Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Laminar Flow Entrance Region/Issues

Please login with a confirmed email address before reporting spam

I am attempting to model a long square flow channel with laminar flow of air. I was wondering though if anyone knows exactly how COMSOL computes the entrance region for laminar flow, or where I can find this information?

4 Replies Last Post 14 janv. 2017, 17:27 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 juil. 2014, 08:21 UTC−4
Hi Brian,

I am not entirely sure what you like to know, but I assume you are asking about the inlet boundary condition with a fixed velocity at the entrance. In this case COMSOL sets the inward velocity of the fluid to the given value. In case of a no-slip boundary condition at the walls this will produce a flow profile which is incorrect near the entrance. There are two workarounds.
1) You could define a volume in front of the channel which allows the correct profile to form and cannot be used for further calculations.
2) You could use a pressure difference as source of your flow. This produces the correct profile and is much more elegant.

By the way, are you sure you would like to use the laminar flow model? Air does usually not have a laminar behaviour. This is especially important when dealing with heat transfers.

Best
Daniel
Hi Brian, I am not entirely sure what you like to know, but I assume you are asking about the inlet boundary condition with a fixed velocity at the entrance. In this case COMSOL sets the inward velocity of the fluid to the given value. In case of a no-slip boundary condition at the walls this will produce a flow profile which is incorrect near the entrance. There are two workarounds. 1) You could define a volume in front of the channel which allows the correct profile to form and cannot be used for further calculations. 2) You could use a pressure difference as source of your flow. This produces the correct profile and is much more elegant. By the way, are you sure you would like to use the laminar flow model? Air does usually not have a laminar behaviour. This is especially important when dealing with heat transfers. Best Daniel

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 22 juil. 2014, 17:08 UTC−4
In my case, I am using a known mass flow rate to designate the flow of air in the channel, not velocity or pressure difference. The channel is very small, I probably should have specified that, 2 mm x 1.25mm (not exactly a square more of a rectangle) and 10 cm in length. At this size and with the mass flow rates I am using it will almost always be laminar flow. I want to specify that air comes into the channel at a given temperature, that there is heat generation along the flow channel, and resolve what the temperature profile of the air and flow channel will look like at steady state. The flow becomes fully developed very quickly, and during this development I know the convection coefficient will be very high.

However, when I place a face at the inlet to the flow channel and specify its normal mass flow rate and temperature, I get results which cannot be correct. That is, the heat generated and the heat absorbed by the air do not match. I assumed this was something to do with the inlet face, such that it physically touches the flow channel and maybe heat conducts out of the face? and this lost heat fills the energy balance. To work around this I do like you had said and place a volume in front of the entrance of the flow channel. I made it only 0.3mm distance away from the inlet and am getting better results from my model.

My question was primarily how does COMSOL compute the convection? I have a resource to which I am comparing these results that uses empirically derived convection coefficients for rectangular flow channels. I have been told that COMSOL simply solves the navier stokes equations to compute convection, but I would like to verify this and possibly figure out why I must put an entrance volume in front of the flow channel to get reasonable results. Thanks for the help!!
In my case, I am using a known mass flow rate to designate the flow of air in the channel, not velocity or pressure difference. The channel is very small, I probably should have specified that, 2 mm x 1.25mm (not exactly a square more of a rectangle) and 10 cm in length. At this size and with the mass flow rates I am using it will almost always be laminar flow. I want to specify that air comes into the channel at a given temperature, that there is heat generation along the flow channel, and resolve what the temperature profile of the air and flow channel will look like at steady state. The flow becomes fully developed very quickly, and during this development I know the convection coefficient will be very high. However, when I place a face at the inlet to the flow channel and specify its normal mass flow rate and temperature, I get results which cannot be correct. That is, the heat generated and the heat absorbed by the air do not match. I assumed this was something to do with the inlet face, such that it physically touches the flow channel and maybe heat conducts out of the face? and this lost heat fills the energy balance. To work around this I do like you had said and place a volume in front of the entrance of the flow channel. I made it only 0.3mm distance away from the inlet and am getting better results from my model. My question was primarily how does COMSOL compute the convection? I have a resource to which I am comparing these results that uses empirically derived convection coefficients for rectangular flow channels. I have been told that COMSOL simply solves the navier stokes equations to compute convection, but I would like to verify this and possibly figure out why I must put an entrance volume in front of the flow channel to get reasonable results. Thanks for the help!!

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 juil. 2014, 05:19 UTC−4
You can view the equation solved in the equation section of the physics node. The laminar flow model solves a navier-stokes equation.
You can view the equation solved in the equation section of the physics node. The laminar flow model solves a navier-stokes equation.

Jesudoss Asirvatham Jeyaraj

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 14 janv. 2017, 17:27 UTC−5
Hi,

I am simulating fluid flow through pipe in Comsol 5.2a using 3d modeling with laminar flow. I have inlet and out let. According to manual, if I give velocity at inlet I can give pressure outlet and vice versa. But if i follow like that it shows as inconsistent values and not solving and gives error statement. It does not show error statement when I give pressure at both inlet and out let. But the results, the velocity plots are not correct. Can anybody give me correctly working sample data at inlet and outlet ?
Hi, I am simulating fluid flow through pipe in Comsol 5.2a using 3d modeling with laminar flow. I have inlet and out let. According to manual, if I give velocity at inlet I can give pressure outlet and vice versa. But if i follow like that it shows as inconsistent values and not solving and gives error statement. It does not show error statement when I give pressure at both inlet and out let. But the results, the velocity plots are not correct. Can anybody give me correctly working sample data at inlet and outlet ?

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.