Видео

I'm happy to hear this. I'll try to make a video with dynamic mesh soon.

Your welcome

Thank you for your comment. First, suppose you are using solid-solid phase change material. In that case, you can still use the solidification and melting model in Ansys Fluent because velocity is zero in all directions when you have solid in the zone. So, you will have only conduction in energy equations and you may select 0 for mushy constant as well. Even you can disable the momentum equation for your simulation. However, if you want to have only conduction and use it with the UDF, I suggest you utilize User-Defined Scalars (UDS). In UDS, you have unsteady, convection, defusion, and source terms. You can solve the equation in your domain and then use the parameter, let's say temperature, in the UDF. I hope this helps you to complete your simulation. I appreciate it if you share my channel with your friends to subscribe.

I am so happy to hear this from you. Sure, I will provide more videos regarding PCM. However, the next one will be about generating a multizone mesh that can be useable for PCM simulation.

You can see the equations of entropy generation in the below article. After creating the equation in Fluent in the Custom Field Function Calculator dialog box, you can calculate the volume integral of entropy generation in your domain. doi.org/10.1007/s00231-019-02615-z

You can see the equations of entropy generation in the below article. After creating the equation in Fluent in the Custom Field Function Calculator dialog box, you can calculate the volume integral of entropy generation in your domain. doi.org/10.1007/s00231-019-02615-z For derivative of temperature, you need to active it as I showed in below tutorial. ruclips.net/video/9Cf3854WTjw/видео.htmlsi=N5A6Sn6WBQdV68dE

No, it's better to calculate the Nu number at each specific length of the tube and then calculate the average of all Nu number values using the methods to get the average value. The average method: www.cuemath.com/trapezoidal-rule-formula/

In ansys, we do not have a length average; so when you use area average, it is supposed to give you the average length of elements. To test my claim, please calculate the area of the line around the pipe, if it gives you the circumference of the tube, so, this method works.

Thank you! we are glad you found it helpful

If this doesn’t work well for you you can add the following command to your tecplot.cfg file. $!COMPATIBILITY USEV10TEXTFORMATTING = FALSE If FALSE this setting allows Tecplot to display text special characters created with older Tecplot versions without automatically converting the text to the new formatting. In Tecplot 9 a single character could be changed by adding the following identifiers in front of that character. The identifier only applies to the character immediately following. ` (back quote) Greek ~ (tilde) math @ (at symbol) User-defined _ (underscore) subscript ^ (caret) superscript

Hi, I guess you have one of the following problems: You may be using the student version, you need to have installed Visual Studio, or you may also need to change the mpi setup in the launch window in Fluent. The problem seems to appear from parallel processing.

@@akamcae you are right...... how can i fix it ? udf?

@@sss-bc6dz Please try one of the following steps: 1) First read your case and data, run one interaction, change the setup like the video here, and then run your simulation one iteration more before getting your data. 2) if your simulation is steady and you want to get data at the end of your simulation, you can read the case and data with a single process instead of parallel processing. Just be aware that before calculating the data by custom field function calculator, just do one interaction, otherwise, you will get a zero value. 3) try to install a newer version of Visual Studio. 4) if none of the above solutions worked, you need to create a UDF file, You can use DEFINE_ON_DEMAND. Good luck

@@akamcae thank you for your reply

We are happy to hear that, thank you.

Sorry, I do not know how to calculate dT-dt. I suggest you get the temperature from a specific point and then provide a function by overfitting using Exel then calculate the derivative of it. Also, you can use numerical methods for calculating a derivative of some data.

For calculating the local Nusselt number at a specific point, instead of taking Tw,avg, you need to calculate the Tw at the specific point. The same procedure for local heat transfer. For the local thermal boundary layer, make a line inside the domain that is normal to the wall, then report the temperature, I guess you can see the thermal boundary layer.

You can calculate the average Nu number along the tube by using the trapezoidal rule. This is a simple method. OR; If you have access to Matlab software, you can use trapz( X , Y ) and then divide it by the length of the tube. See Eq. 14 in this paper doi.org/10.1007/s00231-018-2345-z.

You can get the results from Fluent and then use Excel to plot an x-y plot.

My pleasure

Hi, You should make a cross-section in your domain and then use Mass-Weighted Average to calculate the temperature on the cross-section surface. That's where you will get the bulk temperature in a specific cross-section. For wall temperature, you need to use Area-Weighted Average over the line or surface of the geometry. Please subscribe to my channel, I will provide a video about calculating these two parameters within the next week.

@@akamcae Thank you very much for your information.

@@ammarlaichi8474 Hi, you can find the video regarding calculating the Nusselt number in the below link. Thanks ruclips.net/video/2StpgKD2ljU/видео.htmlsi=4lFfz-Vo7ux_HMUh

Hi. Yes, it is possible. You can see the equations of entropy generation in the below article. After creating the equation in Fluent, you can calculate the volume integral of entropy generation in your domain. doi.org/10.1007/s00231-019-02615-z

I have not worked on rotating impeller. You can search and find an article that explains the equation of entropy production in a rotating impeller. When you find the equation. You can transfer the equation as a UDF code or as a custom field function. However, I think, the equation should be general for all applications, which is explained in the Bejan book. When you find the equation, I think you can calculate the equation using a volume integral in reporting section of Fluent software. You can see the equation of entropy generation in the below link. This is a published article by me. link.springer.com/article/10.1007/s00231-019-02615-z

Hi, please send an email to me. I will send you a file. My email address: najafi.fme@gmail.com

You're welcome!

You're welcome!

Please I need how we calculate nusselt average pleasssse

@@papahabibi1402 Hi, you can send me an email (najafi.fme@gmail.com). I can not tell you in the comment. We can have an online meeting and I will explain it to you.

Hi, first you should create a custom field function with your definition (dT/dy....) then calculate the value of your function from the wall, which the location is y=0. To report the value from that location, you can create a line or point by Fluent on the surface (wall).

@@akamcae I want to find dT/dY not dT-dY

@@maniksinghsambyal550 It is dT/dY but when you use the Fluent, it shows like dT-dY.

Hi, please send me an email to send an article for you. My email: najafi.fme@gmail.com

Hi, there is a Macro called DEFINE_INIT (initializes variables) in UDF that the Dialog Box of this Macro can be activated in User-Defined Function Hooks. You can see the definition of this UDF in the ANSYS FLUENT UDF Manual.

For normal derivative, you should use the UDF code. I have not calculated the normal derivative but you can see some samples of UDF code about the normal derivative in the UDF Manual of ANSYS Fluent and CFD Online website.

Sorry, for not getting back to you sooner. You can report the Time-Averaged Wall Shear Stress as below: 1) For steady simulation: Go to Report> Surface Integrals> Report type (Area-Weighted Average)> Field Variable (Wall Fluxes...)> select Wall Shear Stress> then Chose the boundary that you want to report the Time-Averaged Wall Shear Stress> Compute. 2) For translate simulation: You can use UDM Macros for saving you data every time step size and calculate the average time with other software, or use the Tecplot 360 CFD post-processing software, which you can import your case and data to it.

Hi, In this method, the value illustrates without units because you make a custom field function which it has not a defined unit in Fluent. To unravel the problem of zero value, you should calculate your run (your all equations or just your energy equation) before you calculate the average value of dT/d(x,y,z) for a particular part of the fluid domain. One iteration for calculating the equations is enough after you read your Case and Data in Fluent. If you need more information regarding this issue, please do not hesitate to reply to me.

​@@akamcae You are right. I first ran a calculation and after convergence is achieved, I followed your steps and didn't rerun calculation. Thank you and best regards!

@@josipbatista3570 Your welcome. In addition, whenever you want to calculate your parameter with the custom field function in your previous run, do the below steps and save your time. 1- Read your Case and Data files. 2- Go to Solution> Solution Controls> Equations 3- Active the energy equation and inactive others equations 4- Press OK 5- Run your solver for one iteration 6- Calculate your parameter with the custom field function Good luck

Your welcome. We can't calculate the second derivatives in the custom field function calculator of the ANSYS FLUENT software (maybe I don’t know), but it will be calculated with the UDF. I think these links (in below) may help you. *www.cfd-online.com/Forums/fluent-udf/37146-how-build-second-derivative-udf.html *www.cfd-online.com/Forums/fluent-udf/137277-second-derivation-temperature.html

@@akamcae Thank you... I wrote a UDF and It seems working: :-) # include "udf.h" DEFINE_ON_DEMAND(store_gradient) { Domain *domain; cell_t c; Thread *t; domain=Get_Domain(1); thread_loop_c (t,domain) { begin_c_loop (c,t) { C_UDSI(c,t,0) = C_T_G(c,t)[1]; C_UDMI(c,t,0) = C_UDSI_G(c,t,0)[1]; } end_c_loop (c,t) } } Ps., For better refinement in gradients, one also can use the reconstruction component of gradients of any variable...

@@Roilbb gradient is a vector. How can you store it in a scaler?

@@waqasmuneer7951 It is magnitude which is stored in UDS. You can also store the components of a vector in UDS as well.

@@Roilbb It is the vector that has been stored in scaler and it is a magnitude of the sclalar which has been stored in UDM. Is your code working fine?

Unfortunately, I don't know, but it might be possible with the UDF definition in the Ansys Fluent.

@@akamcae thanks a lot sir please do you have any UDF documents support ! i really appreciate ur help tnx

@@newyorkbrooklyn987 No, I am sorry.

For calculating the mean of parameters, I think that you can write your custom field function then calculate its as the Area-weighted Average (for example) in reports section of Fluent.