Видео

couldn't find overset mesh option as shown at 38:18

Hello, You can always apply a user-field function in which you can define the desired motion to the airfoil (the overset motion). Perhaps you can create a field function of a sine wave with respect to time and then assign it to the motion through TOOLS>Motions>(Your Specific Motion-rotation, translation etc...)> click on the motion and assign the function (for rotation you have to click on rotation rate then click the box with three dots in the properties window and for translation you can assign a user-function through clicking on it >translation velocity>click on three dots and assign your function.) Hope I was clear enough. If you have any questions please let me know! Cheers. Good luck!

@@ownershen97 Hello, yes this is extremely helpful. thank you so much for the quick reply! I have done this but then do I apply that motion to the overset region? And I am also struggling to set the physics models, do I have to select specific models so that I am able to see the airfoil moving when I run the simulation? Thanks again nd apologies for taking your time.

@@alejandrabotbol308 Hey Alejandra, You should apply the motion to the overset region. Considering that there is a time-varying parameter in your simulation, you should select a model that is able to show you time dependent motions induced into the flow. Note that ReynoldsAveraged Navier-Stokes(RANS) models are not very applicable in your case so you should either use Laminar or Large Eddy Simulation (LES) for your models (FYI:Laminar is not necessarily a model) with a fine enough mesh to resolve the turbulent scales. Preferably, you should use LES. Also, if you would like to see the motion, you can take 'snapshots' of a scene for each specified time-step through adjusting the settings in scene>update section. Try to set up velocity and pressure scenes and see the changes in the flow. Cheers!

@@ownershen97 Thank you for all of the advice, it is extremely helpful. I will try all of that!

Hi have you solved this problem of airfoil flapping? if yes I need a little help. Thanks

The things you have mentioned are all in the video;yet, here is a summary: 1) Continua: Selection of physics/models you would like to utilize. 2) Region: Defined parts where the physics chosen in Continua will be solved. 3) The fluid region is extracted to specify it. Otherwise there won't be any fluid region. It is better to create CAD in Star CCM+ as imported files can cause problems in which case you have to use surface wrapping/surface repair tools to fix the geometry. Of course, it is your choice to pick your CAD program. Again, everything you mentioned is in the video... You should watch it carefully.

Hey, glad it helped =) The coefficients are 0 since we only want the fluid region to cool the CPUs. Otherwise giving a value to these coefficients will implement a cooling effect which will give wrong results.

Hey, Correct me if I misunderstood you question: You would like to separate the airfoil into two sections; lower and upper surfaces. In that case, in 3D-CAD, before you subtract the airfoil shape from the overset region, you can create a point at the centroid of the airfoils side surface (right-click on the surface --> Reference Geometry-->Point from Entity Center-->Centroid-->Apply. Then, noting the coordinates of this point (copy them by clicking the '...' on properties window), right click 'Features' and proceed to select 'Create a Plane'--> 'By Definition'. Paste thee coordinates you have copied and play with the directions until you create a plane which cuts the airfoil in half. Once you have done that you can now slice the airfoil in half using Slice Booelan operation (Keep both sides while performing the said operation). Make sure you label the upper and lower surfaces correctly. Hope this helps...Cheers

Hello The Velocity IC for fluid physics should be zero for better solution since the pipe does not run along a straight path. For better clarification: If you assign a initially velocity of 5m/s in X direction, every point in the fluid will be 5m/s in +X direction which is wrong in this case because there is a region where the fluid should flow in -X direction (because of the U bend changing the direction of the flow). The problem here is that I gave the Courant number a considerably big value for faster solution time. Check 1:09:35 for Courant number. It should be around 1 for good solution; less than 1 for better yet longer solution time. Hope this helps =)

Hello Vesper, If you changed the fluid medium in Continua, then you won't have to take further steps for the boundary conditions. You can simply give temperature condition into the inlet and outlet boundaries. This can be done by going to the "Regions" tab and proceed to the region with inlet and outlets and change the conditions. In order to monitor the temperature of the fluid near the outlet, you can create a point probe from the derived parts and place it near the outlet. Then, create a maximum (or minimum) report and choose that point you have previously created for the report. Finally, right click the report and say 'monitor and plot'.

@@ownershen97 I understand the inlet, outlet setting you mentioned before. However, as the fluid flows, the temperature of the fluid must be transferred outside or outside, and the remaining temperature must escape, which is the same. My condition is to flow water into the pipe and the inlet temperature of the water is 60℃.I'd like to check the remaining exit temperature after transferring heat, but the exit temperature is still 60℃. p.s. If there is something I am doing wrong, could you please forward the ccm file and check the problem?

​@@vesper_007 I suppose it could be due to failed interface between the fluid and the pipe. I could help you with the file yet I'm having trouble with licensing preventing me from opening SIM files. However, you can try the steps below and see if they work. In the "Regions" tab, select the two boundaries (while holding control) shared by the fluid and the other object and right click one of them while they are both selected and select interface. Or simply try the imprinting process I've shown in the video where the program creates contacts based on imprints made in CAD. Also, do your residuals converge? The solution may not be correct as well due to increased transient heat transfer. You can change from "Steady" to "Unsteady" simulation in continua tab. Make sure all of the continua have the same option selected. Hope this helps

Glad it helped =) Good luck with your thesis!

you can also watch Ozgur Ertunc's video which is in his channel.

Hey I suppose you are referring to the Heat Transfer Coeff. for fluid flow. You should first calculate the fluid's Re number to determine whether its laminar or turbulent. Then, you can use predetermined formulas to calculate Nusselt number. Then, you can proceed to calculate heat transfer coefficient using previously found Nusselt number through a relation which I unfortunately don't have the memory of. . You should take a look at Internal Forced Convection from various books. I recommend Heat and Mass Transfer_Fundamentals and Applications by Yunus A. Cengel and Afshin J. Ghajar (the book used in my uni) however most of the heat transfer books should cover this topic. Hopefully this was helpful. Take care

a trick : you can watch series at flixzone. I've been using them for watching all kinds of movies during the lockdown.

@Matteo Brayden yea, have been watching on flixzone} for since december myself =)

@Matteo Brayden definitely, I've been watching on Flixzone} for since december myself :D

Thank you sir.. means a lot!

Hello The reason is that I assume that there is no forced convection over the surfaces of the system. (However, as you may know, there are fans that actually cool the CPUs down) So, I set the heat transfer coefficient (or convection coeff.) to 0 and let the liquid cooling do its thing and assess how well it cools the system. Now, if you would like to see the effects of convection over the surfaces, then you can set the convection coefficient; however, you may want to make some analytical calculations/simple simulations to find out the coefficient. Then you can set the boundary condition for convection. In addition, lets assume that I know the convection coefficient caused by the liquid inside the pipe. Then I do not need to assign a fluid region and only assign the coefficient at the inner wall of the pipe. This way, you can simplify your simulation and hence decrease computational costs. Hope this helps =)

@@ownershen97 Hello Onur. I researched some thesis about radiators and have to create cfd simulation of a car radiator. If you have any information, Can you inform us briefly about this issue. How should I do this simulation. Thanks for your informing.

@@erenmeliksurmeli9796 Hello I dont have much experience with car radiators but I think it would be fundamentally similar to this simulation. I suppose you can initially draw out the geometry of your radiator, extract the pipes/ducts from it for your fluid to flow through and create interfaces between the pipes and your radiator. There is a tutorial exactly on this topic in star ccm+ help menu (under tutorials) called 'Duel Stream Heat Exchanger: Car Radiator.' If I remember correctly it implements some porous regions in it which maybe of use.

Btw, I am doing a heating system case. There is no fluid/gas in this system except environment. I have a problem that there is only energy on the residual plot (no appearance of continuity and others). Is it reasonable?

@@yajiaoliu5970 Ofcourse since you dont have any fluid there is no need for momentum transfer. This is the ideal case generally since addition of the fluid domain kinda causes problems in terms of multiple time scales (Somewhat the problem in the simulation I have done.)

​@@ownershen97 Thank you for your kind reply!

I could assign a convection boundary layer to assign the environmental temperature but I suppose it wouldn't make a huge difference regarding the temperature distribution throughout the system. Thanks for the heads up though! :)

Hello, I'll try to make more videos for sure;yet, unfortunately I do not have the knowledge to use Python to code for Star CCM+. Maybe in the future when I learn to code on Python =).

Thank you sir

You can indeed transfer CAD files from various programs where Star-CCM+ can distinguish parts provided that you have identified your part previously. Though I'm not sure about regions; I've never worked with NX Siemens (yet I think you meant parts?) In fact I did few of my projects from Inventor 2019 and transferred them to Star-CCM+ no problem (saved as IGES file in Inventor and imported it in Star-CCM+). However, the issue is that there can be problems in the geometry where you may have to use Surface Repair Tools available in Star. Thank you. Always happy to help =)