Post By Wayne White
How many of us can interpret "surface-extrude 1" from "DeleteFace2" without first interrogating the model? Not many; when surfacing, remember use folders!
Add a folder to the design tree by a RMB. Choose Add to New Folder form the menu.
Also, check out Body Delete.
RMB on your remaining surface bodies and choose Delete Body. This allows you to clean up the tree further without affecting the solid body. Ultimately, you want solid geometry. The surfaces just hang around as references. Here, we eliminate them to ‘tidy up’ that tree.
I recently attended a Mini-maker faire. “Maker Faire” is an organization and a brand. Started by the people at Make Magazine back in 2006 in San Mateo, CA (close proximity to the TechShop origins), the popularity spread to NYC and now groups may organize their own “licensed” local maker fair – a “mini” maker faire.
The events are family oriented with young children often involved. At the faire I attended, nearly every corner had something interesting for kids to do. There must have been 1,000ft2 of Legos, including a table full of parts to build whatever you want. There was a Star Wars R2D2 for them to talk to and a great display involving a large metal sheet wall and pvc pipe pieces attached to magnets. Basically, the pipe sections stick to and slide around the large “whiteboard” to allow the creation of a roller coaster of tubes for a small ball to pass through. The kids loved it.
There were three 3DP displays, including one with a Kinect tied to a Makerbot Replicator along with a homemade turntable. A person stood on the turntable and as they spun around, they were scanned by hand with the Kinect. After a few minutes of cleanup in the software, the scanned image was sent to the Replicator. There were Arduino vendors and Raspberry Pi kits, artists, a hotdog vendor, and the whole thing was passed over by a Parrot-style quadricoptor controlled by an unseen person.
But… no 3D CAD. I was approached by one gentleman inviting me to the grand opening of his maker space. He informed me that they have a 3D printer, a laser cutter, a welder, and an assortment of hand tools. He had never heard of SolidWorks. There were no displays at the faire involving 3D CAD with the possible exception of the 3DP displays and then it was just software for scan cleanup. I thought maybe the reason is that the concept is difficult for children, but then why the Arduino and Raspberry Pi kits? After all, 4th graders are working on algebra at night.
My guess is the emphasis with Makers is, well, "making" – hands-on. 3D CAD is just something between a Maker and a 3DP part in their hand to build a robot or a piece of art. And by no means is the power of SolidWorks necessary for most of this activity. From Legos to 80/20 extrusions, these mass-produced, angular objects assemble quickly and inexpensively into fantastic creations. No 3D CAD required.
However, as we know, later when the parts get more functional and the design becomes important, these kids are going to want the power of SolidWorks. Showing kids what is possible is part of the Maker goal and the world isn't rectangular. So, find a local maker faire and get involved - they need our expertise.
Post by Tim Pulaski
Weld lines are a natural and expected occurrence in plastic injection molding, especially when multiple gates are involved. Though this may be the case, it sure would be nice if they weren’t. In addition to being visually undesirable, weld lines produce areas of local weakness which can have a drastic effect on a plastic part’s ability to perform as designed. If not controlled properly, weld lines in the wrong areas can lead to significant (read: costly) mold rework.
A popular and effective way of controlling weld lines in a mold with multiple gates is through the use of valve gates, which can be independently controlled to inject plastic into the cavity at specific intervals. Determining these intervals is the tricky bit, and can be prone to much real world trial and error. Solidworks offers a solution to this problem in the form of Solidworks Plastics Premium, which allows you to analyze and control the filling of your plastic parts in a familiar interface.
It’s very simple to do – take this car bumper fascia for example.
Figure 1 Gate Locations
It’s a large fiberglass part, so not only are weld lines a concern, but having enough pressure behind your melt to fill the extremities can also pose a challenge. For this part, 4 gates were used (blue stars in Figure 1). If I were to inject through all 4 gates simultaneouesly, I could expect numerous weld lines:
Figure 2 Simultaneous Injection
Each color represents the filling contribution of a single gate.
Right down the center too! No good. The required pressure to fill this part in this manner came out to be around 12,000psi as well, which while not impossible is certainly excessive. Let’s try a different approach using valve gates.
The procedure I used was to first inject plastic into the part using only the single, left-most gate shown in the previous images. From the results of this analysis I was able to determine the time at which the flow front passed the second gate, and used this time to specify when the second valve gate should open. I repeated this procedure for the two remaining gates in sequence to produce a cascade effect which eliminated a number of large weld lines and relocated others to more reasonable locations.
Figure 3 Cascade Injection (using valve gates)
Additionally, the required pressure plummeted to a mere 5300psi, less than half what was required using simultaneous injection. Pretty cool!
This is a great tool if you work with large or complicated models often as you can test out different methods and approaches to various problems with ease.
I’ve explained the pieces that make up a drawing(Blog Post on Drawings), which will help in understanding why you may run into an error message about a missing sheet format or the "wrong" sheet format is added to new sheets in a drawing.
The SolidWorks drawing file references the template which includes the sheet format – the title block, borders, etc. The sheet format is stored with the drawing template, not the drawing. The drawing template is saved with the drawing. If you save a drawing template without a format, SolidWorks will ask you to select one when you start a drawing with that template. Some of you reading this may even prefer that function and already have your system set up in that way.
The consternation arises when an existing drawing is opened, a new sheet is added, and SolidWorks explains that it cannot find the sheet format or uses a different format. “But it’s right there!” Actually, it’s not “there”. If it was, SolidWorks would find it. The problem is that the location for the sheet format for new sheets in the drawing aren't coming from the opened drawing. Let me explain:
When the drawing is created, the sheet format is in location A – for example c:\program data\solidworks 2012\my sheet formats. Years later, I get ambitious and clean up all of my old SolidWorks 2012 folders. The sheet format info is saved with the drawing (or template). So when the sheet formats from 2012 are deleted, we don’t get an error like we do when we open an assembly that's missing parts or when we open a drawing that cannot find the model it is supposed to detail. However, when a new sheet is added to the drawing, it looks in location A where the sheet format used to be. The solution of course is to modify the old sheets to point to the format in the new location or add the sheet format back to the old location. However, there’s no way to tell where the old location was if you don’t remember – it’s not in the list of file references because it isn’t a file reference. Also, one of the default install directories for sheet formats is typically hidden in Windows Explorer - C:\ProgramData\SolidWorks\SolidWorks 20XX\templates\Drawing.drwdot. Change the View settings in Windows Explorer to Show Hidden Directores to find that one.
Once you understand this, you can leverage the behavior. For example, many companies use a different format for sheet 1. So it stands to reason that if you create a drawing template that uses “sheet format 1”. Then, move “sheet format 1” to a new folder and replace “sheet format 1” with a different sheet format needed for all subsequent sheets, but saved as the same name - “sheet format 1”. All new sheets will have the new “sheet format 1” - and you can collect that reward offered for the missing sheet format.
"It's Season" can mean different things. If you live in a coastal community, “season” refers to the time of year when the temporary residents are in town – south in the winter, north in the summer. Other things come to mind if you’re a hunter or if you’re a SolidWorks reseller.
When a new SolidWorks version comes out, I call it upgrade season. In preparation for upgrade season, I’ll offer a few tips:
- Check the SolidWorks website for graphics card and operating system compatibility.
- While you’re on the website, make sure you know the login to the Customer Portal. Customer Portal Info
- If you are a regular reader, you know I stress back-ups. With upgrades, that includes backing up your SolidWorks settings using the SolidWorks Settings Wizard.
- SolidNetworkLicense Manager (SNL) will work with older versions, so upgrade that first.
- Workgroup needs to be at the same major version as SolidWorks. EPDM has to be the same or can be newer.
- Keep in mind that Workgroup cannot be accessed by an older version of SolidWorks. (Read that closely)
- Make sure your subscription is up to date before embarking on any upgrade. If you aren’t on subscription, there will be significant downtime while we sort things out.
- Decide what you want to do with toolbox. You can upgrade or install a new one.
It may be time to invest in a virtual machine software package - not a new computer, but software. Think about how easy it would be if you had a chance to practice (Allen Iverson anyone?). It's a minimal investment for maximum peace of mind. Set up a client and a server on one machine and upgrade.
Finally, please give us a call before upgrading and let us know. We can check subscription status, let you know about any common issues we’re seeing (if any), and talk to you about the specifics of your upgrade. Season is a busy time of year, so let us help keep those unexpected ToDo items off your list.
It's available right now. Enjoy, you early adopters.
SolidWorks Labs (labs.solidworks.com) is no more. It has vanished into the ether. Tree House, Watch It widget, and Drawings Now are gone. It was a nicely kept secret for total SW geeks. SolidWorks would like you to go to My.SolidWorks for such things, but it ain't the same. Labs offered ideas - all good, some half-baked, often broken ideas that were "kewl". Admittedy, there didn't appear to be much effort exerted on Labs for quite a while. But the goal of My.SolidWorks is to have a gathering place for lots of SW Users and Labs wasn't that type of place.
Adaptive methods help you obtain an accurate solution for static studies. There are two types of adaptive methods in SolidWorks Simulation: h-adaptive and p-adaptive method. The concept of the h-method is to use smaller elements in regions with high relative errors. The p-adaptive method increases the polynomial order of elements with high relative errors - higher order elements which you understand from reading the previous Simulatrion blogs - or you're just a smarty pants.
With h-adaptive, SolidWorks Simulation uses the stress in the study to determine areas where it should refine the mesh. With p-adaptive, there are three criteria to use for the convergence evaluation. The order of the suspect mesh elements is increased until it reaches convergence - the point where more calculations produce the same results or worse, the point beyond which we get the wrong answer.
Things get a little complicated from here. From the earlier posts on meshing, you can understand the pitfialls for each - more elements versus higher order elements. The cool thing about h-adaptive is you can coarsen the mesh (is "coarsen" a word?) in areas to speed up things. With p-adative, you only run more calculations where it's needed. However, as with other meshing options, there is no "best" option. I won't go into more detail on the adaptive meshing. You'll know it if you need it. My overall message on meshing is twofold:
1) If you want to "try it" to get your model to mesh, then you're not approaching the mesh from the correct perspective.
2) Provide some basics on meshing to pique your curiosity and counter some of the incorrect assumptions out there.
Now go simulate something.
Once the model is simplified to accommodate the best balance between mesh density and result accuracy, the next step is to optimize the mesh. Recall from 1-minute calculus that we need to use something we know to approximate something we don’t. Thinking along those lines, the more regular our tetrahedrons or triangles are, the more accurate the result will be. It's easy to calculate the area of right triangles and isoceles triangles. "Regulah" as they say in the general vacinity of SolidWorks' HQ is translated as mesh quality or "aspect ratio". Aspect ratio is calculated three ways by SolidWorks - edge length, normals length, and inscribed/circumscribed circles check. For higher order calculations, there is yet another mesh quality check - Jacobian, but we're not going there.
inscribed/circumscribed circles check
(can you tell which one is better?)
Of course, measuring the aspect ratio of millions of elements is a detailed process. Since we can't affect that calculation, know that basically SolidWorks is making sure that most of the mesh elements can be used in the calculation.
Let's pause for a moment to discuss "higher order" as I mentioned above regarding a Jacobian quality check. By higher order, we are referencing the exponents of the equation describing the curve we need to create a mesh element that approximates the shape of the model face (only 1 comma in that whole sentence). An equation representing a straight line has no exponent (mx + b = y). As soon as you step to x^2, the "line" curves. A "higher exponent" like x^3 or x^4 is synonymous with "higher order". I don't want to get side-tracked with this, so I'll keep it simple - we've been discussing triangles and tetrahedrons, but we could use mesh elements that are parabolic to approximate curved surfaces - higher order elements. Now back to our mesh quality discussion.
So, why does SolidWorks calculate mesh error for us? I'll start by explaining what makes a "good" mesh. Let's look at the thickness of a part, as an example. We want our elements’ size to match that thickness - exactly one "triangle" high.
Or consider an inside or concave radius. That curve requires a curved element. If we are using a relatively large element, then the elements will have to curve, so we either need a higher order element or better yet, a smaller element so the discretization error is minimal. Otherwise, we get this "jagged" result around the radius:
If we don’t pay attention to these mesh considerations, Simulation will do its best to calculate what we ask, but the result may be elements in the mesh with aspect ratios that negatively affect our results. Remember, the more simple the shape, the more regular the shape, the more accuate the result becomes (perfect vacuum and frictionless ice). Said another way, the approach to calculating the volume of these two pyramidal shapes is fairly similar, but one result is going to be much more accurate...
To avoid providing obviously wrong results, SolidWorks performs a mesh error check. We can help ourselves, though. To address the need to minimize the mesh density, and optimize the element aspect ratio, we can use mesh refinement techniques, the basis of which I've been attempting to explain. Here's what SolidWorks offers in order of refinement:
- Coarse to Fine adjustment of global mesh
- Mesh parameters (adjust the global size and tolerance)
- Automatic transitions (make sure the nodes from element to element match up)
- Curvature-based mesh
- Incompatible Mesh option (nodes don't match up)
- Mesh Controls (adjust the mesh in specific areas of the model)
- Mesh Methods (discussion coming soon in a blog post)
So, if you run a simulation in SolidWorks and get a mesh error that stops the calculation, that list is where you start to correct the issue.
My last post discussed meshing, more specifically Standard versus Curvature based meshing, so I’ll continue from there. We can continue the mesh discussion with mesh quantity or mesh quality. I’ll go with quantity.
More is not always better with mesh element quantity. You may recall from the patent-pending one-minute calculus post that we are approximating an infinite number of mesh elements, so in theory we can have that many. Like the choice between Standard and Curvature based mesh, we need to ask ourselves why SolidWorks Simulation doesn’t just use “a lot” of elements. The answer is efficiency, or in the FEA world the term is convergence. Each new element introduces up to 10 new calculations - nodes. Each new calculation increases the resources needed to make the calculation and the time it takes to run. Each time a calculation is made, error is introduced. After a certain number of elements are used, the return on investment is very small and in some cases, wrong. Convergence is a separate topic, so I’ll just note that more mesh elements is not always better and we need to minimize the number of elements to the best of our ability to get the best results.
One way to minimize the number of mesh elements is to use defeaturing. It’s right at the front of the Simulation training manual, but on our side of the support inbox, we see it over and over again – models that are too detailed. Overly detailed models create issues for meshing. Detailed models require very small mesh elements or at least mesh controls in localized areas where the geometry isn’t suited for the global mesh type.
The problem begins with the thought that we want to simulate the exact product in the real-world environment. Physicists use things like frictionless ice and perfect vacuums for a reason – that’s the only way the math works. Another example is modeling water in a tank to find the stresses in the tank. There’s no need at all to model the water – we apply those forces to the tank with Simulation.
What we are creating is an approximation of the real world, so there’s no need to use a model with all of the details needed for use in the real world. Using an overly detailed model will not improve the outcome. Besides, the results – regardless of how detailed our model and forces are – will always need to be correlated. No one builds a product without physical testing (no one that stays in business). The key is to create a Simulation setup that closely predicts the physical testing outcome.
There is a Defeature tool in Simulation, but defeaturing can be done in other ways. Some people create simplified configurations for their Simulation runs. Suppressing features is another way. Suppressing or removing parts of an assembly is also a good choice. For example, the vanes in a water pump are needed for a flow analysis, but not to simulate the mounting point stresses on the housing.
So, idealize models, take out what you don’t need, and then work on the mesh.