Effect of a bar in air flow

[Dmanley]

Hello,

Could someone help me out with this little question?

I'm trying to find out what the length of a turbulent area would be behind a bar in a given air flow?

My situation is a bar in a laminar air flow unit in the pharma industry.

The bar is 20 mm in dia and the air flow is laminar at 0.45 m/s.
At what distance would the air flow return to laminar after the affect of the bar in its path?

If some one could point me in the direction of solving this that would be great.
If any more information is needed just ask.

Thank you,
Derry

 

[40818]

Very difficult subject.

Though lacking in theory, the following link will hopefully point you in the right direction.

http://www.eng.fsu.edu/~shih/succeed/cylinder/cylinder.htm

Some of the pointers:

Wake: Consider a fluid particle flows within the boundary layer around the circular cylinder. From pressure distribution measurements, the pressure is maximum at the stagnation point and gradually decreases along the front half of a cylinder. The flow stays attached in this favorable pressure region as expected. However, the pressure starts to increase in the rear half of the cylinder and the particle now experiences an adverse pressure gradient. Consequently, the flow separates from the surface and creating a highly turbulent region behind the cylinder called the wake. The pressure inside the wake region remains low as the flow separates and a net pressure force (pressure drag) is produced.

Vortex Shedding: The boundary layer separates from the surface forms a free shear layer and is highly unstable. This shear layer will eventually roll into a discrete vortex and detach from the surface (a phenomenon called vortex shedding). Another type of flow instability emerges as the shear layer vortices shed from both the top and bottom surfaces interact with one another. They shed alternatively from the cylinder and generates a regular vortex pattern (the Karaman vortex street) in the wake. The vortex shedding occurs at a discrete frequency and is a function of the Reynolds number. The dimensionless frequency of the vortex shedding, the shedding Strouhal number, St = f D/V, is approximately equal to 0.21 when the Reynolds number is greater than 1,000.

Maybe if you ask nicely, there might be somebody who would knock up a quick CFD run of your problem and give an actual answer?

 

[MadSplinter]

Well, you need more data to find a solution.
In particular, you need the viscosity of the fluid and the temperature, i.e. you need to calculate the Reynolds number of your flow. The Reynolds number defines the ratio between inertial forces and viscous forces of the flow. In general, if the Reynolds number of the flow is sufficiently low, the flow will stay laminar even after the bar. For a flow in a duct, the maximum Reynolds number for a laminar flow is about 2100. Look at this wikipedia page for reference:

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

__________________________________________

Everything should be made as easy as possible, but not easier

Albert Einstein

 

[rb1957]

does the bar do anything (like adding fluid/gas into the airflow) ? ... this'll probably affect the flow around the cyclinder.

do you have an idea about how quickly you need the flow to return to laminar ? maybe it'll be obviously too short or far enough that it'll be ok.

i liked the idea about a quick CFD model ... almost an ideal application.

i suspect that you won't like the answer, and may need to trick the cyclinder ... does it have to be a cyclinder ? (a small amount of airfoil profiling will probablby help. maybe golf-ball dimples too ?? can you test it out ?

 

[Dmanley]

Thanks all, for your responses!

I'll explain the situation more and it may help get my issue across.

I'm in manufacturing facility which fills sterile vials in a Grade A (Class 100) environment, as the vials move through the filling machine and then on to a freeze dryer, the vial pass through a complex loading systems made up of conveyors and pushers.
As there can be different size vials used in the systems, support bars are used to hold guide rails in place along these conveyors.
The vials pass under these bars as they move through the system.
The sterility of these vial is crucial to the production process, there fore Grade A and Laminar Air Flow must be maintained around the vial.

When I refer to laminar air flow here, I don’t mean in an engineering sense, such are boundary lay etc., I am referring to the use of unidirectional air flow to help keep any particulates, dirt, etc moving away from the vials.

The problem is how high above the vial must the bar be?

The fluid is air at ambient conditions at a velocity of 0.45 m/s as stated above.

What I would love to get from this is a picture showing varying sizes of bar, also showing the length of the wake (?) behind it.

Once again thank you for all you comments, I would love to be in a position to model this in CFD but alias we have a though time getting an upgrade to Win XP from 2000 never mind forking out on training and a package.

From my college years I remember seeing pictures similar to what I describe above, but I could be incorrect.

Thank you,
Derry

 

[rb1957]

i think your best answer is contacting a CFD "shop" lots of people will do this analysis (for a price). you might be able to interest a nearby college/university to study the problem for you (i think they might like having a real world application for the students to model).

i think this is an ideal problem for CFD, and you'll get lots of nice pictures of the wake for different positions of the bar.

since the "bar" is structural, you could consider different shapes, eg plate (rectangular X-section).

in the absense of real numbers, make it as high as you can ! maybe there are other ways to support the rails, that don't intrude on the space above the vials (tension bracing off to the sides). maybe you could add a "splitter" sheet under the bar, to separate the disturbed flow from the airspace above the vials ?

 

[Dmanley]

Thanks for the response rb1957,

I just may send the info to our local college.

About the splitter sheet, you would have to clean under that sheet quite regularly hence the reason we use clean air.

Thanks again to all for there input.

Regards,
Derry

 

[IRstuff]

I'm not sure that the question you've posed makes sense to me in the context that you've posed.

Is there some supposition that a non-laminar flow is non-sterile? The air is the same air, regardless of whether it's laminar or not. Either the bar is sterile and adds nothing to the air flow or it's not. What happens further downstream seems irrelevant to me.

TTFN

FAQ731-376

 

[rb1957]

i think what they're concerned about is the turbulent wake increases the chance of something coming in contact with their carefully sterilised vial

 

[IRstuff]

But, that would be the case no matter how far downstream the vial is located?

If contaminants enter the airstream, making the airflow laminar again won't get rid of the contaminants, once they're in.

TTFN

FAQ731-376

 

[crysta1c1ear]

As I see it, air molecules are travelling at around 330 m/s, as they can transport sound by bumping into each other and sound travels at around that speed.

You can estimate the speed of air molecules another way. Their kinetic energy in any given direction is say 300 kelvin and their mass is 29 atomic mass units. By knowing the mass and kinetic energy, it is simple to compute the velocity.

So if there is an airflow of 0.45 m/s, I see that as in some ways irrelevent compared to the random motion of the air molecules themselves.

Suppose you stand by some motorway roadworks. There might be more cars travelling past in one direction than another, ie a net flow of cars down the road in one direction of 0.45 miles per hour. The fact that there is a net flow of cars down the road does absolutely nothing to stop a wanted man from driving along in the opposite direction.

Now let's say that all the cars travelling down the road (in one direction) have been through a police check searching for the wanted man. That still doesn't help prevent him from travelling up the road at 70 miles an hour,
even though the net flow of traffic is down the road at an average of 0.45 miles per hour.

If you are going hunting, it is easier for animals to smell a human if he is upwind of them than downwind of them. It is common sense that the random motion of the air molecules carries more of the smell downwind than upwind. How far upwind does somebody need to be to avoid detection? That is a sort of probability question and clearly would depend on the wind speed.

=

I am aware I haven't come remotely close to answering your original question either. Sorry!

 

[crysta1c1ear]

Oops, I took too long writing that and by the time I posted it, IRstuff and rb1957 have said the same sort of thing much more elegently and succinctly.

 

[MadSplinter]

By the way, modeling the shape of the bar as a symmetric airfoil makes the drag 10 times lower, with an obvious effect over the lenght and the thickness of the wake.
Since the bar has a diameter of 20 mm, even in the case of a turbulent flow, its effects would be practically negligible.

__________________________________________

Everything should be made as easy as possible, but not easier

Albert Einstein

 

[mohr]

Hi Demanley :

Try with Naca TR - 1911 of A. Roshko. It may be useful for
you .
Cheers

 

[vortexman]

Hi Dmanley,

I think you want to know the length of the wake behind a cylinder in an oncoming flow that is not quite laminar, but with a low intensity of turbulence. This is a fairly hard problem for CFD, but a very easy problem experimentally. In fact, this has been tested many, many times, and I would be surprised if you couldn't find the wake length as a function of Reynold's number via a Google search. If you have a Fluid Mechanics textbook, or some kind of mechanical engineering or chemical engineering handbook, you would probably find it in there also.

Good luck.