|
[Sponsors] |
June 2, 2009, 10:23 |
3D flow in a pipe
|
#1 |
New Member
Ravi Kapoor
Join Date: May 2009
Posts: 6
Rep Power: 17 |
Hi,
I want to analyse velocity profile of fully developed(laminar) single phase flow in a pipe. For that I have created a pipe in GAMBIT and exported in Fluent with the inlet and outlet conditions as pressure conditions. We know the analytical solution in a pipe and it shud be parabolic at the outlet(when flow is fully developed) When I am working with low reynold's number, the results are pretty much accurate(RMS error=1.5%) but when I work with high Re # like 1000, I get an error of around 20%. For example, for Re=500, the entrance length required is 3m (D=0.1), so I am woking on L=8m so that fluid is fully developed at the outlet and I have accurate results bt still I am getting error for high Re#. The error increasses as Re# increases. Moreover, Can I create different interval spacing in axial direction and on the surface nodes in GAMBIT. Like, I have dia=0.1 and L=8m, so if I use Interval count of 60, then it will be fine for the face while coarser for the axial direction whreas if I use interval size of 0.001, it will be okay for the face but too much fine for the axial direction. So, Please help. Thanks |
|
Tags |
3d geometry, laminar, pipe, single phase |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Simulating fully developed flow in a pipe | Tim Donohue | CFX | 1 | November 20, 2007 21:24 |
some weird results about a simple pipe flow: | Ning | FLUENT | 2 | March 11, 2007 14:23 |
Flow in a Pipe having protrusion inside it | Hari Analakkat | Main CFD Forum | 0 | January 25, 2006 07:22 |
Flow through circular curved stationary pipe | R.Sripriya | Main CFD Forum | 0 | May 13, 2005 07:07 |
Pipe flow | John Grimm | Main CFD Forum | 8 | March 11, 2002 13:15 |