3
$\begingroup$

When a wave moves faster than the local speed of sound ($c_s$) in a fluid, there is a shock wave since the fluid is unable to respond to the moving wave. Even if velocity ($v$) is constant, if pressure ($p$) varies, there can be a shock if the local speed of sound is exceeded (i.e. $ v \geq c_s \rightarrow M \geq 1)$. Therefore, to confirm, isn't there always a shock wave present as long as the wave is supersonic? Does simply the type of shock (e.g. normal, oblique) change as alluded to here?

Is there any difference from $M = 1$ shock waves and $M > 1$ shock waves? For example, this article describes a shockless transition from supersonic to subsonic flow, but what does that conceptually mean?

$\endgroup$
2

1 Answer 1

3
+50
$\begingroup$

It's not entirely clear to me what you are asking. You seem to be asking several things at once, and your request for 'a thorough explanation of the influence of mach number on shockwaves, and how a shockless transition from supersonic to subsonic can be achieved' seems very broad. For instance, there are whole text books dealing with supersonic flow and the nature and behavior of shocks.

However, I will try to answer your questions as best I can, to see if I can help clear up any misconceptions:

isn't there always a shock wave present as long as the wave is supersonic?

I assume that here you are referring to a pressure wave travelling through a stationary gas? I believe, if the wave is travelling faster than the upstream speed of sound, then it must be a shock, since no other type of pressure wave can travel faster than sound. Note that it is only supersonic from the point-of-view of the upstream gas, which is flowing into the shock. The shock is moving subsonically, relative to the downstream gas (i.e. the gas that is leaving the shock). This is because the local density increases across the shock, so the local speed of sound downstream from the shock is higher than the shock velocity.

Does simply the type of shock (e.g. normal, oblique) change as alluded to here?

I am a bit unclear as to what exactly you are asking about here - does the type change due to what conditions? You seem to be referring to the inlet cone of a supersonic jet engine. In this case, the goal is to decelerate the flow from supersonic to subsonic (you want subsonic conditions for combustion in most typical engines) with minimal loss of energy. This is done by firstly sending the flow through a series of oblique shocks and then finally through a (weak) normal shock. This is good because oblique shocks tend to be weaker than normal shocks and incur smaller losses. If there was no inlet cone, then the supersonic flow would have to go through a huge normal shock in front of the engine, which would be very lossy and inefficient.

Is there any difference from M=1 shock waves and M>1 shock waves?

I think you may have a misconception here. Shock waves are supersonic flow phenomena, so technically you can't have a shock at M=1. At M=1, you are just on the verge of causing a shock, but you won't actually have a shock until you exceed the sound barrier. Sorry if that seems a bit pedantic, but I think it's an important point: there are no 'M=1' shocks, only M>1 shocks. If the flow is at M=1.00000001, then it's still M>1. (note that here I am referring to the local Mach number of the flow, not the global 'average' Mach number, e.g. for an aircraft)

The key difference to understand is between normal and oblique shocks. Normal shocks are stronger and always go from M>1 to M<1, i.e. they take flow from supersonic to subsonic. Oblique shocks are less strong. They can take flow from M>1 to M<1, but they can also go from M>1 to M>1, i.e. from supersonic to supersonic (but lower M). In fact, in the direction normal to the shock, an oblique shock is also taking flow from supersonic to subsonic, however the flow downstream of the oblique shock is still supersonic, because there is a component of flow velocity tangential to the shock, which is not altered by the shock. This Wikipedia page would be a good place to start, if you want to learn more about oblique shocks.

For example, this article describes a shockless transition from supersonic to subsonic flow, but what does that conceptually mean?

I didn't read the whole article, but it looks like he is referring to slowing down a supersonic flow at the inlet of a supersonic jet engine, as I mentioned above, but in a way that avoids having any shocks, which would be ideal, because then it would avoid those losses. This seems plausible, because if you look at 1D compressible flow theory, flow velocity responds to changes in the flow passage area - supersonic flow should slow down if the flow area is reduced. So, it seems like he may have found some clever geometrical technique that enables the flow to be slowed down to subsonic without having to go through those entropy-generating shocks (which I am sure would be of interest in supersonic engine design).

I hope that helps.

Edit:

For 1D isentropic flow, this is the key equation that relates duct area to velocity:

$$\frac{dV}{dA}=\frac{V}{A(M^2-1)}$$

where $V$ is the velocity, $A$ is the duct area and $M$ is the Mach number. You can see from this that if $M<1$, $\frac{dV}{dA}$ is negative and if $M>1$, $\frac{dV}{dA}$ is positive. Thus, it is possible in theory to decelerate a supersonic flow isentropically by reducing the flow area, without needing any shocks.

If you want to know more about this, then I think reading through a few chapters of a compressible flow text book would help you a lot and would help answer many of your questions.

$\endgroup$
11
  • $\begingroup$ Just to clarify: it could be a pressure wave or any object (e.g. rock) moving at supersonic speeds relative to a stationary gas that could cause a shock wave, correct? Perhaps I am misunderstanding the concept of oblique shocks and what's happening at a molecular level (since entropy is scalar and not a vector). Could you explain the mechanism of how particles can move parallel but not in the normal direction to a shock? Also, displaying the explicit 1-D equations you're referencing and where the area term arises (and how this can avoid shocks/entropy) would solve the problem for me. $\endgroup$
    – Mathews24
    Commented Oct 13, 2018 at 20:50
  • $\begingroup$ @Mathews24 if there is any motion relative to a gas that is at supersonic speeds, then there will almost always be a shock involved somewhere, whether it is the gas moving, or an object moving in the gas. I think the Wikipedia page would be a good place to start to improve your understanding of oblique shocks. I will add the link in my answer. $\endgroup$
    – Time4Tea
    Commented Oct 14, 2018 at 12:35
  • $\begingroup$ @Mathews24 providing a full derivation of 1D compressible flow theory is way beyond the scope of this site. Most textbooks on compressible flow will have an entire chapter that deals with 1D adiabatic flow, as the maths is fairly involved. What I will do is quote the one equation that relates Mach number to the area of a duct - for the rest, I recommend you consult a text book. $\endgroup$
    – Time4Tea
    Commented Oct 14, 2018 at 12:38
  • 1
    $\begingroup$ @honeste_vivere I have added a note in my answer to clarify that I was referring to the local Mach number there, not global. A shock can only form if the local Mach number of the flow exceeds M=1. But yes, I agree with your point that you can have local regions of supersonic flow, even if a larger body such as a piston or aircraft is moving globally at a speed less than Mach 1. $\endgroup$
    – Time4Tea
    Commented Oct 16, 2018 at 20:47
  • 1
    $\begingroup$ @Mathews24 I think a clearer way to state it would be that, across a shock, the flow in the direction normal to the shock transitions from M>1 to M<1. So, at some point within the shock, it passes through M=1 as it decelerates. $\endgroup$
    – Time4Tea
    Commented Oct 17, 2018 at 21:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.