Pressure drop in a pipe - Different results Fluent vs. Analytical

liliana · August 29, 2018, 13:21

Hello all,

I am simulating a very simple caso of a flow in a 2D horizontal pipe (0,2 m diameter and 1m length).

Fluid Density = 827 kg/m³
Fluid Viscosity = 0.026 Pa.s

Inlet = parabolic velocity profile with vmax = 0.016 m/s
Outlet = pressure-outlet with P = 0

The flow is laminar.

So I calculated by hand the pressure drop along the pipe, since it is laminar, fully developed and newtonian fluid with:

dP = rho * (L/D) * (64/Re) * ((v_ave)^2)/2

Then I got arount 0.17.

In Fluent I got 0.083. For this, I analyzed static pressure. Even with total pressure, the result is not the same (0.098).

Does anyone have a similar problem?

LuckyTran · August 29, 2018, 15:27

Just a sanity check, what is the massflow rate that Fluent outputs?

What's your mesh look like? Did you already do a grid dependence study? You'll be surprised how fine a mesh you need for this simple problem.

Also just fyi, this type of problem is really good for simulating using the periodic boundary condition rather than velocity inlet and pressure outlet.

liliana · August 29, 2018, 15:31

Quote:

Originally Posted by LuckyTran

Just a sanity check, what is the massflow rate that Fluent outputs?

What's your mesh look like? Did you already do a grid dependence study? You'll be surprised how fine a mesh you need for this simple problem.

Also just fyi, this type of problem is really good for simulating using the periodic boundary condition rather than velocity inlet and pressure outlet.

I did a grid independence study. It is ok. The mesh is fine enough.

The mass flow rate at inlet is 1.32 kg/s.

blackmask · August 29, 2018, 20:44

You must have made a mistake. If the flow rate Q=1.32 kg/s, then
$v_{avg}=\frac{4Q}{\rho \pi D^2} = \frac{132}{\pi 827} m/s\approx 0.0508 m/s$
Which is three times larger than the maximum value you claimed. For parabolic velocity profile, the average velocity should be half the maximum. Did you multiply the result reported by FLUENT by $2\pi$ ?

liliana · August 29, 2018, 20:45

Quote:

Originally Posted by blackmask

You must have made a mistake. If the flow rate Q=1.32 kg/s, then
$v_{avg}=\frac{4Q}{\rho \pi D^2} = \frac{132}{\pi 827} m/s\approx 0.0508 m/s$
Which is three times larger than the maximum value you claimed. For parabolic velocity profile, the average velocity should be half the maximum. Did you multiply the result reported by FLUENT by $2\pi$ ?

I have found the problem. I did not use axisymmetry. Now it matchs perfectly.

Vishsel · May 14, 2021, 09:45

Hello everyone,

My case is to compare the fluent and openfoam pressure drop results for simple straight pipe.

Pipe datas - Diameter 20mm and 0.2m long.

Solver - simpleFoam

And the pipe is splitted into three parts (60mm,80mm,60mm totally 200mm). and i am using cyclicAMI boundary type for interiors.

I am using k-omega turbulence model and dictionary files are as per in the tutorial. Before using this, i have used realizable k epsilon, but it leads to higher dp, so i changed to k-omega and it gives better results in many projects.

Ansys fluent results and experimental results are exactly matched. Yes, the total pressure drop of openfoam is also reaches very close to an experimental results.
But the cutoff dp (i.e individual pipe dp) was not matching with fluent. In percentage wise, the cutoff dp is high (greater than 10%).

Fluent results and experimental results:
pipe-1 = 3.5 mbar, pipe-2 = 3.82 mbar, pipe-3 = 2.8 mbar, total dp = 10.12mbar
OF results:
pipe-1 = 4.02 mbar, pipe-2 = 3.4 mbar, pipe-3 = 2.25 mbar, total dp = 9.67mbar

y+ plus value for the pipe is max:2.18 and min:0.16, minimum cell size is 0.005mm.

U file

Code:

internalField   uniform (0 0 0);

boundaryField
{
   massflow-inlet
    {
        type            fixedValue;
        value            uniform (0 0 -2.65);
    }
    
    pressure-outlet
    {
        type            inletOutlet;
        value           uniform (0 0 0);
        inletValue      uniform (0 0 0);
    }
    
    wall-pipe-1
    {
        type            fixedValue;
        value            uniform (0 0 0);
    }
    wall-pipe-2
    {
        type            fixedValue;
        value            uniform (0 0 0);
    }
    wall-pipe-3
    {
        type            fixedValue;
        value            uniform (0 0 0);
    }
    pipe121
    {
        type            cyclicAMI;
    }
    pipe122
    {
        type            cyclicAMI;
    }
    pipe231
    {
        type            cyclicAMI;
    }
    pipe232
    {
        type            cyclicAMI;
    }
}

p file

Code:

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 99.858;

boundaryField
{
    massflow-inlet
    {
        type            zeroGradient;
    }
    pressure-outlet
    {
        type            fixedValue;
        value           $internalField;
    }
    wall-pipe-1
    {
        type            zeroGradient;
    }
    wall-pipe-2
    {
        type            zeroGradient;
    }
    wall-pipe-3
    {
        type            zeroGradient;
    }
    pipe121
    {
        type            cyclicAMI;
    }
    pipe122
    {
        type            cyclicAMI;
    }
    pipe231
    {
        type            cyclicAMI;
    }
    pipe232
    {
        type            cyclicAMI;
    }
}

k file

Code:

internalField   uniform 0.0095;

boundaryField
{
    massflow-inlet
    {
        type            turbulentIntensityKineticEnergyInlet;
        intensity       0.05;
        value           uniform 0.0095;
    }

    pressure-outlet
    {
        type            inletOutlet;
        inletValue      uniform 0.0095;
        Value              uniform 0.0095;
    }
    wall-pipe-1
    {
        type            kqRWallFunction;
        value           uniform 0.0095;
    }
    wall-pipe-2
    {
        type            kqRWallFunction;
        value           uniform 0.0095;
    }
    wall-pipe-3
    {
        type            kqRWallFunction;
        value           uniform 0.0095;
    }
    pipe121
    {
        type            cyclicAMI;
    }
    pipe122
    {
        type            cyclicAMI;
    }
    pipe231
    {
        type            cyclicAMI;
    }
    pipe232
    {
        type            cyclicAMI;
    }
}

omega file

Code:

internalField   uniform 127;

boundaryField
{
    massflow-inlet
    {
        type            fixedValue;
        value           $internalField;
    }
    pressure-outlet
    {
        type            zeroGradient;
    }
    wall-pipe-1
    {
        type            omegaWallFunction;
        value           $internalField;
    }
    wall-pipe-2
    {
        type            omegaWallFunction;
        value           $internalField;
    }
    wall-pipe-3
    {
        type            omegaWallFunction;
        value           $internalField;
    }
    pipe121
    {
        type            cyclicAMI;
    }
    pipe122
    {
        type            cyclicAMI;
    }
    pipe231
    {
        type            cyclicAMI;
    }
    pipe232
    {
        type            cyclicAMI;
    }
}

nut file

Code:

internalField   uniform 0;

boundaryField
{
    massflow-inlet
    {
        type            calculated;
        value           uniform 0;
    }
    pressure-outlet
    {
        type            calculated;
        value           uniform 0;
    }
    wall-pipe-1
    {
        type            nutkWallFunction;
        value           uniform 0;
    }
    wall-pipe-2
    {
        type            nutkWallFunction;
        value           uniform 0;
    }
    wall-pipe-3
    {
        type            nutkWallFunction;
        value           uniform 0;
    }
    pipe121
    {
        type            cyclicAMI;
    }
    pipe122
    {
        type            cyclicAMI;
    }
    pipe231
    {
        type            cyclicAMI;
    }
    pipe232
    {
        type            cyclicAMI;
    }
}

fvSolution

Code:

solvers
{
    p
    {
        solver          GAMG;
        tolerance       1e-08;
        relTol          0.05;
        smoother        GaussSeidel;
        cacheAgglomeration on;
        nCellsInCoarsestLevel 20;
        agglomerator    faceAreaPair;
        mergeLevels     1;
    }

    "(U|k|epsilon|omega)"
    {
        solver          smoothSolver;
        smoother        symGaussSeidel;
        nSweeps         2;
        tolerance       1e-07;
        relTol          0.1;
    }

}

SIMPLE
{
    nUCorrectors   2;
    nNonOrthogonalCorrectors 0;
    
    residualControl
    {
        p               1e-10;
        U               1e-10;
        "(k|epsilon|omega|f|v2)" 1e-10;
    }
    
}

relaxationFactors
{
    fields
    {
        p               0.3;
    }
    equations
    {
        U               0.5;
        k               0.7;
        epsilon         0.7;
        omega            0.7;
    }
}

fvSchemes

Code:

ddtSchemes
{
    default         steadyState;
}

gradSchemes
{
    default         Gauss linear;
}

divSchemes
{
    default         none;
    div(phi,U)      bounded Gauss linearUpwind grad(U);
    div(phi,k)      bounded Gauss limitedLinear 1;
    div(phi,epsilon) bounded Gauss limitedLinear 1;
    div(phi,omega)  bounded Gauss limitedLinear 1;
    div(phi,v2)     bounded Gauss limitedLinear 1;
    div((nuEff*dev2(T(grad(U))))) Gauss linear;
    div(nonlinearStress) Gauss linear;
}

laplacianSchemes
{
    default         Gauss linear corrected;
}

interpolationSchemes
{
    default         linear;
}

snGradSchemes
{
    default         corrected;
}

wallDist
{
    method meshWave;
}

Could you please help me out with this issue, how should i improve or match the OF result with experimental result ?

Thank you in advance,
Vishsel

August 29, 2018, 13:21	Pressure drop in a pipe - Different results Fluent vs. Analytical	#1
liliana Member Liliana de Luca Xavier Augusto Join Date: Feb 2013 Posts: 64 Rep Power: 13	Hello all, I am simulating a very simple caso of a flow in a 2D horizontal pipe (0,2 m diameter and 1m length). Fluid Density = 827 kg/m³ Fluid Viscosity = 0.026 Pa.s Inlet = parabolic velocity profile with vmax = 0.016 m/s Outlet = pressure-outlet with P = 0 The flow is laminar. So I calculated by hand the pressure drop along the pipe, since it is laminar, fully developed and newtonian fluid with: dP = rho * (L/D) * (64/Re) * ((v_ave)^2)/2 Then I got arount 0.17. In Fluent I got 0.083. For this, I analyzed static pressure. Even with total pressure, the result is not the same (0.098). Does anyone have a similar problem?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Fluent cyclone pressure drop	mkal	FLUENT	1	February 20, 2020 00:55
Matching analytical and simulated pressure drop	harsh_kgp	FLUENT	1	October 18, 2017 05:48
Discrepancy between the measurements (pressure drop; VOF)	blackemperor	FLUENT	2	March 6, 2016 03:40
Running UDF with Supercomputer	roi247	FLUENT	4	October 15, 2015 13:41
Pressure drop across pipe	mnolan93	FLUENT	1	December 15, 2014 07:26

August 29, 2018, 15:27		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,683 Rep Power: 66	Just a sanity check, what is the massflow rate that Fluent outputs? What's your mesh look like? Did you already do a grid dependence study? You'll be surprised how fine a mesh you need for this simple problem. Also just fyi, this type of problem is really good for simulating using the periodic boundary condition rather than velocity inlet and pressure outlet.

August 29, 2018, 20:44		#4
blackmask Senior Member Join Date: Aug 2011 Posts: 421 Blog Entries: 1 Rep Power: 21	You must have made a mistake. If the flow rate Q=1.32 kg/s, then $v_{avg}=\frac{4Q}{\rho \pi D^2} = \frac{132}{\pi 827} m/s\approx 0.0508 m/s$ Which is three times larger than the maximum value you claimed. For parabolic velocity profile, the average velocity should be half the maximum. Did you multiply the result reported by FLUENT by $2\pi$ ?