A new way to explain airflow

Elastic collision simulations used to model the kinetic theory of gas. I introduce the concept of a speed signal to explain the exchange between pressure and velocity in airflow. It is the underlying mechanism that leads to Bernoulli's equation, but it is not limited to flow with simplifying assumptions such as the flow being steady, incompressible or inviscid. I show the effect of temperature and density on the Maxwell-Boltzmann distribution. I have included the effect of duct geometry on choked flow and supersonic expansion, but an explanation of flow through a converging duct or venturi will be covered in a future video.

[AlphaPhoenixChannel]

at least three times in here i kinda gasped as i connected something new. - this is a fantastic way to think about these dynamics. Reasoning out why venting compressed gasses get cold is normally something that requires me to do some mental gymnastics but this signal concept makes it completely trivial. I love it.
I'd love to see a 2D (or 1D) circle filled with particles that get released to vacuum, but instead of coloring them by speed color coding them based on their starting radius. I had always imagined that the fast moving particles squeeze through to escape, but when the MFP is low they can't - it's the energy that bounces through.

[superfao]

This is the first time in 20 years that I understand why an increasing cross section in supersonic flow results in increasing speed. Awesome explanation!

[domonkosludvig]

Thank you. I always could imagine why things worked the way they did, but it hurt to watch wrong videos on the topic, and people accepting those "its magic" kind of answers.

[bradgirod]

If i ever teach fluid dynamics, i will use these videos on the sections where we learn about choked flow. These are simply wonderful.

[jellyfish]

Fantastic video, as an engineer, it is a very intuitive explanation. I would suggest a few things: 1) Consider the effect of compression when uploading the video, maybe there are ways to improve clarity. 2) Go slower and give more time for visualizing the animations, a lot is going on and you are moving fairly quick. 3) It would be great if you could connect Pressure, Temeprature, and Density (maybe other variables as well) to properties of this analogy (Collosion forces, average velocity, moleules per unit area). Thanks and loving it!

[WetDoggo]

I'd prefer to call it an impulse instead of "sign".
Excellent representation, that's basically how I always imagined how gases and liquids behave on this scale.
Really cool to see the emergent properties arise

[llahneb]

Spectacular demonstration of an otherwise complex phenomenon into something that is intuitively within reach. You’ve got a great skill, I hope you keep these up.

[adrianinvents]

Thanks for excellently demonstrating this wonderful way of thinking about fluid dynamics. Far more intuitive than most explanations.

[tinyear]

Something I keep circling back to after watching this:

The entire system you model here is exquisitely sensitive to mean free path and boundary feedback.

Without short-range coupling (via collisions), there is no sustained signal transfer, no stable flow, and no coherent pressure differential.

Which makes me wonder — how often do we over-simplify large-scale models (like planetary atmospheres or space propulsion) by forgetting the absolute dependence of these effects on the medium conditions?

It's one thing to describe flow behavior inside an atmosphere or pressure vessel...

It's another thing entirely to assume that behavior persists unchanged into regions where the mean free path dwarfs the system itself.

Your work makes that razor-edge distinction beautifully clear.

[MigotRen]

This was a nice watch. I was never satisfied with the the Bernoulli explanation of "well it has lower pressure cause it moves faster". I always viewed it more from the approach of kinetics. Nice to have such a clear visual reference.

[jojodi]

Fantastic follow up video thank you! What's awesome is that all of this is very intuitive. If we look at the statistics of a small region of space and consider many interacting patches we can see how things like lattice Boltzmann methods work. Thank you again!

[KurtCollier]

you know you just watched a next-level explainer when one of your favorite science youtube guys sends you to a video, you watch it, and change how you think about a thing. then you look down in the (less than 100 comments at this point) and see one of your other favorite youtube science guys telling you that this video is reallly helpful and good. I need an extra thumbs up button for you. Since i don't have one of those. i'm just gonna watch again because I am confident I messed something simple.

[Petch]

This one is just for the algorithm, such that more people can see this fantastic video. 🥰

[fusha_]

It's fascinating how simple equations of thermodynamics can describe the combined behavior of billions of molecules.

[poobertop]

Very insightful, thanks Will. Despite all I've watched on rocket nozzle design, I never once consider the backwards travelling molecules, so thanks for clearing that up for me.

[tinyear]

"This is excellent — what you've modeled here bridges the gap between kinetic theory and intuitive flow dynamics beautifully.

What strikes me is how the speed signal concept dissolves the artificial boundary between 'static' pressure and 'dynamic' flow — showing they are both aspects of energy distribution mediated by molecular collisions within a boundary.

It really reinforces that flow is not the result of some abstract 'force' acting at a distance — but a local consequence of asymmetry in energy density within a confined space.

It makes me wonder if future propulsion systems might lean less on the classical 'reaction mass' model, and more toward managing localized field asymmetries — effectively shaping displacement through internal energy structuring rather than relying solely on expelling mass.

In any case — this is the kind of content that helps people move from textbook memorization to true cause-and-effect understanding.

Looking forward to your future Venturi exploration!"

[mitchellsteindler]

Incredible video. My understanding of the pressure decreasing with velocity has recentky evolved to "well before the molecules were moving around randomly. When the flow through a smaller cross section, now they are are moving mostly in the direction of flow. So they spend less time bouncing off walls in the reduced cross section pipe so pressure is lower". I think this is a great visual explanation. It would have been cool to see a count of collisions per inch of boundary in the container vs in thr reduced cross section pipe. Multiply each collision by their velocity component normal to the boundary and you calculate a sort of pressure.

[jj]

You made it so much more intuitive and interesting than how I have heard of stuff like this before.

[DerTypDa]

I particularly enjoyed the way this model explains a gas in a pressurized container cooling down as the container is opened up. The highest-speed signals are the first to move out, leaving only low-speed molecules behind, leading to a drop in temperature. That feels a lot more intuitive to me than the energy density model.

[youngbloodbear]

Great way to connect the mechanism behind waves!