Problem with parabolic flow profile in a tube

[Henrik6]

Hi smart people!

I am a chemist (doing a Ph.D in Analytical chemistry) and I pump plugs of chemicals through tubes (super fun!).
I would like to have a flat flow profile, not a parabolic one like I am having.

My question is this:
Is there a shape (of tube) where I can get a flat flow profile?

The liquid at the wall is traveling slower than in the middle. Should I make a conical "tube" to force the liquid at the wall to flow faster and catch up with the middle?

All help appreciated!

-H

 

[zdas04]

Working on a PhD in Chemical Engineering and you aren't aware of the No-Flow Boundary? Two molecules in contact must be traveling nearly the same speed. If your pipe wall is stationary (relative to the fluid), then you have to have some amount of fluid going that speed (zero). The next "layer" is going a touch faster than zero, etc. The resulting flow profile is not parabolic (and I would expect a PhD ChemE to know that as well), but it is bowed and has a maximum velocity some distance from the pipe wall. Hopefully you can now answer your own question (because student posts are not allowed and this thread will not be here long).

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[Compositepro]

If you make your fluid shear-thinning (like mayonnaise), it will have a more plug-like flow. Gelatin is one type of thickener. Also, high Reynolds number will promote plug flow.

 

[MikeHalloran]

Technicon used air bubbles to provide longitudinal separation between slugs of diluted blood.

There is a Freon that's liquid at STP that can be used for the same purpose, and is heavy enough to make settling and recovery easy, and is of course almost incompressible.

In either case, because of contact with the wall as Zdas noted, there is a net toroidal flow within the bolus of interest. If that's a problem for you, you need to find a universe where the physics is different.



Mike Halloran
Pembroke Pines, FL, USA

 

[Henrik6]

Thanks experts I appreciate you taking the time to answer

Im working in the field of High Pressure Liquid Chromatography (HPLC) where our "tubes" are filled with semi-permiable particles and our plugs are parabolic. Changing the liquid is unfortuneltly not an option (sorry Compositepro). So what I understand from you is that I have no chance of changing the velocity distribution in the tube by going with another geometry (i.e conical)?

There is a closely related field where the flow is electroosmoticly driven. They have ionised liquid and charged wall surface (called Capillary Electrophoresis). They have "plug flow" due to the entire plug experiencing an electric field.


Perhaps my lack of knowledge in your domain is making it hard for me to correctly describe the problem. Apologies for that. I have attached a figure, my case on the bottom, the desired case on top.


----
Thanks for the Galileo quote Mr. Simpson, Ill keep it in mind
MikeHalloran: I tried to ask my supervisor to change the fundamental rules of reality, but he said he did not have time this week

 

[btrueblood]

Are you pushing fluid with a syringe, as shown in the bottom figure? Is the issue then not the flow profile, but the shape of the meniscus? You could try changing the chemistry (surface energy) of the tube wall to get the wetting angle of the fluid as close to 90 degrees as possible...

Otherwise, the only way to have a flat flow profile in normal fluid flow is to consider very high Reynold's numbers for short distances downstream from the entry.

 

[BigInch]

"If things aren't exciting, you're just not moving fast enough." Mario Andredtti

 

[katmar]

The velocity profile is affected by the Reynolds number, in that for low Reynolds numbers (<2000) the flow is laminar and the variation in velocity between the layers is larger. Any fluids text will show you this. So what you want is a high Reynolds number which tends to give a flatter velocity profile. For a fixed mass flow rate (i.e. kg/s) the Reynolds number will increase as either the diameter or the viscosity decrease. So these two variables, diameter and viscosity are all that you have to play with.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[MikeHalloran]

Can your particle material be supplied as a paper-thin sheet?

If so, you could place it between rigid plates, each having half a manifold on two parallel edges. Insert a sheet gasket on each face of the porous medium and clamp it tight, and you've got a pressure flow analog of electrophoresis.



Mike Halloran
Pembroke Pines, FL, USA

 

[btrueblood]

There was some research done some 20+ years ago on a microgravity electrophoresis seperation device. It used moving walls (think of two parallel running conveyor belts) to keep the flow profile uniform.

 

[btrueblood]

Thinking back...not sure if that separator device ever got tested. It was a NASA "Get-Away Special" (aka GAS can) experiment that was to have flown on a Space Shuttle mission. Challenger exploded a few months later and a lot of experiments got shelved in the mayhem that followed.

 

[3DDave]

You could make the flow annular by introducing an obstacle in the tube. Fore example flow straighteners will cause the outlet to be very uniform in flow characteristic.

But if you want true plug flow, introduce plugs.

Is the reason for the question to produce discrete segments of different compositions in the same tube?

 

[racookpe1978]

If you using a plunger, shape the top of the plunger with a recess so the "slug" of the moving mass of liquid ahead of the plug approximates a flatter surface. With a fluid getting pushed into a second fluid, it will never be "flat" as drawn.

If the fluid is being pushed into a vacuum, then you'd have a chance if you can exactly predict the front "curve" of the fluid using a flat-surface plunger. Then make the recess to flatten-out the curve.