Turbulent to Laminar Air Flow Conversion

[Helepolis]

Hello all,

As the title suggests I'm trying to convert a turbulent air flow to a laminar flow in a closed system (oscillating flow).
The turbulent airflow occurs as a result of "rough" geometry, but further down the road the path clears out but the flow still caries the turbulence.
I need the flow o be laminar only in one direction.

If I'm not mistaken, it is possible to stabilize the flow using a specific geometry, something like the intake cone of jet planes.
But unlike the jet planes, in my case I'm working with very low flow rates and a small pipe cross section of 25mm max.

In case someone asks, when the air is pushed in one direction a bolus of particles is injected into the stream and the goal is to carry the bolus with minimal dispersion.
At the end of the previous step the air is sucked back, but this time there is no significance for the flow regime.

If someone could direct me to some relevant material it would be appreciated.

 

[hendersdc]

What velocities are you dealing with in the pipe? If it is high enough there is nothing you can do to prevent turbulence.

To get an idea if it is possible, here is a rough formula for Reynolds number, if it is below 2100 the flow could be considered laminar in your pipe.

Re = (U * Dh) / v

Where:

U = Average velocity over the flow cross sectional area [m/s]
Dh = hydraulic diameter which for a round pipe is just the ID of the pipe [m]
v = kinematic viscosity of your fluid [m^2/s]

Also, when you talk about dispersion of the particles, are you concerned about radial dispersion, longitudinal dispersion, or both?

 

[Helepolis]

About the Re number, the flow is in the range of 1200-600 so it is laminar, but at some point there are geometrical features that disturb the flow and there my knowledge ends, in how to quickly restore the flow back to laminar.
During my degree we studied fluid mechanics but only for flows in pipes without physical disturbance (only surface roughness and bends), so my knowledge is partial on this matter.

For the dispersion, I'm more concerned with the longitudinal although if the bolus reaches the boundary layer it will stretch the bolus to a cone shape (Poiseuille flow profile), so I'm not sure how i can keep to bolus with minimal dispersion.

 

[GregLocock]

Loosely speaking, if you decelerated the flow to near zero, and gave it some time for the vortices to die out, and then accelerated the flow again, would that result in laminar flow? If so then a smooth 'antiventuri' might do the job.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Sparweb]

Settling chambers are used for this, as GregLocock describes.
Also note that flow straighteners (guide vanes) may also be needed, because the currents may have a rotation which will not be dispersed in the settling chamber, and turn into a swirl again as the air exits.

STF

 

[rb1957]

as sparweb said, like in wind tunnels.

another day in paradise, or is paradise one day closer ?