---

Hallo,everyone

I am doing a simulation on Multi-layer welding with ANSYS by using the double ellipsoid heat source and Element Birth Technique. In the simulation process there has been a very strange phenomenon. If I use the BFE command to apply the heat generation rate to the element, some elements, which are not activated, have results -700°C temperature, but the final result is correct. However, if I use the BF command to apply the heat generation rate to the node, the phenomenon of -700°C does not exist, but the final result is wrong. What is the reason? How should I solve it? Thanks a lot

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Do you use an iterative solver and -700 is a premature value of the inactive elements? If so, it is not necessary wrong.

------------------
Grüße, Moe

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

hallo, i have use he JCG solver. And after the element is active. the nodal temperature get right.see the graph. the negative temperature is appear, when the element is not active.i set the initial and room temperature to 20°C.do you mean that the negative temperature can be ignore? why this phenomenon appear?

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Zitat:

---

Original erstellt von farahnaz:
Do you use an iterative solver and -700 is a premature value of the inactive elements? If so, it is not necessary wrong.

---

hallo, i have use he JCG solver. And after the element is active. the nodal temperature get right.see the graph. the negative temperature is appear, when the element is not active.i set the initial and room temperature to 20°C.do you mean that the negative temperature can be ignore? why this phenomenon appear

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

General speaking, solvers may need a number of iterations to converge to a realistic range. I experience it in CFD simulations. How is it in your case? Are the final results reasonable?

PS: Did you attached a graph? Can't find!

------------------
Grüße, Moe

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Unbenannt.PNG

Zitat:

---

Original erstellt von farahnaz:
General speaking, solvers may need a number of iterations to converge to a realistic range. I experience it in CFD simulations. How is it in your case? Are the final results reasonable?

PS: Did you attached a graph? Can't find!

---

sorry. it is my first time in this forum. for the base plate the final temperature is very similar to the experiment. but for the weld bead the final temperature is not right. some layerŽs temperature are higher than the experiment and some are lower

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Zitat:

---

Original erstellt von frank12:

for the base plate the final temperature is very similar to the experiment. but for the weld bead the final temperature is not right. some layerŽs temperature are higher than the experiment and some are lower

---

Difficult to say! Using CFD, the correlation with experiment is not always perfect.

The temperature gradient should be high around beads. Try to improve the FE-model by local mesh refinement and increasing the element quality if possible. Additionally, double check the material properties and how accurate they are.

------------------
Grüße, Moe

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Zitat:

---

Original erstellt von farahnaz:

Difficult to say! Using CFD, the correlation with experiment is not always perfect.

The temperature gradient should be high around beads. Try to improve the FE-model by local mesh refinement and increasing the element quality if possible. Additionally, double check the material properties and how accurate they are.

---

i have refined the mesh and the material properties is right.now i think the problem is about the command BF and BFE. Do you know the different between the both command?some papers use the BF and others use the BFE, i donŽt know the different expect that, BF is apply in node and BFE is applied in Element.

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Never tried, but I think the difference is that BFE is applied to elements vs BF to nodes.

https://www.sharcnet.ca/Software/Ansys/17.0/en-us/help/ans_cmd/Hlp_C_BFE.html
https://www.sharcnet.ca/Software/Ansys/17.0/en-us/help/ans_cmd/Hlp_C_BF.html
..

------------------
Grüße, Moe

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP

---

Zitat:

---

Original erstellt von farahnaz:
Never tried, but I think the difference is that BFE is applied to elements vs BF to nodes.

https://www.sharcnet.ca/Software/Ansys/17.0/en-us/help/ans_cmd/Hlp_C_BFE.html
https://www.sharcnet.ca/Software/Ansys/17.0/en-us/help/ans_cmd/Hlp_C_BF.html
..

---

thank a lot.finally i find out the problem. when i apply the smaller heat generation rate,the negative temperature will not appear.but if the apply the bigger heat generation rate, the negative temperature appear again. do you think that the problem is about the material properties?

Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP