In this video, learn about the various stress quantities available, and carry out contour plotting.
- [Instructor] Stress contour plots are very useful in understanding the overall distribution of stress. However, if we want to get a detailed understanding of a local stress region, then X-Y plots of stress versus distance are invaluable. Now what I want to do is to plot maximum principal stresses. In our result set here, we haven't got that setup at the moment, so we can do that. Right mouse click on results, say define stress plot. In the definition and the display options, let's choose first maximum principal stress.
We can check the box there, and that's now being created, and we see immediately the plot is being created here. So what's active at the moment are the maximum principal stresses, we can see the stress range in here. If we want to we can tidy this up, so let's go again to chart options, and we can say, let's switch off the de-form shape, and in the settings we can say we want the fringe option to be discrete, and we want the boundary conditions again to be a mesh.
Click that. And that would be our typical starting point. Let's just zoom in and see what that looks like on the little end. So, these are maximum principal stresses, and the sense of these is going to be stresses, for example, running around the surface in here. So principal stresses are useful. They tend to follow the flow of things such as, for example, the edge direction there. Now what we want to do is produce some x y plots of the stresses in these different regions.
So with the principal stress active, we're going to spawn some x y plots. Now the idea of the x y plots is they're spawned off whatever stress state we've got, so for example if I double clicked on the von-misis stress, and then created some x y plots, they would be based on the x y stress. So again, let's double click, make the principal stresses active, and then create our displacement plots off that. So if I right mouse click on this stress result here, and click on probe, new dialogue box coming up, which controls the x y plotting.
Now let's change the radio button to on selected entities, and it's looking for geometric entities to overlay the x y plot on. So let's choose the inside arc of the little end, which is this edge here. You see edge one now pops up in here. Let's click update, and as you can see there's a spreadsheet now giving us the value of stresses, and it's the stress contour that we choose, or the stress quantity that we selected, which is maximum principal stresses, x and y and zed positions, and also the node counts.
So there's a tremendous amount of information here giving us detail stresses about that split line there, it's called split line three, that particular edge curve there. Now there's various things we can do here. We can, for example, directly create a plot. If I do that, get a very nice x y plot showing, the variation of stress, principal stress, around that edge. Now it's a parametric distance, and it's based on that edge length, so that would be parametric length zero, and then this is the key parametric length we're interested in, the area that's got the most action, which is parametric distance one.
So you can see that profile there follows the stress build-up pattern in here. It's very exact, very precise, and a lot of checkers, a lot of reviewers, will want to see this. I tend to think of the contour plot as being subjective, and then the x y plot is very objective. We've also got this little crosshair in here, which is very useful, if I want to know what is that big value? It will give me at the bottom here the parametric distance and the value. So if I move over here we lose the crosshair, but you can see that the value is here, so we keep an eye on that value, move up to here.
Now it's about .89, and 34,666, for example. So again, I can kind of monitor the stress results anywhere in here. Now with that plot we can save it as, which will save it as a comma separated variable file, and then we can take that into an Excel spreadsheet. We can print it out, and it will print an image of the plot. The plot actually stays persistent while we've got the database open. So if I create other plots, they will also be available, so it's very useful.
Down in here I also have further controls. I can directly save it off the same actions we saw before, or I can save it as a sensor. A sensor is then something that's remembered by a SOLIDWORKS simulation, so this particular line here will be remembered as a feature, a sensor, so we can come back to this plot state at any stage. So rather than having to re-pick it, we can name it, and again that's really useful. We can also capture this as an image, which again for report purposes is very useful.
Now I could choose another region on selected entities, so let me this time pick this line which is running up the center line. Now we have to be careful here we don't stack these up by mistake, it's kind of a skinny little dialogue box, and if we're not careful, we think we've only picked edge two. So we want to deselect edge one, and then we just have edge two there. So we click update, then we can choose a new plot of that region, and now you can see it clipped at 32,000, goes up to 40,000, so again be careful it's not going from zero.
But it's showing the stress variation running across that plot there, so we can see it's a little bit jagged, we could have improved the mesh fidelity to try and make that smoother. Again, I can pick off particular values here with the crosshairs, which is extremely useful. Now one thing that this indicates to us is that the peak stresses shown on this kind of region running through here, but if we look, the actual peak stresses, more like, it's about five to 10 degrees offset from the vertical there. That's a clue for when, if we want to refine this analysis, what we could do is go back into the geometry, put another split line down there.
The split line is then going to control where we're going to be getting results from. So we talked about modifying geometry for loads, bad conditions, and meshing. Once we know where the action is, we can also then go back, refine the geometry to pick up best positions for stress quantities and so on. And if you're doing that, then you're developing a very sophisticated analysis model and you're really going to please a checker, because you're putting that control in exactly where the key stress results are going to be.
So again, these two plots are persistent, so I can show them both at the same time, and I can spread them around and review, and see what's going on. If I wanted to take a screen capture, then again, I've got lots of controls, lots of ways of showing what's going on. Now I'm going to delete that plot. We're going to show another technique that we can use for finding stress values. Let's zoom in a little bit tighter, into this bottom region in here. Now at the moment I've focused on, on selected entities, so these are pre-defined entities, as I've mentioned earlier, very, very convenient.
If I don't have any pre-defined entities, I can go to at location. What I can do here is to literally pick off element by element or node by node. Now what quantity I'm going to pick off, whether it's nodes or elements, depends on the fundamental underlying plot controller set up here. In my case, I set up nodes, so I'm going to be looking at nodal results. So with that set up, let's pick some stresses at positions out here.
Pick out a node there. You can see there's a little dialogue box that pops up, which shows me the node ID, the x y zed location, and then the stress value. And then that's, the next one, say on the center line there, roughly, on the line of action. I can cut a go along like this. Now I can click these and drag these into nice positions, but when I reach a certain number, so kind of setting up ready for the report, get beyond that six and it all goes crazy, and then they're just scattered all over the place.
But that's okay, because what we've got now is, I got to reposition these, but you'll notice a little pin there. If I say pin it, that's going to stay put. And then, it's actually quite quick to set this up, and it's a very, I think, quite an elegant way of actually showing where these big stresses are. So pin that one there. That one there, you can obviously be a little bit artistic with the way that you lay these around. That fella can come over here, pin him. Now if I also bring back my x y plot, it's really a very nice description of what's going on.
There's the x y plot running down in here, and here we're pulling off the hard numbers. Now at the moment it's a little bit verbose because we've got the element ID, or the node ID, I should say, got the x y and zed location, we don't really need that, so let's deselect those, and then we're just down to the value, which is a much cleaner picture. I can probably get my x y plot on there at the same time, so, that in a report is pretty definitive of what's going down that line there, so again, a checker reviewer can see exactly what's going on.
And obviously you can play games with that, you can set up very powerful, very useful kind of plots to show what's happening here. I've finished with that, I'm going to shut down the graph, and I just check the box to cancel out from the data probe. So in conclusion, x y plots are very powerful, they enable a very precise understanding of stress distributions. In addition the data probe can annotate contour plots with key data. If we carefully position the geometric curves and edges, that will really enhance the power of this tool, so as I mentioned, it's worth splitting geometry, then re-meshing and reworking the analysis to improve the overall quality of the stress results.
- Setting up Simulation properties and defined views
- Preparing the geometry
- Setting up a local coordinate system
- Splitting surfaces
- Defining the constraint and the loads
- Running analysis
- Contour control
- XY plots