-
-
February 21, 2024 at 9:25 am
Emil Axelsson
SubscriberTL;DR : I have no idea why the velocity (and streamline) is bending downwards in the beginning of the outlet, as this to me is an unexpected result. Do you have any idea and how to fix this? I've had same phenomenon in various mesh sizes.
Hello fellow Ansys community! I'm fairly new to CFD, and in the process of trying to simulate hydrogen gas-flow through a pipe in 2D.
The problem: The velocity (and streamline) of my results are unexpectedly bending towards the bottom of my pipe, even after rifining the mesh and redoing the calculations - and I have no idea why! I'm wondering if any of you more experienced users have any ideas, as this is unexpected!
The solver:
The pipe inlet is 7 mm diameter and inlet velocity ~ 12 m/s hydrogen gas. Larger pipe is 95 mm diameter. The incoming gas (and inlet pipe temperature) is 150 C. The larger pipe is divided into three parts, two unheated (160 C) and one heated (900 C) part, see picture. In the solver I have used the Pressure-based solver with Laminar flow, and Second order upwind for Density, Momentum and energy. Second order for Pressure.The gas: ideal gas density for hydrogen, constant viscosity, Piecewise polynomial Specific heat, constant Thermal conductivity, constant molecular weight.
The solid (walls): Is assumed to be aluminium with constant density, specific heat and thermal conductivity.The operating conditions are low pressure 5500 Pa defined at center of outlet, and gauge pressure at outlet is 0 Pa. No gravity.
The initialization is; X velocity = 1 m/s, Y velocity = 0 m/s, Temperature = 900 C, Gauge pressure =0 Pa.
All solutions (continuity, momentum x, momentum y, energy) converge to residuals 1e-6.
Thank you in advance,
Emil
-
February 21, 2024 at 9:41 am
Rob
Forum ModeratorYou've got some reading to do :) "Coanda effect", and "pitchfork bifurcation". Welcome to the weird and wonderful world of fluids. Have a careful look around the nozzle region, you want to be reviewing velocity vectors and the pressure.
-
February 21, 2024 at 9:45 am
Emil Axelsson
SubscriberHello Rob!
I'm definetly new to the world of fluids! Thank you for your kind response, I will definetly look into these phenomena!Kind regards,
Emil
-
-
February 21, 2024 at 10:02 am
Rob
Forum ModeratorYou're welcome. Doing a mesh study is the correct first response to seeing something like you've got, asking questions is good.
-
- The topic ‘Unexpected bend in velocity and streamline of laminar flow in straight pipe’ is closed to new replies.
-
6455
-
1906
-
1457
-
1308
-
1022
© 2026 Copyright ANSYS, Inc. All rights reserved.
