Hello
I have just created a "wave plate" (a wavy disk) on solidworks and I would like to study its deformation when subjected to compression (like a spring).
But the problem is that, because of the ripple, I can't use the simulation tool.
My goal would ideally be to simulate two infinitely rigid disks coming together and thus compressing my part...
Thank you for your help
Kind regards
Hello
It's doable in Simu SW!
But infinitely rigid does not exist in mechanics. You just have to indicate the materials of each of the components.
Can you post your ASM but using the Pack and GO function and all in zip.
I don't remember what Pack and GOI is called in the versions before 2017 It's in file ===>???
Hello
It seems to me that it's a take-home composition zozo_mp'm talking about ^_^..
Hello again,
Since my file is not large, I can send it as is. This is not an assembly, just the part that I wanted to use the "SOLIDWORKS Simulation" add-in on.
Cdlt
Good evening
There are a few quirks in your 3D model, more on that later.
The simulation you have made works but is not in line with real life.
On the other hand, your imposed movements are not good as well as the pressure zones.
The simu won't be complicated, I'll do it to you in the evening and explain how I did it. This will allow you to reproduce with your previous version because I am in 2018 version.
Question: Is there a single washer or multiple washers stacked on top of each other?
To Plutarch
Good evening
Thank you very much for the time you devote to my problem!
The situation is as follows: this wave washer is the first of a multi-plate clutch consisting of a stack of friction linings and steel discs. So on one side there is a steel disc and on the other a fluid that exerts pressure. The system actually allows for progressiveness during the clutch phase.
Anyway, to simplify I suppose that my part is as if "placed" on a steel disc on which it flattens under the effect of the pressure of the fluid.
Cdlt
Hello Lepenico
The simulation is over, you will find two videos that show the deformation closest to reality. And the screenshots that show the logic adopted
Indeed because of a mistake when talking about pressure.
The pressure is the pressure exerted inside the piston. The piston has a surface it is the resultant pressure in bar X surface that gives the force of the piston. It is this strength that we need to know.
In the meantime I put 120N X 6 = 720 N but which are only distributed on the metal surfaces that are in contact with each other.
Indeed, you have made two mistakes but they are in the same logic. You have put surfaces to receive the force and for the contact surface. In fact, an arc only comes into contact at one point, but as the part flattens you have a surface that enlarges slightly at the end of the crushing.
I put a 5 mm contact arbitrarily, we can refine later.
To delimit this surface, I made sketches in dividing lines. These lines allow you to select (delimit) a surface to be able to select a small area in the simulation without any modification of the internal structure of the part, and therefore without affecting the simulation.
You didn't have to make your washer on the surface, especially since the simulation module doesn't like it. Because surfaces are not volumes. I just replaced with a swipe without changing your two sketches.
Another minor point, you use a lot of 3D sketching to make a vertical axis, it's not the fastest. But hey, everyone does as they want but almost no one does it otherwise the SW engineers would have fired the 2D function
As we don't have the same version, I'll make you a little tutorial (as @ac cobra does so well).
There you have it, there you go
Good evening Zozo_mp,
Thank you for your precious help! I managed to replace it with a sweep but I can't create a dividing line. I have the choice between "silhouette", "projection" or "intersection", I took intersection and I selected the derived sketch2 and the top face of the washer. However Solidworks tells me that the sketch does not have a contour geometry... In your tutorial, you use the origin and plan 3 but I don't know how this plan was defined. I attached a screenshot because I'm not sure I was clear.
Kind regards
C normal because the line is not a line, you have to make a sketch and it is the sketch that will be a separate area from the rest. So if I take your jpeg again, you need either two continuous lines 2 or 3° apart (only above the surface to be divided) and then make a circular repetition of these two lines on each bump. So lines on the bumps on the top and lines on the bumps below.
In the attached image I put rectangles but with lines it also fits. It is important to understand that these traits allow SW to make selections otherwise he selects the whole face. With the lines, we better understand the term "dividing line"
This allows you to divide the large area as you want
Hello Zozo_mp,
Thank you for your clarifications. I was therefore able to do a deformation study and I obtain displacements of the same order of magnitude (see attached image).
However, I am not sure of the constraint between the hem and the puck. I put a coaxiality but suddenly the parts are not attached for the simulation.
The problem with the contact area is that it gets bigger as the pressure force increases. I don't know if this increase is negligible or not and if there is a way to model it.
What tool do you use to create the video that shows the piece deforming (always nice to illustrate)?
Kind regards
Hello
I don't know how you made the contact area (2 parallel lines or rectangle). It is sufficient to bring the two lines closer together for each zone to reduce the contact zone.
In the result, we can see very well the incidence according to the width of the famous areas.
I did a new simulation without a cimbleau so as not to put any artifacts in the system. I replaced it with areas of the puck's infield (small diameter).
I haven't refined it but I'll see how to do even better in terms of realism, i.e. as close as possible to real life. Because the washer changes these two diameters if you crush it completely, but it doesn't count in centimeters.
If you want to better see the deformations at the crest of the undulations, just do a simulation on a quarter of the part. By placing the intersection lines correctly, it is as if we only have a cross-sectional view.
That said, if we look at the numbers, if we put a total force of 2100 N, we have:
For a sheet thickness of only 1.9 we have only 0.186 mm of displacement while we arrive at 6.912 for a yield strength of 7.000e+08.
This shows that your material is not very elastic and therefore the change in curvature at the point of tangency with the piston or anvil is "upsilonnesque".
I'm going to make a very complete tutorial on the subject of crush wave washers because you're not the first to pose this problem, especially for ball bearing lock washers.
Kind regards
PS: You have a defect in your part because the inside diameter is not cylindrical but patatoidal. It is better to remove material to rebore the inside diameter in a way. This makes it easier to choose the support points for imposed movements (on cylinders for example). Look here at what the tutorial I'm doing first for you and the community might look like (better).
Good evening
I recreated the washer by extrusion to avoid this non-cylindrical problem. As a result, the seat can also be avoided. The results seem consistent.
I enclose the model.
Kind regards
Hello Leprenico
Very good idea to use the deformation to obtain the undulation On the other hand, you have made approximations by not limiting the deformation area. We can discuss this later because if you look closely, your waves have flat areas. (I redid the piece entirely with a 3D spline on a surface and now the ripples are perfect with control of the curvature. I have the part available)
On the other hand, it's not by chance that I put the centering seat
This allows the tops of the ridges to rotate in two directions simultaneously as the puck flattens.
Indeed, in simulation, it must be remembered that you cannot put constraints as in the assembly but you put imposed displacements which amounts to almost the same except that depending on the case you have to redo in part what you have constrained in the assembly.
Thus, solidarity (and/or without penetration) does not mean welded.
The disadvantage of the way you made your constraints is that the edges are fixed (and therefore cannot move in rotation which distorts the movements compared to real life. So out of the four zones of the ridges, only one must be fixed, the others must be with translational displacements (use the advanced function). In the first according to plane direction 1 and the second according to plane direction 2 put 1mm.
This remark also applies to the centering constraint that you did not put (since the bottom ridges are fixed). This is a mistake because when your washer flattens it lengthens by 0.12 mm and the diameter varies by 0.02 so if everything is fixed its deformations not taken into account will have an impact on the result of the deformation. This can be seen very well in Von Mises or in Voyages.
If you make an approximation, it's not a problem if you want to be rigorous, it's better to make the right moves. (This is good design hygiene to adopt from the beginning of your career) For example, if you had axes with parts connected to each other, you would have deformations greater than the tolerance of the bores, for example.
Kind regards