|
[Sponsors] |
Non-newtonian polymer flow - incredibly low timestep needed |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
February 15, 2017, 10:05 |
Inlet velocity/flow condition - velocity rises incredibly in the first few steps
|
#1 |
Member
Join Date: Jun 2016
Posts: 31
Rep Power: 10 |
***Title edited***
Hi all, I'm getting to know OpenFOAM Extend and its non-newtonian flow solver properties. I've tested nonNewtonianIcoFoam, pisoFoam and pimpleFoam and they work well as long as the viscosity isn't too high. The nonNewtonianIcoFoam tutorial for example runs fine for all three solvers, but as soon as I use the values for HDPE for the power law or Cross power law (which are very high), the velocity values shoot through the roof and I don't know why. The solver then is forced to use a very small time step in the orders of 1.0e-6 to 1.0e-9 and the velocity values drop after a while, but not to the value I fixed. I need to run it a few seconds, so just waiting is not really an option. I also fumbled with the relaxation factors down to 0.3 for p and U, but it didn't help. The same goes for fixed pressure at inlet and outlet. I tried to initialize the flow field using a newtonian flow (water) and switching to non-newtonian flow after it is developed, but it was all the same. The values for HDPE I found were: CrossPowerLawCoeffs { nu0 nu0 [ 0 2 -1 0 0 0 0 ] 135; //1.282e5 / 950 nuInf nuInf [ 0 2 -1 0 0 0 0 ] 1.0e0; m m [ 0 0 1 0 0 0 0 ] 2.8; n n [ 0 0 0 0 0 0 0 ] 0.227; } and powerLawCoeffs { nuMax nuMax [ 0 2 -1 0 0 0 0 ] 1e2; nuMin nuMin [ 0 2 -1 0 0 0 0 ] 1.0e-1; k k [ 0 2 -1 0 0 0 0 ] 2.0e1; //k/rho!!! 20000/1000 n n [ 0 0 0 0 0 0 0 ] 0.41; } What I could try is to gradually increase the newtonian viscosity until I reach an approximate range of the non-newtonian viscosity and switch, but I don't know if it will work. I'll try to upload my geometry and the whole case later. Does anybody have an idea, why the velocity just explodes (fixed inlet: 0.01 m/s, resulting velocity: around 1.0e8) from switching to the non-newtonian model and what could help my case? Edit: I have noticed that the velocity magnitude in the output text stream rises at the beginning when I use a fixed inlet velocity or flow rate. How come??? Even when I have a look at the tutorial case offsetCylinder from nonNewtonianIcoFoam, the displayed velocity goes from 1 in the first step to over 8000 in the fourth step and then drops again! Similar things happen when I use a newtonian fluid. What is happening there? No wonder the solver is having trouble! Does anyone know what this is? Edit 2: I now found something very interesting: Low Reynolds number question Probably the answer to the phenomenon. Last edited by tdof; February 19, 2017 at 08:39. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Low Speed Flow Boundary Conditions | Paul Bateman | FLUENT | 2 | January 10, 2013 13:01 |
pump prevalence at low mass flow rates | Caterina | Main CFD Forum | 0 | January 8, 2011 16:05 |
urgent help needed with 2d compressible flow | James | FLUENT | 2 | June 20, 2007 04:22 |
Flow visualization vs. Calculated flow patterns | Francisco Saldarriaga | Main CFD Forum | 1 | August 2, 1999 23:18 |
CFD code for very low Re Flow | Raj | Main CFD Forum | 2 | May 31, 1999 22:06 |