Eng-Tips is the largest forum for Engineering Professionals on the Internet.
Members share and learn making Eng-Tips Forums the best source of engineering information on the Internet!
-
Congratulations JStephen on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Welding simulation
- Thread starter jungi
- Start date
Are you looking for the software to add elements as the weld bead is placed and fills the gap between plates? Or are you just trying to simulate the thermal flow? Does it need to do plastic flow? What, a little more precisely, are you trying to find out?
Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group
Magnitude The Finite Element Analysis Magazine for the Engineering Community
Garland E. Borowski, PE
Borowski Engineering & Analytical Services, Inc.
Lower Alabama SolidWorks Users Group
Magnitude The Finite Element Analysis Magazine for the Engineering Community
- Thread starter
- #3
Thanks for reply GBor,
I'm looking for software adding elements…. my problem is that my part "explodes" after welding… and the FEA stress simulation of the completed weld shows no problems with residual stresses… that’s why I want to have look at what happens, with the part, during the welding… and I hoped a proper simulation could help… maybe you have any suggestions?
Best regards
Jungi
I'm looking for software adding elements…. my problem is that my part "explodes" after welding… and the FEA stress simulation of the completed weld shows no problems with residual stresses… that’s why I want to have look at what happens, with the part, during the welding… and I hoped a proper simulation could help… maybe you have any suggestions?
Best regards
Jungi
Hi,
I know Ansys can do something very close to what you need (but not "exactly" what you need...) with its "birth-and-death" functionality. I say "not exactly what you need" because, in order to make an element "live", it must already be there, so you deterministically "fix" the position of the elements BEFORE beginning the simulation.
Regards
I know Ansys can do something very close to what you need (but not "exactly" what you need...) with its "birth-and-death" functionality. I say "not exactly what you need" because, in order to make an element "live", it must already be there, so you deterministically "fix" the position of the elements BEFORE beginning the simulation.
Regards
There are a couple of others that do what cbrn suggested, but they are not easy to set up. You may want to check on AMPS, but this may be a little difficult. Ansys is an option. Algor has a similar function to Ansys.
Are you saying you have a physical part that is breaking after you've welded it and at a much lower load than you would expect it to? And you are trying to determine why?
Are you saying you have a physical part that is breaking after you've welded it and at a much lower load than you would expect it to? And you are trying to determine why?
- Thread starter
- #6
Hi Gbor,
The part, actually a shaft, breaks apart before the load is applied... the guys in the workshop just lifted it from its support then it happened... the weld goes over the circumference of the shaft, is not very wide but quite thick - multiple layer. I want to see what happens with the shaft exposed to changing thermal stress during the process of welding…
The part, actually a shaft, breaks apart before the load is applied... the guys in the workshop just lifted it from its support then it happened... the weld goes over the circumference of the shaft, is not very wide but quite thick - multiple layer. I want to see what happens with the shaft exposed to changing thermal stress during the process of welding…
Hi,
ah, now it's clearer... Well, if you are going to simulate that with FEM, good luck, because you will face a quantity of challenges:
- build a model with crack-growth possibilities
- build a model with birth-and-death possibilities
- get an extremely detailed/precise constitutive law for both the base material and the welding material, taking into consideration that the properties will be temperature-dependant
- build a model which can be solved efficiently because you will have to perform transient coupled-field analyses (thermo-mechanics), but on the other hand you will have to use some lots of elements for the HAZ and around the crack(s) if any
Before doing so, if you can, I'd try to determine the causes of failure by metallurgic inspection. Did you already radiographed the specimen? Was the welding controlled? And if yes, how? Do you know if the specimen already had defects? Cracks? Crack initiators somewhere in "critical" locations? Did you examine the fracture surfaces?
With data of these kinds, a good metallurgical specialist can in 90% of cases determine the mechanism and thus the cause of the failure.
Regards
ah, now it's clearer... Well, if you are going to simulate that with FEM, good luck, because you will face a quantity of challenges:
- build a model with crack-growth possibilities
- build a model with birth-and-death possibilities
- get an extremely detailed/precise constitutive law for both the base material and the welding material, taking into consideration that the properties will be temperature-dependant
- build a model which can be solved efficiently because you will have to perform transient coupled-field analyses (thermo-mechanics), but on the other hand you will have to use some lots of elements for the HAZ and around the crack(s) if any
Before doing so, if you can, I'd try to determine the causes of failure by metallurgic inspection. Did you already radiographed the specimen? Was the welding controlled? And if yes, how? Do you know if the specimen already had defects? Cracks? Crack initiators somewhere in "critical" locations? Did you examine the fracture surfaces?
With data of these kinds, a good metallurgical specialist can in 90% of cases determine the mechanism and thus the cause of the failure.
Regards
- Thread starter
- #8
Hi,
I’ve checked the „birth and death“option for ANSYS, it should work… I also have quite precise material properties, temperature dependency inclusive… I think I can still take a look at the residual stresses during a transient simulation without taking cracks into consideration… just to see how big the maximal stresses are, and where they occur… it would be generally interesting to see if such a FEA simulation is possible and how much it fits the reality…
Regards,
I’ve checked the „birth and death“option for ANSYS, it should work… I also have quite precise material properties, temperature dependency inclusive… I think I can still take a look at the residual stresses during a transient simulation without taking cracks into consideration… just to see how big the maximal stresses are, and where they occur… it would be generally interesting to see if such a FEA simulation is possible and how much it fits the reality…
Regards,
Intercooled
Mechanical
Hi at all,
to simulate the welding process there is a module of MSC.Marc, based on Goldak theory.
The calculus is a thermal-structural coupled FEM simulation, the elements are activated in time and temperature through the path of welding.
I used Marc for a lot of simulation, but i'd never used it for Welding. Now i use Ansys, but for weld simulation there is only a spot weld element; i have only a mechanical license so for a transient thermal-structural coupled simulation a must do an upgrade to multiphisics license to simulate a similar calculus.
Best regards.
to simulate the welding process there is a module of MSC.Marc, based on Goldak theory.
The calculus is a thermal-structural coupled FEM simulation, the elements are activated in time and temperature through the path of welding.
I used Marc for a lot of simulation, but i'd never used it for Welding. Now i use Ansys, but for weld simulation there is only a spot weld element; i have only a mechanical license so for a transient thermal-structural coupled simulation a must do an upgrade to multiphisics license to simulate a similar calculus.
Best regards.
- Status
Similar threads
- Question
- Locked
- Question
