high viscosity flow with low Reynolds' number

[someguy79]

How can I adequately calculate pressure drop in a line if viscosity is ~3000cP and Reynolds' number is under 10? When I look at a Moody plot, it doesn't seem to be anywhere near the plotted curves?

Using the common relation of 64/Re=f doesn't seem to be giving reasonable results either.

Based on previous jobs, this should not result in very high pressures, or much pumping power.

I don't expect that pressure drop on the discharge side of my metering pumps will be a problem, but the NPSHa could be.

 

[Latexman]

You can use Poiseuille's law for laminar flow directly without calculating f. Refer to Crane TP410 or other fluid flow references.

Good luck,
Latexman

 

[katmar]

Using the relation of f=64/Re is the correct way to go. It might be a bit of a shock to see friction factors close to 10, but that is the reality. As you say, you can still get relatively low pressure drops because the velocity will be low.

If you combine this with the Darcy-Weisbach equation you in fact get Poiseuille's Law.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

It is Newtonian flow, right?

"If everything seems under control, you're just not moving fast enough."
- Mario Andretti- When asked about transient hydraulics

http://virtualpipeline.spaces.msn.com

 

[someguy79]

It is a Newtonian fluid.

Thanks for the suggestions. I'll have to re-check my numbers, again.

 

[worrell]

Newtonian fluid- one who's viscosity is not dependent upon its velocity. (I tell my kids that.)

 

[btrueblood]

How do you wake them back up again? :0

 

[someguy79]

I checked over my numbers and everything was fine there. So I got a hold of the chemical supplier. As it turns out, the fluid (Dow Corning 544 Antifoam Compound) is shear-thinning. Mystery solved! Biginch, you had the right question. I was hampered with some bad info and bad assumptions.

When I asked if they have any quantitative characterization for that behavior; they said it's not typically necessary in normal use.

Unless I find some good and easy to use info quickly, I'm going to have to guess on pipe & valve sizing based on past projects. I don't think my boss is going to pay, or wait, for lab testing and another round of calcs.

Thank you all for your help.

 

[BigInch]

Most fluids with those kind of viscosities are at least somewhat (if not a whole lot) non-Newtonian. Even if they are mostly Newtonian, you can easily get some wierd results from other factors you might not be expecting, such as heating of the outer layers near the pipe wall thereby reducing the viscosity there while the inner fluid stays cooler. I've had a large diameter bitumen pipeline that could actually be in transition-turbulent flow in the outer layers while flowing as a laminar plug at the centerline!


"If everything seems under control, you're just not moving fast enough."
- Mario Andretti- When asked about transient hydraulics

http://virtualpipeline.spaces.msn.com

 

[jmw]

If it is pseudoplastic, then above a certain shear rate it will effectively behave as a Newtonian fluid.

Conversely, if the pipes i not heated and poorly insulated then the effective pipe diameter can become very much smaller than you think. Chocolate lines are a classic example, the chocolate cools at the pipe wall, its viscosity increases, it slows down and it cools some more. Most of the lfow takes place in the very centre of the pipe line at far higher flow velocities than you'd expect based on pipe size.

JMW

http://www.ViscoAnalyser.com