Hi Cyprien, thank you very much for this video, it help me a lot in my current problem i told you. I hope i create the rest too :) cordial greetings, stay healthy Theo
Yes, that’s because the time function I defined goes too fast to 100% of the value (10 sec, so literally in one step), thus it’s exactly the same as defining a constant source at the max value. I guess the purpose here was mainly to teach how to do it ;-)
Hey Cyprien, your email calling out to this video helped a lot I was just looking for means to perform transient heat exchange and this is a powerfull tool. I'll try to apply this step by step guide in a more complex model (with more bodies, lots of contacts) and let's see if it works. Thanks a lot.
Hi Cyprien, thanks for your great videos. I have been using commercial FEM softwares for years and I'm discovering SalomeMeca/CodeAster since a few weeks... what a good surprise! Transient thermal studies works fine for me. I need to fix something with steady state analysis (initial state and increment field are marked as "required" with red asterisk and can not be empty).... maybe a problem with the distribution I'm using (Salomé-méca 9.10 for Windows).
Hi Cyprien, I'm new to this software and I find a lot of material done by you. Thank you for this incredible service! I'm following this tutorial and trying to replicate it for another analysis. I have to simulate the temperature propagation through an aluminum part, from an electric arc as a heat source (GTAW Welding). Sadly I don't encounter anyone that has done this so far. I want to ask particulary about the electric arc heat source, it should be done through radiation and not "contact" as you have on this tutorial. Can you tell me which function should I select on "BC and Load" to simulate this case? Or what should be the best approach as to simulate a moving the electric arc around an aluminum tube. Thank you very much!
I appreciate your answer Cyprian! I found that is possible to simulate welding on Salome Meca, even multipass welding (!!). On their website there is a full document about it. The challenge is to have a moving heat source along a meshed weld bead. I believe that should be done through a function. It's tricky though as I'm very new to this software. If you are doing any tutorial in the near future, please consider a something similar to this one (transient thermal analysis with a moving heat source) since the website forum is full of questions and no tutorial can be found on the internet. Many thanks again! Nicolae
Hi, Nicolae. I'm Mikhail from Minsk, Belarus. I'm interesting in welding processes. Please, tell me on what website have you got a full documentation about simulation welding on Salome Meca? Thanks in advance!
You mean doing a CFD simulation? To do forced convection, you generally have to model the fluid and fans, etc... you’ll have to use Code_Saturne to do it and that would be a whole new simulation.
Will the Code_Saturne be able to simulate both conduction and convection? Is it suitable for estimating temperature of electronics parts, in some assemblies, in enclosure? Or it's impossible or requires being expert in physics and programming? This example seems to be easy to use and good for things like simulating temperature rise under pulse power loads but not sufficient for estimating steady state temperatures in real life situations, as eg. if the part were in enclosure and convection would be hard to guess.
@@asdfzxcv6268 yes, code_saturne can do that. You have do do a study coupled with Syrthes( which is another EDF code) . I think there is a PDF tutorial on the code-saturne website to show how to set that up. You do not need to be a programming genius to follow that, but you need to know what you are doing and have some basics of Linux Ubuntu. Good luck!
Hi Cyprien! I must write again because I saw something in the tutorial that I don't quite understand. When doing the transient analysis, you must define the initial temperature through "ETAT_INIT". What I dont understand is why you put "0" for temperature. What I learned is that temperature units are in Kelvin. Are you saying that your initial temperatureis 0 K (-273 ºC)? Or maybe I'm assuming something wrong.. Thank you! Best regards
Temperature is always calculated as a difference, so the unit doesn’t matter. Etat_init is the value of the temperature field at the initial moment, so if I set it to zero, it means I am calculating in degree C and initial state is zero degree C.
@@Feaforall I believe then that i've been having troubles with the temperasture units. I assumed so far that temperature was displayed in Kelvin, given that I'm using SI units (m, kg, s, K). Specially with para vis, you read the temperature as Celsius. How can I know if its either Celsius or Kelvin?
You can use the convection load with an ambient temperature which depends on a function. The convection is what approximate the contact between your model and the external air, but this needs the approximation that the air temperature around is constant all over the model. If this is not the case, you’ll have to do a CFD heat-structure coupled simulation (Code-aster can’t do that)
@@Wikimika1411 You're welcome. In addition, i recommend you to read this article on the blog, and especially the part about convection: feaforall.com/heat-transfer/ That should help you to understand better how heat transfer works
Hi Cyprien! Another great tutorial! You used SOURCE to apply thermal load on a group of elements. Its possible to apply to a group of nodes? SOURCE only have "Group of Elements" as a option. In Ansys i can use the "F,,heat,value" to apply heat loads directly on nodes. Its possible to do this in Aster? Thanks!!
@@Feaforall actually there is: # Create a group after selecting the nodes local_nodes=welding_mesh.MakeGroupByIds("local_nodes",SMESH.NODE,selectednodeIDs) # Create a group of elements which contain the nodes local_elements= welding_mesh.CreateDimGroup( [ local_nodes], SMESH.VOLUME, 'local_elements', SMESH.ALL_NODES, 0) but i really wanted to apply directly to nodes..... :/
@@rodrigofarias900 According to the doc, SOURCE has to be applied to elements, not nodes. There doesn't seem to be an option for nodes: www.code-aster.org/V2/doc/v13/en/man_u/u4/u4.44.02.pdf But maybe I am just not aware of it... Try to ask on the Code_aster Forum, they may have more answers there!
Excellent tutorial. I am just beginning to use Salome Meca and this is exactly what I needed. Many thanks for your efforts.
Hi Cyprien,
thank you very much for this video, it help me a lot in my current problem i told you. I hope i create the rest too :)
cordial greetings, stay healthy
Theo
Thanks Cyprien. One comment though- I got the same result even when I did not define the source function.
Yes, that’s because the time function I defined goes too fast to 100% of the value (10 sec, so literally in one step), thus it’s exactly the same as defining a constant source at the max value. I guess the purpose here was mainly to teach how to do it ;-)
Hey Cyprien, your email calling out to this video helped a lot I was just looking for means to perform transient heat exchange and this is a powerfull tool. I'll try to apply this step by step guide in a more complex model (with more bodies, lots of contacts) and let's see if it works. Thanks a lot.
Very glad to hear that Lucas! Let me know how it goes ;-)
Perfect , execellent tutorial !
Is it possible to apply current to modeling? If I'd like to analyze temperature of joule heating? (Copper busbar's temperature flowing current)
Hi Cyprien, thanks for your great videos. I have been using commercial FEM softwares for years and I'm discovering SalomeMeca/CodeAster since a few weeks... what a good surprise! Transient thermal studies works fine for me. I need to fix something with steady state analysis (initial state and increment field are marked as "required" with red asterisk and can not be empty).... maybe a problem with the distribution I'm using (Salomé-méca 9.10 for Windows).
Hi Cyprien, Thank you for nice tutorials! Could you upload some additional videos about joule heating simulation of busbar by open source software?
It helps a lot, as always. Thanks
Hi Cyprien, I'm new to this software and I find a lot of material done by you. Thank you for this incredible service!
I'm following this tutorial and trying to replicate it for another analysis.
I have to simulate the temperature propagation through an aluminum part, from an electric arc as a heat source (GTAW Welding). Sadly I don't encounter anyone that has done this so far. I want to ask particulary about the electric arc heat source, it should be done through radiation and not "contact" as you have on this tutorial. Can you tell me which function should I select on "BC and Load" to simulate this case? Or what should be the best approach as to simulate a moving the electric arc around an aluminum tube. Thank you very much!
I think code_aster cannot do radiation: code-aster.org/forum2/viewtopic.php?id=24342
There is another code called Syrthes which supposedly can do it but I don’t know how much is available about that online...
I appreciate your answer Cyprian! I found that is possible to simulate welding on Salome Meca, even multipass welding (!!). On their website there is a full document about it. The challenge is to have a moving heat source along a meshed weld bead. I believe that should be done through a function. It's tricky though as I'm very new to this software.
If you are doing any tutorial in the near future, please consider a something similar to this one (transient thermal analysis with a moving heat source) since the website forum is full of questions and no tutorial can be found on the internet.
Many thanks again!
Nicolae
Hi, Nicolae. I'm Mikhail from Minsk, Belarus. I'm interesting in welding processes. Please, tell me on what website have you got a full documentation about simulation welding on Salome Meca? Thanks in advance!
Is it possible and/or difficult to add forced convection for this simulation?
You mean doing a CFD simulation? To do forced convection, you generally have to model the fluid and fans, etc... you’ll have to use Code_Saturne to do it and that would be a whole new simulation.
Will the Code_Saturne be able to simulate both conduction and convection? Is it suitable for estimating temperature of electronics parts, in some assemblies, in enclosure? Or it's impossible or requires being expert in physics and programming?
This example seems to be easy to use and good for things like simulating temperature rise under pulse power loads but not sufficient for estimating steady state temperatures in real life situations, as eg. if the part were in enclosure and convection would be hard to guess.
@@asdfzxcv6268 yes, code_saturne can do that. You have do do a study coupled with Syrthes( which is another EDF code) . I think there is a PDF tutorial on the code-saturne website to show how to set that up. You do not need to be a programming genius to follow that, but you need to know what you are doing and have some basics of Linux Ubuntu. Good luck!
Hi Cyprien! I must write again because I saw something in the tutorial that I don't quite understand. When doing the transient analysis, you must define the initial temperature through "ETAT_INIT". What I dont understand is why you put "0" for temperature. What I learned is that temperature units are in Kelvin. Are you saying that your initial temperatureis 0 K (-273 ºC)? Or maybe I'm assuming something wrong.. Thank you!
Best regards
Temperature is always calculated as a difference, so the unit doesn’t matter. Etat_init is the value of the temperature field at the initial moment, so if I set it to zero, it means I am calculating in degree C and initial state is zero degree C.
@@Feaforall I believe then that i've been having troubles with the temperasture units. I assumed so far that temperature was displayed in Kelvin, given that I'm using SI units (m, kg, s, K).
Specially with para vis, you read the temperature as Celsius. How can I know if its either Celsius or Kelvin?
@@nicolaerodideal9718 www.eng-tips.com/viewthread.cfm?qid=386704
Hi, I've a question. How can I do Transient Thermal Simulation not from the source but from the air? I can't find in anywhere.
You can use the convection load with an ambient temperature which depends on a function. The convection is what approximate the contact between your model and the external air, but this needs the approximation that the air temperature around is constant all over the model. If this is not the case, you’ll have to do a CFD heat-structure coupled simulation (Code-aster can’t do that)
@@Feaforall Temperature is constant all over (800 degrees). Thank you very much for Your respond.
@@Wikimika1411 You're welcome. In addition, i recommend you to read this article on the blog, and especially the part about convection: feaforall.com/heat-transfer/
That should help you to understand better how heat transfer works
Hi Cyprien! Another great tutorial!
You used SOURCE to apply thermal load on a group of elements.
Its possible to apply to a group of nodes? SOURCE only have "Group of Elements" as a option.
In Ansys i can use the "F,,heat,value" to apply heat loads directly on nodes.
Its possible to do this in Aster?
Thanks!!
I think there is probably a command to change a group of nodes into a group of elements... but never tried
@@Feaforall actually there is:
# Create a group after selecting the nodes
local_nodes=welding_mesh.MakeGroupByIds("local_nodes",SMESH.NODE,selectednodeIDs)
# Create a group of elements which contain the nodes
local_elements= welding_mesh.CreateDimGroup( [ local_nodes], SMESH.VOLUME, 'local_elements', SMESH.ALL_NODES, 0)
but i really wanted to apply directly to nodes..... :/
@@rodrigofarias900 According to the doc, SOURCE has to be applied to elements, not nodes. There doesn't seem to be an option for nodes: www.code-aster.org/V2/doc/v13/en/man_u/u4/u4.44.02.pdf
But maybe I am just not aware of it... Try to ask on the Code_aster Forum, they may have more answers there!