CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   ChannelOodles Correcting driving force for a constant mass flow rate (

maka September 20, 2005 10:39

channelOodles: I'm trying t

I'm trying to edit the channelOodles to use it for channelDNS. I found some part of the code that is about correcting driving force, this part does not exist in icoFoam, which means: it is either included because of the channel or because of LES? I'm a ware that the driving force in the channel enters in the non-periodic part of pressure gradient, but I do not know about such correction. Can any one explain why we need this part in the channel code or give any reference? Many thanks!

// Correct driving force for a constant mass flow rate

// Extract the velocity in the flow direction
dimensionedScalar magUbarStar =
(flowDirection & U)().weightedAverage(mesh.V());

// Calculate the pressure gradient increment needed to
// adjust the average flow-rate to the correct value
dimensionedScalar gragPplus =
(magUbar - magUbarStar)/rUA.weightedAverage(mesh.V());

U += flowDirection*rUA*gragPplus;

gradP += gragPplus;

Info<< "Uncorrected Ubar = " << magUbarStar.value() << tab
<< "pressure gradient = " << gradP.value() << endl;


mattijs September 20, 2005 12:58

All to do with the 'channel' b
All to do with the 'channel' bit, not with LES.

One specifies an average flow-rate (Ubar in constant/transportProperties) and the pressure gradient is (hopefully slightly) adjusted to keep the wanted flow-rate.

maka September 21, 2005 04:56

In my case, I'm interested
In my case,

I'm interested to keep the dimensionless driving pressure gradient (non-periodic part) (scaled by u_tau and half channel width) to be -1 while specifying Re_tau for the simulation. The Ubar can come as a result of the simulation. I think in this case, I do not need the correction step, am I right? On the other hand I was thinking to add this deriving pressure gradient as a source term in Su, what do you think? Many thanks for your help.

Best regards,

eugene September 21, 2005 05:18

Yes, you are correct, I have d
Yes, you are correct, I have done this myself.

The pressure gradient was excluded from the momentum matrix for a reason, I just cant remember why. It should be simple to try both options and see if there is a difference.

All times are GMT -4. The time now is 16:59.