Can an IC Engine's Exhaust Flow Reach Supersonic Speeds?

[vinay ks]

i wanted to know if the flow in the exhaust pipe of an ic engine cross the sonic speed? i have been seeing a lot of papers presented about the supersonic internal flows on the net and hence i wanted to know some of the practical situations where such flows can be seen?

 

[boneh3ad]

I'm not sure but my gut feeling is no. I don't know all the pressures involved throughout the exhaust system of a typical internal combustion engine, though, so I'm open to the possibility.

I'd you are just looking for examples, how about a scramjet or a shock tube? The flow in the barrel of a gun is often supersonic. The flow in a rocket engine is supersonic before reaching the exit of the nozzle.

 

[AlephZero]

You only need human blowing power to create shock waves, so it doesn't seem unreasonable that they might exist in an IC engine exhaust. It probably depends whether the exhaust was designed to sound quiet or noisy.

http://www.bbc.co.uk/news/science-environment-13574197
http://www.researchgate.net/publication/253351132_Visualization_of_weak_shock_waves_emitted_from_a_trombone

 

[boneh3ad]

I need to go look at that article when it get on campus again tomorrow and can access it. That seems very suspect to me given that, with the atmosphere as the back pressure, a human would have to generate a total pressure of 1.89 atm to produce a supersonic flow. That seems like a tall order. I'm very interested in seeing what the paper says though. If it's true, that's pretty neat.

 

[AlephZero]

That seems very suspect to me given that, with the atmosphere as the back pressure, a human would have to generate a total pressure of 1.89 atm to produce a supersonic flow. That seems like a tall order.

A trombone is a resonant system, and the blowing is pulsed at the resonant frequency. You aren't blowing against a steady back pressure.

If would be nice if somebody did a similar study of organ pipes. You can get some very loud noises with blowing pressures that are normally below 20 inches water gauge, and often below 10.

Even the most extreme pipe organs, built more to prove how rich the sponsor is than to make music, only use of the order of 100 in w.g maximum.

 

[boneh3ad]

I looked at the 1995 article linked in the BBC report (the one from your researchgate link apparently either isn't available to me through my campus library's online portal or else is not available in full text online). It looks like the pressure rise across the wave is 2 kPa to 4 kPa depending on the two data sets they reported. If it is a shock, that would imply a Mach number of about 1.04, and that is the same order of magnitude as what the BBC article reports as being the wave speed from the article I can't find. Unfortunately the researchers didn't do anything like placing two microphones close together and measuring the actual wave propagation speed, as that would definitively prove it. Otherwise, for such a weak potential shock wave, it could seemingly just be a sharp pressure rise and not an actual shock.

I am not an acoustics guy, so I don't know exactly his theoretical basis and I don't have that reference he cites for his method of characteristics treatment, but it seems legitimate from what I could tell from reading that one article. I know my own hasty statement earlier was born of me not really thinking through what was going on and had more to do with reaching sonic velocities in the flow in a steady sense rather than in what was going on, so I suppose it does make sense that this would be possible.

That said, a car exhaust system may not be all that similar since this trombone example relies on the nonlinear propagation of the wave generated by the mouthpiece and the pipe and an internal combustion engine is significantly different than that and the exhaust system almost certainly is designed to avoid any such resonant effects.

 

[Flyboy]

If memory serves, the exhaust flow from many late World War II era piston engines, and designs under development at the time, had supersonic exhaust flow at full power while on the ground. The Merlin engine is a good example of this, especially given the "short stack" exhaust arrangement.

 

[MisterDynamics]

When fluids at supersonic speeds flow over a given structure, the flow on the structure’s surface is usually always subsonic. But the flow above the surface at thicker parts of the fluid’s Boundary Layer will flow at supersonic speeds.

Structures to permit supersonic fluid flow are designed with a shape that bypasses the Oblique shock waves of supersonic fluid flow in which the Sound Barrier held behind the supersonic fluid’s motion allowing the fluid above the surface to penetrate past the Sound Barrier.

A good example of supersonic fluid structures is with supersonic aircraft. The airfoil profile shape is like that of a diamond. This allows the Oblique shock waves of the supersonic fluid flow to dampen at the leading edge of the wing while the Sound Barrier is bypassed before the fluid flow.

The actual fluid flow speed upon the surface of the wings and jet nozzle exterior and interior duct will remain at subsonic speeds while the fluid flow above or around them (thickening Boundary Layer) will move at supersonic speeds along with the supersonic aircraft.

A rifle bullet is another good supersonic fluid flow structure in which the fluid flow on the bullet’s surface remains subsonic while the flow around the surface of the bullet’s Boundary Layer is at supersonic speeds along with the bullet itself.

Same with the barrel of the rifle’s interior structure, along with the lands & grooves to rotate the bullet within the barrel for bullet Rigidness in Space stability, the gas flow over the rifle’s barrel interior surface is subsonic while the area around the barrel’s interior surface has supersonic gas flow.

It does not seem likely that an internal combustion piston engine’s exhaust exhibits supersonic exhaust gas flow.

Although some internal combustion piston engines might have supersonic exhaust gas flow such as piston engines with a very tiny cylinder bore and exhaust manifold might expel exhaust gas pressure at supersonic speeds.

But for the typical gas piston engine this is highly unlikely.

Either way, regardless if a gas piston engine’s exhaust gas flow is supersonic or not, the surface of any supersonic flow structure will always be kept at subsonic speeds while the fluid flow around or above the surface will travel at supersonic speeds.

Speed of Sound in Ft/Sec (Cs) = [√(460 + °F)] x [49.022] Regards,

- MisterDynamics -

January 07, 2014

 

[boneh3ad]

When fluids at supersonic speeds flow over a given structure, the flow on the structure’s surface is usually always subsonic. But the flow above the surface at thicker parts of the fluid’s Boundary Layer will flow at supersonic speeds.

Well, yes, that's kind of the definition of the boundary layer. The flow slows down as it gets closer to the surface, and at some point it falls below the local speed of sound (which increases as you approach the surface since the temperature increases).

Structures to permit supersonic fluid flow are designed with a shape that bypasses the Oblique shock waves of supersonic fluid flow in which the Sound Barrier held behind the supersonic fluid’s motion allowing the fluid above the surface to penetrate past the Sound Barrier.

A good example of supersonic fluid structures is with supersonic aircraft. The airfoil profile shape is like that of a diamond. This allows the Oblique shock waves of the supersonic fluid flow to dampen at the leading edge of the wing while the Sound Barrier is bypassed before the fluid flow.

There is no such thing as the sound barrier. The sound barrier was a phenomenon postulated by some early aerodynamicists as a result of the fact that drag appears to approach infinity as you approach Mach 1. Mathematically, this is described by the Prandtl-Glauert singularity. As it turns out, however, the Prandtl-Glauert transformation is a linearized relation of a highly nonlinear phenomenon, and as you approach the speed of sound and the physics get increasingly nonlinear, Prandtl-Glauert gets increasingly invalid. There is no infinite pressure rise and there is no sound barrier.

Objects designed for supersonic flow are designed in a variety of ways for a variety or reasons depending on the situation. Using your wing example, supersonic airfoils are typically quite close to being biconvex (essentially a rounded diamond) not because it somehow helps them "penetrate the sound barrier", but because in a supersonic flow, that is essentially the most efficient airfoil you can have outside of a simple flat plate. You obviously can't use a thin flat plate for structural reasons, so they go with the slightly thicker biconvex shape. There is nothing about them that "allows the Oblique shock waves of supersonic fluid flow to dampen at the leading edge of the wing", as that statement doesn't really even make sense.

For an object traveling at supersonic speeds (or with air moving at supersonic speeds over it) oblique shocks are inevitable in any situation that tends to turn the flow into itself (compress it). In fact, oblique shocks are quite beneficial as opposed to the alternative: bow shocks. Supersonic vehicles are often slender to avoid the formation of bow shocks at the tip and leading edges. The wave drag from a bow shock is substantially greater than that from oblique shocks.

 

[MisterDynamics]

boneh3ad,

Thanks for the correction.

My my previous comment is full of errors and and over-simplified. It's been a while since I did any work with High Speed Aerodynamics for I have been working with Electrical & Piston Powerplants for the longest time.

Tell me what you think of the following below.

I will just quote aerodynamics text directly from my 'Jeppesen Airframe & Powerplant General Textbook' (1997) in the High Speed Aerodynamics Chapter; Sections: Normal Shock Waves, Expansion Waves and High-Speed Airfoils.


NORMAL SHOCK WAVES: If a blunt airfoil passes through the air at a supersonic velocity, the shock wave cannot attach to the leading edge. Instead, the shock wave forms ahead of the airfoil and perpendicular to the airstream.

When the airstream passes through a normal shock wave, its direction does not change. However, the airstream does slow down to a subsonic speed with a large increase in its static pressure and density. A normal shock wave forms the boundary between supersonic and subsonic airflow when there is no change in direction of air as it passes through the wave.

In addition to forming in front of the leading edge, normal shock waves also form on an airfoil in transonic flight. For example, as an airfoil is forced through the air at a high subsonic speed, the air passing over the top of the wing speeds up to a supersonic velocity and normal shock wave forms. Once the shock wave forms, it slows the airflow beyond the wave to a subsonic speed. These shock waves form on top of the airfoil first and then on the bottom. As airspeed increases beyond the transonic range, both shock waves move aft and attach to the wing's trailing edge to form an oblique shock wave.


EXPANSION WAVES: When a supersonic stream of air turns away from its direction of flow to follow the surface of an airfoil, its speed increases and both static pressure and density decrease. Since an expansion wave is not a shock wave, no energy in the airstream is lost.


HIGH SPEED AIRFOILS: Transonic flight displays the greatest airfoil design problems because only a portion of the airflow passing over the wing is supersonic. When an airfoil moves through the air at a speed below its critical Mach number, all of the airflow is subsonic and the pressure distributions are as your would expect. However, as flight speed exceeds the critical Mach number for an airfoil, the airflow over the top of the wing reaches supersonic velocity and a normal shock wave forms. Normally, the airflow over the top of a wing creates an area of low pressure that pulls the air to the wing's surface. However, when a shock wave forms on the top of the wing, airflow passing through it slows causing the air's static pressure to increase. This destroys the area of low pressure above the wing and allows the air to separate from the surface. This shock-induced separation causes a loss of lift and can reduce control effectiveness.

Because the supersonic flow is a local condition, its effects can be reduced by the use of vortex generators. A vortex generator is a small airfoil mounted ninety degrees to the surface of the wing. It has a low aspect ratio and produces a strong vortex or flow of controlled air that moves high energy air from the airstream into the boundary layer. This vortex delays airflow separation. To obtain maximum benefit, vortex generators are mounted in pairs so the vortices are combined. Although they actually add drag at low speed, the benefit at high speed is a good tradeoff.

Airfoil sections designed for for supersonic flight are typically a double wedge or biconvex shape with sharp leading and trailing edges. Their maximum thickness is at the 50% chord position. As soon as either of these airfoil sections pass through the transonic range, oblique shock waves attach to the leading and trailing edges. The expansion waves form at the point where the airflow must deflect to follow the surface. Since there are no normal shock waves on either airfoil section, there is no subsonic airflow.



Regards,

- MisterDynamics -

January 08, 2014

 

[AlephZero]

That said, a car exhaust system may not be all that similar since this trombone example relies on the nonlinear propagation of the wave generated by the mouthpiece and the pipe and an internal combustion engine is significantly different than that and the exhaust system almost certainly is designed to avoid any such resonant effects.

One way to get a handle on the exhaust question is estimate the speed of the piston, since any expansion of high pressure exhaust gas is wasted energy so far as the engine is concerned.

Typical numbers for a small European car : stroke 85mm, max power at 6000 RPM = 628 rad/sec, so assuming simple harmonic motion (which is crude, but the right order of magnitude) the max piston speed is 628 x 0.085/2 = less than 30m/s, which is way below the speed of sound.

 

[boneh3ad]

In addition to forming in front of the leading edge, normal shock waves also form on an airfoil in transonic flight. For example, as an airfoil is forced through the air at a high subsonic speed, the air passing over the top of the wing speeds up to a supersonic velocity and normal shock wave forms. Once the shock wave forms, it slows the airflow beyond the wave to a subsonic speed. These shock waves form on top of the airfoil first and then on the bottom. As airspeed increases beyond the transonic range, both shock waves move aft and attach to the wing's trailing edge to form an oblique shock wave.

Shocks do form on a transonic airfoil for the stated reason, but they are not, in general, normal shocks. Under the right conditions you may form a normal shock there, but they will generally be oblique.

Other than that, the rest of what you wrote seems to be correct, albeit simplified. For example, on a supersonic airfoil, there is not necessarily only oblique shocks attacked to the leading and trailing edges. If the angle of attack gets large enough, you can see expansion waves form instead.

 

[boneh3ad]

Shocks do form on a transonic airfoil for the stated reason, but they are not, in general, normal shocks. Under the right conditions you may form a normal shock there, but they will generally be oblique.

Other than that, the rest of what you wrote seems to be correct, albeit simplified. For example, on a supersonic airfoil, there is not necessarily only oblique shocks attacked to the leading and trailing edges. If the angle of attack gets large enough, you can see expansion waves form instead.

The problem I see here is that a piston need not be itself moving supersonic in order to generate a shock wave, though that is not the same thing as a supersonic flow, obviously. For a piston in a tube it would certainly have to be moving faster than the speed of sound to set up a supersonic flow behind any induced shock wave, but in an engine I really hesitate to make that conclusion since the exhaust gas gets sent out of the combustion chamber through smaller diameter tubes, where it will accelerate and eventually go through a diverging section.

If that initial constriction is such that it chokes the flow, it is conceivable that under the right conditions you could get a supersonic flow coming out the other end into the exhaust system, but given all the losses in such a system, it would really surprise me.

...but I won't discount the possibility.

 

[Physics_Kid]

isnt this fairly easy to calculate. piston certain rpm = piston speed, the flux between piston area and movement has to equal that on on any slice of exit (given you fix some vaiables), probably most speed right at the exhaust valve. perhaps easy to calculate using a incompressible fluid, but gases are compressible, so i suspect there is a threshold rpm for any given motor where exhaust speed at the exit valve goes supersonic.