ABAQUS CFD verification with Darcy Pressure equation

[Chrissy123]

Hello - can anybody help me on this please ? I am modelling a simple cylinder with laminar flow in ABAQUS CFD. I am using Reynolds number of 25 to simulate laminar flow with an inlet velocity of 1m/s using a fluid with density of 1 and kinematic viscosity of 0.012. I have worked out the length of the pipe as 0.375 to allow for full flow so just to add extra length I've gone to 0.562 for the length of pipe. So far so good - I've done a mesh convergence for the velocity and have a good velocity profile to display laminar flow with a max velocity in the centre as 1.98 m/s. I am now trying to do some verification regarding the pressure drop so I have looked at the contours for pressure and my CFD is giving me a value of 6.14; I am using Darcys equation of deltaP = f*(L/D)*(rho*v^2)/2 to calculate analytical - using 64/25 (i.e. 64/Re) for f, L=0.562,D=0.3 (hydraulic diameter), rho and v both equal to 1, which gives a value of 2.398. So analytical is very different to the CFD - what am I doing wrong ? Any help would be greatly appreciated. I have tried making the pipe longer but this just gives the same proportionality for the difference in the results - thank

 

[Dave442]

I might be wrong but I think you're meant to use flowrate in m3/s rather than velocity in m/s for v.

Also, I think Darcy equation applies only to fully-developed laminar flow and doesn't take into account that you have an entrance length where the flow develops.

You could use an equation to define a fully developed velocity profile at the inlet to avoid this. Or you could only consider pressure loss across the length of pipe where the flow is in fact fully developed.

 

[Dave442]

Maybe I am thinking of a slightly different form of the equation:

dP = L*(128/pi)*[(mu*Q)/D^4]

I guess just double check you're using an appropriate mean velocity. Rather than the value you define at the inlet. Since it varies as a function of tube radius.

 

[Chrissy123]

Thanks all - I have found the solution - Darcy only applies to fully developed flow so I increased my duct size considerably and now I'm getting good agreement with analytical values - thanks Dave - I've effectively changed it to look at only fully developed flow -

 

[Dave442]

i think if you can just neglect the entrance length, which can be calculated.

you should be able to calculate dP between the start/end of the fully-developed region.

then use that for your validation - saves having to extend your domain.

glad you got sorted anyhow!