[Nek5000-users] Pressure drop in simple cylinder flow

[nek5000-users at lists.mcs.anl.gov]

Hello Neks,

As a first excercise learning to mesh using the tools provided in Nek5000,
I ran a simple laminar cylinder flow simulation (geometry obtained by
extending the geometry of cyl3dr.rea from /examples/expansion) and clearly
see a fully-developed flow at the outlet. The maximum velocity is, as
expected, exactly twice the uniform inlet velocity, and I see a parabolic
velocity distribution along the radial direction. However, the pressure
drop is not in line with the  theory.

For example, from Cengel, deltaP=8nuLV/(R*R). nu is the kinematic (equal to
dynamic in this case) viscosity, L is the length of the cylinder = 20, V is
the average velocity = 1, and R is the radius of the cylinder = 1.

Also, Re=10=V(2R)/nu, giving nu=(1/5) leading to
deltaP=(8)(1/5)(20)(1)/(1*1)=32. But, the pressure drop (falls linearly, in
line with theory) in my simulation is only 16. I increased the cylinder
length by two times, and again the pressure drop is half of what is
expected. Can anyone tell me if there is something I am not accounting for?
I am new to CFD and this is my first ever problem. I am pasting my input
parameters below.


Thanks,
Ajay
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170814/38bdd195/attachment.html>