Understanding the Venturi Effect in Fluid Dynamics

[chephy]

Hi,

I have a pretty elementary question, but I'm having trouble finding an adequate explanation on the web. I'm trying to figure out how a carburetor works, and am having difficulty figuring out how a venturi tube works.

On the one hand, as air flows through a narrow section of a pipe, it accelerates, which, according to Bernoulli should lead to a drop in pressure in the narrow section.

On the other hand, as the air is forced through a narrow tube, it gets "squeezed" (it's compressible, after all), so that should lead to an increase in pressure.

So, what is really going on?

Thanks in advance. (I hope I am posting this in the correct section. This is not for homework or any course, I'm just trying to figure out a concept...)

 

[rcgldr]

I'm trying to figure out how a carburetor works

A carburetor involves more than just venturi effect. When you have the open end of a pipe exposed to a cross wind, the deflection of that wind away from the hole at the open end of the pipe creates a low pressure vortice, strong enough to draw fluid even if the cross-wind is at or even slightly above ambient pressure. You can duplicate this using a straw in a container of water, and using a blow dryer to create the cross wind. With ear plugs and a secure unbreakable container, you can get more effect from a leaf blower.

Getting back to the venturi effect, other than initial start up, mass flow is constant at all points within a constrained flow (otherwise mass would be accumulating at some point within the flow). Since the mass flow is constant, it's speed is higher in narrower sections of a pipe. Assuming that the pipe doesn't cause the acceleration to higher speeds, then the only source for acceleration is a pressure differential within the gas (or fluid) itself, higher pressure in the larger diameter slower moving flow, lower pressure in the smaller diameter, faster moving flow.

In the real world, mass flow is constant, but pressure is reduced as a flow travels along a pipe because of friction between the pipe and a fluid or gas, and viscosity within the fluid or gas. This reduction of pressure enhances the venturi effect.

As an example of venturi effect, here is a link to a "pump" that connects to a tap, and uses venturi effect to create low pressure within a chamber to draw fluid in from a side connector, and then both flows exit via another opening at the far end from the tap.

http://andysworld.org.uk/aquablog/?postid=247

If you follow the Canadian patent link to the diagrams you'll get to a picture of the internals. Figure 4 shows the device in it's venturi mode. Figure 5 shows the device with the far end closed off, so that the device can be used to fill an aquarium as well as drain it.

internals.html

 

[russ_watters]

Except in the case of very high speed flow, Bernoulli's Principle assumes that the fluid is not compressible. In air, up to about 220 mph, the difference is negligible. So the only way for mass flow to be constant in the system is for the velocity through the obstruction to be higher than the velocity in the free pipe sections.

 

[chephy]

Thanks for the explanations.

So let me just get one question straight: is the air less dense in the narrow section of the carburetor? If not, how can it have lower pressure there? If yes, it turns out that air actually expands when it's forced through a narrow opening. That seems kind of counterintuitive (of course I know that counterintuitive does not always mean wrong).

Also, is there any explanation why this actually happens? I mean, the mass flow and energy conservation argument explain that acceleration must be caused by a pressure differential, but what creates the pressure differential?

 

[russ_watters]

So let me just get one question straight: is the air less dense in the narrow section of the carburetor?

By the basic form of Bernoulli's principle/equation and for practical purposes, no. In reality, a teeny weeny ittty bitty little bit.

If not, how can it have lower pressure there?

The pressure change required to generate a speed of a few hundred miles per hour is a very tiny fraction of atmospheric pressure. So you can apply the equation and measure the pressure change without considering what it does to the density of the air.

Also, is there any explanation why this actually happens? I mean, the mass flow and energy conservation argument explain that acceleration must be caused by a pressure differential, but what creates the pressure differential?

In order for a fluid to move through a pipe, a pressure differential must exist. How you create a pressure differential has a lot of possible answers. A fan, a pump, a pressurized tank, a down-stroke on a piston, etc.

Also, remember there is more than one kind of pressure. It may make more sense to think of it this way: total pressure in a pipe is constant. When air flows through a venturi, static pressure is converted to velocity pressure.

 

[rcgldr]

Is the air less dense in the narrow section of the carburetor?

Yes, the reduction in pressure would correspond to reduction in density, but this reduction in pressure is small compared to the ambient pressure of air, so the change in density is also small. Note that the venturi tubes in a carburetor used to draw fuel only occupy a fraction of the total cross sectional area of the carburetor's intake. The rest of the carburetor is designed to avoid overly restricting the air flow or overly reducing it's pressure. That aquarium pump I referred to is a much better example of venturi effect.

What creates the pressure differential?

Bernoulli doesn't cover this. In the case of a carburetor, the engine acts as an air pump, reducing the pressure on the intake side, which results in a flow into the engine's manifold and cylinders. Generally, the creation of a pressure differntial involves some amount work, which Bernoulli doesn't cover. Bernoulli only covers how a gas or fluid will respond to a pressure differential once it's created and/or maintained, and it's just an approximation that doesn't cover issues like turbulence.

Also, remember there is more than one kind of pressure. It may make more sense to think of it this way: total pressure in a pipe is constant. When air flows through a venturi, static pressure is converted to velocity pressure.

In the real world, if there's a flow in a pipe, then total pressure decreases with distance traveled, the energy being converted into heat because of the friction between a pipe and a fluid or gas.

The density change isn't all that small either. It's just that pressure differentials are usually small compared to an ambient pressure of 14.696 psi == 101325 pascals, which relates to the effect on density. For example the wings on a 747 only result in about .5 psi pressure diffenential, and this is divided into lower than ambient zone above and higher than ambient zone below, so the deviation from ambient in each zone is well below .5 psi.

 

[chephy]

Is the air less dense in the narrow section of the carburetor?

By the basic form of Bernoulli's principle/equation and for practical purposes, no.

Yes

I think, I am confused now... :)

If the change in density is so small, then what causes the creation of "partial vacuum" in the carburetor that some books refer to? And how come expansion - and thus cooling - of the air in the carburetor is given as an explanation for carb ice, if the air density is not appreciably decreasing (i.e., air is not expanding)?

 

[vibjwb]

Another way of looking at this is the top of an airplane wing.
The top of an airplane wing is like one side of a venturi. The air speeds up as it moves over the top of the wing causing a low-pressure area on the surface and if the bottom of the wing is flat the pressure stays the same so the wing will be sucked up by the low pressure on top. With a carburetor there is a low-pressure on the surface of the venturi as the air speeds up going threw it. A carburetor has a small hole in the venturi area, which sucks gasoline in because of the low-pressure at the surface.

 

[rcgldr]

If the change in density is so small, then what causes the creation of "partial vacuum" in the carburetor that some books refer to?

The partial vacuum occurs because the engine is "sucking" air into the cylinders. As the pistons move down during the intake cycle, the air is drawn into the cylinders because the pressure is reduced.

A water pump can suck water, reducing it's pressure with very little change in density since water is nearly incompressable. For example, if you increase the pressure of water from ambient at 14.7 psi to 1000 psi, it's density only increases by .32% (by a factor of 1.0032). A gallon of water at 1000 psi reduced to 14.7 psi would only expand by 2.5 teaspoons (768 teaspoons per gallon x .0032 = 2.5 teaspoons).

And how come expansion - and thus cooling - of the air in the carburetor is given as an explanation for carb ice, if the air density is not appreciably decreasing (i.e., air is not expanding)?

Air approximates an ideal gas where pressure x volume x temperature (kelvin) is constant. Even if expansion is very small where volume doesn't significantly change, then a reduction in pressure corresponds to similar reduction in temperature on the Kelvin scale. If the air temperature is already low, then it only takes a small relative change on the Kelvin scale to reduce temperature to the point that wator vapor in the air freezes and accumulates on the venturi tubes and other parts of the carburetor.

The top of an airplane wing is like one side of a venturi.

Venturi effect requires a enclosed flow, such as a pipe or tube. A wing operates in an open environment. The air follows the curved surface because a void (zero pressure) would be created if the didn't follow that surface (it may follow that surface with turbulent, circular flow if the angle of attack it too high). From Wiki article on wing:

In that case a low pressure region is generated on the upper surface of the wing which draws the air above the wing downwards towards what would otherwise be a void after the wing had passed.

http://en.wikipedia.org/wiki/Wing#Science_of_wings

 

[xxChrisxx]

Jeff you are making this far more complicated than it needs to be bud. It's only leading to confuse the OP in practical terms, what you are saying is fine to someone familiar with the basic principles. The OP is obviously a layperson with regards to this (no offence intended OP) on a technical level your answers are far too detailed.

Chephy up to mach 0.3 air can be assumed to be incompressible. Meaning for the flow speeds through your carb you can assume constant density.

Air pressure is made up of 3 'parts' according bernoulli:
Static pressure - pressure when the air is just sitting around doing nothing.
Dynamic pressure - pressure caused by the air moving.
Pressure due to gravity - does what it says on the tin.
These added together are constant.
v^2/2 + gz + p/density=constant
Dyn. Grav. Static.
What happens in a carb is the venturi effect. it uses Bernullis principle and conservation of flow.

So air travels through the top of the carb, when it reaches the narrow section the air increases in velocity (more dynamic pressure), this is becuase to get the same amount of air flowing thorugh the smaller area, it must speed up. This means the static pressure drops to below atmospheric, causing it to draw in fuel. It's the act of the air speeding up through the narrow section that makes this work.

Thats it, in a very basic nutshell.

 

[rcgldr]

Link to diagram of typical carburetor, note that the fuel exits via a tube that protudes into the "booster". The open end of that tube experiences an additional reduction in pressure because the end of the tube deflects the flow away from the opening. So a typical carburetor has two stages of venturi plus a protuding tube that all combine to reduce pressure.

carburetor_basics.html

 

[chephy]

This is all quite helpful; I'm getting a better understanding, thanks to everyone for replies.

The partial vacuum occurs because the engine is "sucking" air into the cylinders. As the pistons move down during the intake cycle, the air is drawn into the cylinders because the pressure is reduced.

Then why do you need the venturi at all? As far as I (now) understand it's to create region of lower static pressure to draw the fuel in. But if the pistons are already doing the "sucking" and reducing pressure, why isn't that enough to pull fuel in as well?

Air approximates an ideal gas where pressure x volume x temperature (kelvin) is constant. Even if expansion is very small where volume doesn't significantly change, then a reduction in pressure corresponds to similar reduction in temperature on the Kelvin scale. If the air temperature is already low, then it only takes a small relative change on the Kelvin scale to reduce temperature to the point that wator vapor in the air freezes and accumulates on the venturi tubes and other parts of the carburetor.

But now I am confused as to whether any expansion goes on at all. xxChrisxx is saying that the air is not compressed in the venturi and that air density is basically constant throughout the carburetor. If the decrease in pressure is not accompanied by an expansion, then why would it be produce a cooling effect? And why would a book I'm reading to figure this out talk about a "high expansion of air through the carburettor venturi"?

Air pressure is made up of 3 'parts' according bernoulli:
Static pressure - pressure when the air is just sitting around doing nothing.
Dynamic pressure - pressure caused by the air moving.
Pressure due to gravity - does what it says on the tin.

This is helpful, thanks.

So air travels through the top of the carb, when it reaches the narrow section the air increases in velocity (more dynamic pressure), this is becuase to get the same amount of air flowing thorugh the smaller area, it must speed up. This means the static pressure drops to below atmospheric, causing it to draw in fuel. It's the act of the air speeding up through the narrow section that makes this work.

Okay. But why does this have a cooling effect?

 

[rcgldr]

Then why do you need the venturi at all? As far as I (now) understand it's to create region of lower static pressure to draw the fuel in. But if the pistons are already doing the "sucking" and reducing pressure, why isn't that enough to pull fuel in as well?

The carburetor is located before the throttle valve. At partial throttle, the throttle valve restricts the flow and most of the pressure reduction occurs at the throttle valve, so the pressure at the carburetor, which is before the throttle valve, is close to the ambient pressure, and not sufficient to draw in fuel.

But now I am confused as to whether any expansion goes on at all. ... air is not compressed in the venturi and that air density is basically constant throughout the carburetor. If the decrease in pressure is not accompanied by an expansion, then why would it be produce a cooling effect?

Reduction in static pressure corresponds to a reduction in temperature. These result in offsetting changes in density, a decrease in pressure would reduce density, a decrease in temperature would increase density. The net result is a very small change in density. Evaporation of fuel and water also contribute to cooling effect.

Why would a book I'm reading to figure this out talk about a "high expansion of air through the carburettor venturi"?

I'm not sure why your book mentions a high rate of expansion, since it shouldn't occur under normal circumstances.

 

[xxChrisxx]

It cools/(can ice) becuase temperature is based on static pressure. Through the throat static pressure drops (it's on the throat where the icing occurs)

Density is taken from total pressure (static + gravitational + dynamic) which is constant.

 

[rcgldr]

The actual response in terms of pressure, density, and temperature are related to the "heat capacity ratio" or "ratio of specific heats" of the air, and the compressability of air.

wiki_compressible_flow.html

http://en.wikipedia.org/wiki/Heat_capacity_ratio

http://en.wikipedia.org/wiki/Compressibility_factor#Compressibility_of_air

xxChrisxx mentioned that air can be considered incompressable below Mach .3. For Bernoulli like flows, a common pressure change correction factor for air that somewhat underestimates pressure change is 1 / sqrt(1 - M^2), where M is the speed in Mach. At mach 0.1, the correction factor is less than 1%, at mach 0.3, it's about 5% (meaning that the actual change in pressure is about 5% greater if air is considered compressable, as opposed to being considered incompressable). Note this correction factor is for speeds at Mach 0.8 or less. Above this speed it gets complicated.

Regarding carburetor icing, I've looked up articles at a few web sites, and it's a combination of cooling effects, and it's not clear how much of a role each effect takes. Most articles imply that fuel evaporation is the primary cooling effect. If the conditions within the intake drop below the dew point, then water will condense and then evaporate when pressure is reduced at the venturi or past the throttle valve. The other cooling effect, related to a decrease in pressure at the venturi and at the throttle valve have already been mentioned.

Even for just dry air flowing through a venturi, I haven't found an article with a formula for calculating the resulting change in velocity, pressure, density, and temperature.

 

[vibjwb]

Jeff Reid check this out
http://en.wikipedia.org/wiki/Venturi_effect
Read this close to the bottom under examples on this web page
Air flows over the top of an aircraft wing. The stable air around the plane and the foil above the cord line of the wing create a venturi around the wind going directly over the wing. Hence creating a low preasure on top of the wing lifting it upward.

 

[rcgldr]

Jeff Reid check this out
http://en.wikipedia.org/wiki/Venturi_effect

Read this close to the bottom under examples on this web page
Air flows over the top of an aircraft wing. The stable air around the plane and the foil above the cord line of the wing create a venturi around the wind going directly over the wing. Hence creating a low preasure on top of the wing lifting it upward.

I added an update to the discussion page. This idea is commonly disputed at many web sites, including this NASA article:

a wing section isn't really half a Venturi nozzle.

nasa_airplane_bernnew.html

At best if you were to fly a plane into a square tunnel, flying close to the ceiling, then perhaps you'd get the 1/2 venturi effect you're looking for.

As an example of a reverse venturi effect, diffusers used on race cars use expanding sections to reduce the pressure of the air ahead of the diffuser. Included in this article about various ways to create downforce on race cars:

http://www.superhachi.com/theory/downforce

Indy Racing League cars do use "venturi tunnels" on their undertrays and/or sidepods (undertray tunnels are not allowed in Formula 1, I don't know about the other restrictions).

 

[xxChrisxx]

Jeff Reid check this out
http://en.wikipedia.org/wiki/Venturi_effect
Read this close to the bottom under examples on this web page
Air flows over the top of an aircraft wing. The stable air around the plane and the foil above the cord line of the wing create a venturi around the wind going directly over the wing. Hence creating a low preasure on top of the wing lifting it upward.

Wiki is not an authority on this.

The effect described is similar to but not the venturi effect. Ventrui effect only describe flow going through an enclosed cross section where skin friction (viscous effects) do not dominate. Wings in free air are not enclosed.

Wings on planes aren't considered to be a 'venturi effect' as you can neither consider them enclosed, or the flow near the surface to be inviscid. This is also the reason that bernoullis equation does not strictly hold true for wings.