES Sigma
• 6:32
  1R1 plus one R at 2:00. OK, so the pressure difference
• 6:35
  changes depending on how sharply curved the interface is.
• 6:37
  Precisely. The smaller the radius of
• 6:39
  curvature, meaning the sharper the curve, the greater the
• 6:42
  pressure difference generated. So engineers very carefully
• 6:45
  designed the fin geometry, the shape of the fin tip and flank,
• 6:49
  so that the liquid interface curvature decreases as the
• 6:51
  liquid flows along the fins path from the tip down towards the
• 6:55
  base or through. This intentional decrease in
• 6:57
  curvature generates a pressure gradient.
• 6:59
  It literally sucks the condensate from the top of the
• 7:02
  fin down into the troughs between the fins.
• 7:04
  This action keeps the primary heat transfer surfaces, the fin
• 7:07
  tips and upper flanks covered by a much thinner, less insulating
• 7:10
  liquid film. I see this is where geometry
• 7:13
  optimization becomes a real art form.
• 7:15
  Then our sources mentioned specific profiles, like the
• 7:19
  atomic profiles from back in 1981.
• 7:22
  These are sort of convex shapes designed specifically to
• 7:24
  maximize this surface tension pole, right?
• 7:26
  Exactly. Those profiles were early
• 7:28
  examples of really trying to tailor the shape for drainage.
• 7:31
  And the results? Well, they prove the concept
• 7:33
  works beautifully. If you compare, say, a standard
• 7:36
  trapezoidal integral fin tube to a highly optimized sawtooth fin
• 7:41
  geometry, the sources show sawtooth tubes can achieve
• 7:44
  approximately twice the condensation heat transfer
• 7:47
  performance. Wow, twice the performance and
• 7:50
  that's mainly from better drainage, not just the area.
• 7:52
  That's the key insight, yes. The performance increase comes
• 7:55
  primarily from these superior surface tension drainage
• 7:57
  mechanisms, much more so than just the modest increase in
• 8:00
  surface area. OK.
• 8:01
  And to make sure that this microscopic squeegee effect is
• 8:04
  actually working and the gravity isn't just overwhelming it,
• 8:06
  especially between the fins, engineers use something called
• 8:09
  the bond number or BDDDF. Right.
• 8:13
  The bond number is our critical quality check for this strategy.
• 8:16
  It's defined as the ratio of gravity forces to surface
• 8:19
  tension versus for our surface tension drainage strategy.
• 8:22
  To really dominate and be effective, the bond number must
• 8:25
  be small. This metric basically tells you
• 8:28
  how tall the fin can be. If the fin height gets too
• 8:31
  large, the distance the liquid needs to be pulled becomes too
• 8:34
  great. Then the surface tension
• 8:36
  pressure gradient kind of dissipates before it can
• 8:38
  effectively drain the film all the way down, and gravity starts
• 8:41
  to win the battle again, pooling liquid between the fins.
• 8:44
  Fascinating. So we've covered the forced
• 8:46
  approach, high velocity annular flow and this elegant passive
• 8:50
  approach using geometry driven surface tension.
• 8:54
  Now let's move to what sounds like the most aggressive method
• 8:56
  of all active enhancement using Electro hydrodynamics, or EHD.
• 9:01
  Here, we're ditching the finesse and actually starting to apply
• 9:03
  external energy, right? Yeah, EHD is fascinating and
• 9:06
  it's definitely more energetic. It uses a high voltage but
• 9:10
  typically low current electric field to actively manipulate the
• 9:14
  condensate film. Maybe think of surface tension
• 9:18
  drainage as gently coaxing the liquid away, like using that
• 9:21
  microscopic squeegee we talked about.
• 9:23
  EHD on the other hand, it's more like literally zapping or
• 9:27
  pulling the condensate off the surface using electrostatic
• 9:30
  forces generated by the high voltage.
• 9:32
  It can be a bit of a sledgehammer approach, but it
• 9:34
  can be incredibly effective. A.
• 9:35
  Sledgehammer, okay. And it requires specific
• 9:38
  conditions, doesn't it? It doesn't work for everything.
• 9:40
  That's correct. It really needs fluids that have
• 9:42
  low electrical conductivity. A dielectrics essentially.
• 9:45
  That's why it's most often studied and applied with
• 9:48
  refrigerants, which often fit that bill.
• 9:50
  The impact, as you hinted, can be massive.
• 9:52
  The reported enhancements for FEM condensation on a flat
• 9:55
  plate, for instance, can be as large as an incredible $2000 tax
• 9:59
  percent. 1000% That's extraordinary.
• 10:01
  It is. And even for smooth in tube
• 10:04
  condensation, which is often harder to enhance, the
• 10:07
  enhancement can reach up to about $6 and 555 times the
• 10:10
  baseline performance without the electric field huge gains.
• 10:13
  But presumably getting that kind of game isn't just a simple
• 10:16
  matter of turning up the voltage dial.
• 10:18
  The sources mentioned a correlation for a refrigerant
• 10:20
  like R-404-A and it looks complicated.
• 10:22
  Oh, it's definitely not simple. It's a really complex
• 10:25
  multivariable correlation. They developed the enhancement
• 10:29
  ratio one ITO it looks something like $1.00 plus C * N two to the
• 10:33
  power, $1.00 to the power, $1.00 to the power time of 2 * R two
• 10:37
  NS to the power N where two or five.
• 10:38
  So what does that tell us? The conceptual take away is that
• 10:40
  the enhancement that boost you get depends on really
• 10:44
  complicated interactions. It depends on the electric field
• 10:47
  parameter related to voltage and geometry, the total mass flux,
• 10:51
  how much vapor is left which relates to the quality and the
• 10:53
  up top and even the Jacob number which connects latent to
• 10:56
  sensible heat effects. You're essentially fighting, or
• 10:59
  rather manipulating flow dynamics, phase change physics,
• 11:02
  and the electric field strength all at the same time.
• 11:05
  It's a big optimization puzzle. Right.
• 11:08
  A complex interplay. OK, so to kind of round out our
• 11:11
  view on optimizing these surfaces, we should probably
• 11:14
  acknowledge that all these sophisticated methods, flow
• 11:16
  control, micro fins, EHD, they all sit atop A foundational
• 11:21
  material science layer. We often assume in our models
• 11:25
  that surfaces are perfectly ideal, maybe optically smooth.
• 11:29
  But reality is different, isn't it?
• 11:30
  Absolutely. Real world surfaces are almost
• 11:33
  always altered by things like oxidation, maybe some roughness
• 11:36
  for manufacturing or use, and those factors can matter quite a
• 11:39
  bit even if you think you're primarily focused on convection.
• 11:43
  Take roughness, for example. A surface is only truly
• 11:46
  optically smooth if the scale of its roughness is much, much less
• 11:49
  than the wavelength of the electromagnetic radiation
• 11:52
  interacting with it. As roughness increases,
• 11:54
  reflection becomes less secular, less mirrorlike, and much more
• 11:58
  diffuse or scattered, and that dramatically changes how the
• 12:01
  surface exchanges heat radiatively.
• 12:03
  And you also mentioned the subtle role of surface layers,
• 12:05
  esecially things like outside films.
• 12:07
  They can potentially sabotage performance even if they're very
• 12:10
  thin. Yes, that's a critical point
• 12:13
  often overlooked. A very thin non metallic layer
• 12:16
  like an oxide film that naturally forms on many metals
• 12:18
  exposed to air or certain fluids can severely decrease the
• 12:22
  overall reflectance of the metal substrate underneath.
• 12:25
  And if reflectance goes down, absorptance goes up.
• 12:28
  Since many heat exchanger designs implicitly rely on the
• 12:31
  metals high reflectance to minimize unwanted radiative heat
• 12:34
  loss or gain, adding even a thin non metallic layer can
• 12:38
  unexpectedly shift the whole heat exchange balance in a way
• 12:40
  you didn't intend. It really emphasizes that
• 12:42
  optimizing A condenser surface requires A holistic view.
• 12:46
  You have to consider all the thermal modes together,
• 12:48
  conduction, convection and radiation, and how the surface
• 12:51
  material influences all of them. This has been a really
• 12:54
  comprehensive look at the physics engineers manipulate for
• 12:56
  condenser optimization. Yeah.
• 12:58
  Let's just quickly recap the four main engineering levers
• 13:00
  we've identified for you, the listener today.
• 13:03
  OK. 1st, we talked about controlling the flow regime,
• 13:07
  really understanding and managing that critical balance
• 13:09
  between gravity and vapor shear, ideally to force high efficiency
• 13:13
  annular flow when possible. Second, there's passive
• 13:16
  enhancement through geometry, using those clever microscopic
• 13:20
  fins to harness surface tension drainage driven by carefully
• 13:23
  engineered curvature. Third, we looked at active
• 13:26
  enhancement, bringing in the big guns with high voltage EHD
• 13:30
  fields to aggressively remove those insulating condensate
• 13:33
  films, leading to potentially tremendous performance boosts.
• 13:36
  And 4th, remembering the foundation fundamental material
• 13:39
  surface control. That means meticulously managing
• 13:41
  things like surface roughness and those thin oxide layers to
• 13:44
  ensure you have the optimal thermal and radiative properties
• 13:47
  for the job. Right.
• 13:48
  And ultimately, it seems the most efficient design you're
• 13:50
  likely to encounter in advanced applications is robably some
• 13:53
  kind of hybrid, isn't it? Maybe utilizing that assive
• 13:55
  geometry and the microscale surface hysics from the fins,
• 13:59
  ossibly augmented by active fields like EHD where
• 14:02
  appropriate. But I think the take away from
• 14:04
  the researchers who classify these flow regimes is really
• 14:07
  crucial. Different physical forces tend
• 14:10
  to dominate different regimes and transitions.
• 14:12
  That's the real challenge, and maybe the final thought.
• 14:15
  Since these competing forces, gravity, shear, surface tension,
• 14:19
  maybe even electric fields, control different transitions,
• 14:21
  like the shift from stratified to annular flow, or how liquid
• 14:25
  drains, we should probably never expect a single simple formula
• 14:29
  or parameter to perfectly predict all thermal behavior in
• 14:32
  every conceivable condensing environment.
• 14:35
  It's a constant reminder really, that high performance
• 14:37
  engineering, especially in thermal fluids, is often about
• 14:40
  understanding and managing complex, sometimes conflicting,
• 14:43
  physical phenomena. Until the next equation designed
• 14:46
  junkies, we are mechanical engineering made simple.