[Cerium]

I have never really looked into it before myself, but this explanation seems to make sense: https://www.grc.nasa.gov/www/k-12/airplane/nozzled.html

Quote: On the other hand, if the converging section is small enough so that the flow chokes in the throat, then a slight increase in area causes the flow to go supersonic. For a supersonic flow (M > 1) the term multiplying velocity change is negative (1 - M^2 < 0). Then an increase in the area (dA > 0) produces an increase in the velocity (dV > 0). This effect is exactly the opposite of what happens subsonically. Why the big difference? Because, to conserve mass in a supersonic (compressible) flow, both the density and the velocity are changing as we change the area. For subsonic (incompressible) flows, the density remains fairly constant, so the increase in area produces only a change in velocity. But in supersonic flows, there are two changes; the velocity and the density. The equation:

- (M^2) * dV / V = dr / r

tells us that for M > 1, the change in density is much greater than the change in velocity. To conserve both mass and momentum in a supersonic flow, the velocity increases and the density decreases as the area is increased.

[Toutouxc]

That's why I was asking for an ELI5 explanation. I know that the equation holds and that compressibility turns everything upside down, I just haven't been able to figure out an intuitive explanation.

I have this idea that in supersonic flow, a pressure wave can't move backwards against the flow, right? Which would mean that any single molecule inside the flow has no way of knowing what's in front of it (because the information simply can't get there), but it feels the pressure of the molecules behind it, so it accelerates towards the void.

The "Fanno flow" article on Wikipedia says that "... For a flow with an upstream Mach number greater than 1.0 in a sufficiently long enough duct, deceleration occurs and the flow can become choked ... Conversely, the Mach number of a supersonic flow will decrease until the flow is choked.", which means that supersonic flow behaves differently in a diverging nozzle than in a simple straight pipe. This is the part that I don't understand. Is the friction inside the nozzle somehow inhibited by the walls of the nozzle gradually moving out of the flow's way or something?

[namibj]

Well, the slow-down won't happen until after it exits the nozzle (c.f. Mach diamonds in rocket exhaust... they are the shocks where the supersonic flow interacts with the ambient air).