Hello
As part of a simulation carried out by a subcontractor, I have to convert an assembly containing welding functions to parasolid format. These weld seams are not solids and disappear during conversion. Is there a (macro?) way to transform these welding functions into solids, (cylinder, rectangular volume, etc...), and thus allow me to convert them into parasolids.
Hello @Franck Dumoulin
AMHA there is no solution for the following reasons.
1°) The solder functions are only indications for MEPs (for most editors) in SW they have no physical reality.
1bis) only the volume functions are converted and the solder is not a volume.
2°) In the best case, these are representations by images, which avoids having large files for functions that are useless in the assembly. (this is only used for MEP)
3°) The most important thing to know is that in the simulation the weld functions are not used for several reasons. Moreover, this is not significant for the strength itself and especially for mechanical welding. (I can explain this statement).
4°) in simulation, only spot welding is usable for a completely understandable reason because it means knowing the two surfaces in contact (size of the point), which SolidWorks does not know how to do for two volumes in contact (see the 3°).
The welding functions independent of the simulation are a big M... in SW and also Autodesk products to believe that none of their customers do welding. It's as if SW didn't know how to do an extrusion, isn't it! ;-)
Kind regards
PS: if someone has a macro that would allow to retrieve the information from the MEP and generate a volume cord in an ASM then I discreetly tell him that his fortune is made (I am willing to share the profits if he has too many) ;-)
Hello @Franck Dumoulin
AMHA there is no solution for the following reasons.
1°) The solder functions are only indications for MEPs (for most editors) in SW they have no physical reality.
1bis) only the volume functions are converted and the solder is not a volume.
2°) In the best case, these are representations by images, which avoids having large files for functions that are useless in the assembly. (this is only used for MEP)
3°) The most important thing to know is that in the simulation the weld functions are not used for several reasons. Moreover, this is not significant for the strength itself and especially for mechanical welding. (I can explain this statement).
4°) in simulation, only spot welding is usable for a completely understandable reason because it means knowing the two surfaces in contact (size of the point), which SolidWorks does not know how to do for two volumes in contact (see the 3°).
The welding functions independent of the simulation are a big M... in SW and also Autodesk products to believe that none of their customers do welding. It's as if SW didn't know how to do an extrusion, isn't it! ;-)
Kind regards
PS: if someone has a macro that would allow to retrieve the information from the MEP and generate a volume cord in an ASM then I discreetly tell him that his fortune is made (I am willing to share the profits if he has too many) ;-)
It works' po
Hello @Franck Dumoulin
AMHA there is no solution for the following reasons.
1°) The solder functions are only indications for MEPs (for most editors) in SW they have no physical reality.
1bis) only the volume functions are converted and the solder is not a volume.
2°) In the best case, these are representations by images, which avoids having large files for functions that are useless in the assembly. (This is only for MEP)
3°) The most important thing to know is that in the simulation the weld functions are not used for several reasons. Moreover, this is not significant for the strength itself and especially for mechanical welding. (I can explain this statement.)
4°) in simulation, only spot welding is usable for a completely understandable reason because it means knowing the two surfaces in contact (size of the point), which SolidWorks does not know how to do for two volumes in contact (see the 3°).
Weld functions independent of simulation are a big ...... (word censored
Thank you for your answer@zozo_mp
Points 1° and 1°bis and 2° are the reason for my post.
Concerning point 3°, the service provider asks me for the weld seams, or rather volumes representing the welds, which he will interpret as a link between bodies for the simulation via a software other than solidworks. I can't judge the usefulness of these, not being an expert in this field.
I had naively hoped that a macro could have allowed me to solve my problem, namely to transform these indications of welds into physical volumes....
@Zozo: for point 3 it really depends on how you weld. If you do full penetration continuously, it is indeed not very useful. On the other hand, for those who make discontinuous seams or who do not weld in full penetration, it can play a fairly important role since we will end up with concentrations of stresses at the level of the welds.
Conclusion for @Franck Dumoulin : the need to have welds in the simulation depends greatly on what I wrote above.
With us, we draw portions of volumes in the cuts for 3D saddle welds (or any other complex weld) and otherwise volumes in revolution or extrusion for the simplest.
It is also true that SW has a welding function that is only visual
I think he could have made an effort on that side too
A weld is a volume and it interpenetrates
Between 2 bodies that are now 1
So 1 piece welded with x body becomes only 1 body
and I think SW could have directly integrated it into the log
which means that it is still one of the functions that is not completed other than for the costing and visual
So yes we have to create an interpenetrating body
Sketching, smoothing or extrusions and bonding bodies (combined)
@+
Hello@gt22
For simulation, it is not necessary to create an interpenetrating body for several reasons.
1°) the first thing SW does at the time of discretization (the mesh is precisely to check that there is no collission) If there is a collision, it stops immediately.
2°) In the parameters we define that bodies are solidary "with or without penetration". This means that if you have two flat surfaces, they are considered as a single volume.
From my point of view, you have to see the head of the rooms to make a decision.
Let's take an example of a UPN (or a 20 mm vertical plate) welded at right angles to a plate or a part, then there is a joint contact without penetration.
In this case, if we want to see the forces on hypothetical welds, the vertical plate would have to be detached by a few tenths of a mm so that we can see the force on the weld itself. But this space distorts the simulation because the supports are no longer the same (hence the interest of putting without penetration which allows to take into account this play and the deformation it generates)
But then big problem with the mesh at the place of the space created, because it will be necessary to make a very fine mesh which will considerably extend the simulation time because it risks multiplying by several thousand or million the number of triangles and DDLs among others.
However, you have to remember two things: there are charts to define the resistance of the weld according to the type of weld (we did this before the computer because 235 Mpa is still 235 mpa even with software) and a static simulation of a weld says nothing about the dynamic resistance. You have to look at the PB of the natural frequencies and also the external vibrations that can act on the parts. I am not even talking about the material of the weld, which is never equal to the part. Only software specialized in framing manages welds correctly.
So we see that the problem is not simple and between us I am wary of simulation as an alibi for solidity. ;-) ;-)
Note that @Franck Dumoulin has not returned to the forum since.
Kind regards
Hello@gt22
For simulation, it is not necessary to create an interpenetrating body for several reasons.
1°) the first thing SW does at the time of discretization (the mesh is precisely to check that there is no collission) If there is a collision, it stops immediately.
2°) In the parameters we define that bodies are solidary "with or without penetration". This means that if you have two flat surfaces, they are considered as a single volume.
From my point of view, you have to see the head of the rooms to make a decision.
Let's take an example of a UPN (or a 20 mm vertical plate) welded at right angles to a plate or a part, then there is a joint contact without penetration.
In this case, if we want to see the forces on hypothetical welds, the vertical plate would have to be detached by a few tenths of a mm so that we can see the force on the weld itself. But this space distorts the simulation because the supports are no longer the same (hence the interest of putting without penetration which allows to take into account this play and the deformation it generates)
But then big problem with the mesh at the place of the space created, because it will be necessary to make a very fine mesh which will considerably extend the simulation time because it risks multiplying by several thousand or million the number of triangles and DDLs among others.
However, you have to remember two things: there are charts to define the resistance of the weld according to the type of weld (we did this before the computer because 235 Mpa is still 235 mpa even with software) and a static simulation of a weld says nothing about the dynamic resistance. You have to look at the PB of the natural frequencies and also the external vibrations that can act on the parts. I am not even talking about the material of the weld, which is never equal to the part. Only software specialized in framing manages welds correctly.
So we see that the problem is not simple and between us I am wary of simulation as an alibi for solidity. ;-) ;-)
Note that @Franck Dumoulin has not returned to the forum since.
Kind regards
@ Zozo
Totally agree with your conclusion
@+
zozo_mp,
Thank you for these answers, even if as I said, I can't judge the relevance of these welds for the simulation. The service provider, who is an expert, wants them. After listening to different sources, including you forum gentlemen, the only thing that everyone seems to agree on is that we can't do anything with this solder function
@+