|September 17, 2012, 11:31||
Help with boundary settings, along with other questions
Join Date: Sep 2012
Posts: 3Rep Power: 5
First of all, I'm using Solidworks Flow Simulation (SW Premium 2011 SP2)
I'm simulating a simple blower fed air knife. The blower that is going to be used is rated to 300 mbar and 4.6 m3/min, so I've set my inlet boundary to "Static pressure" 131325 Pa (atmospheric pressure plus 30 kpa), and my outlet boundary to "outlet volume flow" 0.0766 m3/s.
Is that the correct way to specify the inlet- and outlet boundaries in this case?
My goals with this simulation (I do of course realize that I have to run more than one simulation) are:
1: Find out how much of a difference in flow distribution i get between not using a diffuser between the plenum chamber and the outlet and by using said diffuser. Without the diffuser, there is really no "plenum chamber", as the whole thing would act as a plenum chamber (i hope). Can I get it "good enough" (which would be something like less than 20% difference in air velocity across the outlet slot)?
2: Find an optimal width of the outlet slot in regards to evenness and outlet velocity.
For the second goal, I have a question:
Is there any way to define an "evenness" goal (even velocity, pressure, e.t.c)? Can I for example define an arbitrary number of points along the slot, and somehow get a goal to show the difference in for example velocity across those points? Or can I select the surface of the outlet lid and use that in any way? The reason I want to be able to do this is that I want to run a parametric study that converges against "evenness".
The model in the picture below is "empty", i.e. there is only one cavity. The "tube" does not go all the way through. The entry is the round lid to the left. The exit is the narrow slot (I've got a hidden lid). Although, in the version with a plenum and diffuser, I've let the tube go all the way through, then the part of the tube that is inside the plates that form the horn is perforated with a bunch of 5 mm diameter holes, to form a diffuser. If it turns out the diffuser is needed (which I think), I want to solve "evenness of flow through the slot" against the size of the diffuser holes.
|September 18, 2012, 08:07||
Join Date: Jul 2009
Posts: 416Rep Power: 13
not quite the best way to do it.
For such an application it is important to know what you are trying to find out. If it is the flow just in the nozzle component you can be ok with just that component but the flow field inside the nozze can change when you consider the surrounding area too. The outflow, distance to the obstacle you migh want to dry, or what ever the application is, can influence the flow in the nozzle due to the back pressure you get compared to a free stream into the environment. But you can start small of course, but remember the result are only as accurate as the boundary conditions. So in the end you will want to model the reality as close as possible.
In general you should always use mass flow or volume flow for the inlet in such a case as this condition at the outlet would force the flow to a certain flow profile at the outlet which is most likely not the profile as it would look in reality. So switch the boundary condition sides.
For your first task the component itself should give you a pretty good idea already with the changed boundary conditions
For the second task you probably should consider using an external analysis that is closer to the reality, maybe even with the obstacle if there is one. The outlet profile and velocity can change due to that change in setup.
There is a goal used in filter material such as in automotive catalytic converters. It is called uniformity index. You can lookup the definition behind it in the parameter in the online help.
A general way to reach such an eveness would be for example to use a parameter goal for minimum, bulk average and maximum and you can define an equation goal that calculates the difference between the minimum and maximum and you can try to reach a very low difference.
In general such a parametric study can ger very complicated as there would probably be a better parameter to optimize or a better design variation than the nozzle width. Certainly the outlet surface is the goal for the eveness and you should use a goal on this surface to check the eveness. But it is not easy. There is a way to do an optimization but it is not that easy and more of a manual work. To complicated to just post it here without any visuals and a lot of typing.
Often such deffuser holes are the way it is done. However the optimization towards velocity at the outlet surface with a goal for the minimum and maximum value will result in a 0 for minimum as there is always 0 at the wall of the geometry for the velocity. In that case I would use an external analysis with pressure goal at the deactivated lid of the nozzle. But it is also not very good for the parametric study as you will have to know the minimum value you are looking for.
I would suggest the better way is to run several calculations with a range of hole diameters and compare the results to get a trend to where you can go with it.
I hope this helps, but it is not an easy task to help in a forum. I would suggest to call your support for any suggestions. They should be open to help you in such a problem and try to find a solution with you.
|Thread||Thread Starter||Forum||Replies||Last Post|
|2 questions about boundary conditions||mateusps||OpenFOAM||8||August 6, 2014 02:29|
|inlet velocity boundary condition||murali||CFX||5||August 3, 2012 08:56|
|Questions about Boundary Layer Thickness and Turbulence Models||famerfamer||STAR-CCM+||3||July 12, 2012 09:47|
|Default settings for 3D boundary layer-Gambit 2.2||Ramine||FLUENT||1||July 31, 2006 12:36|
|Trimmed cell and embedded refinement mesh conversion issues||michele||OpenFOAM Other Meshers: ICEM, Star, Ansys, Pointwise, GridPro, Ansa, ...||2||July 15, 2005 04:15|