Hi all. I started using Nastran 4 months ago, and since then I've become a power user of it, using it consistently for 20+ hours a week. I truly think this program has a lot of potential, and it really is a shame that the workflow within Nastran is hindered by often-times small user interface quirks that result in frustration or downright poor user experience. I browsed the idea section here for Nastran to see if some of my issues had been already suggested, I often didn't find threads specifically talking about my issues/ideas, whilst some other threads are similar or identical in nature, but are so old they are basically archived and they still haven't been acted upon. So, today I'd like to bring you my collection of QOL (quality of life) improvements that would see Nastran become a really A tier product, and I thought rather than making 10+ threads I'd make one big thread and see what the response is to it. If the moderators tell me this thread should really be broken up, whether into a few posts or all 13 different topics, then I'll do so. Many of these suggestions tie into each other too. 1) Inability to add several items from Nastran Model Tree at once It is currently not possible to drag multiple items (be it forces, contacts etc.) from the model tree into the analysis. If you try to do it, it will only place them one by one. If you want to place multiple at once, you have to select them and copy them, then paste it on the analysis above - the only catch is that for most things in the model tree, this is only possible if something was already placed/is already there, otherwise you have to drag something in there first. Example: you cannot paste contacts in the analysis if there are currently no contacts present (the menu for them is missing). You have to drag an existing contact from the model tree onto the idealizations (you cannot paste a contact onto the idealization menu, but somehow you can drag it on there), then copy/paste the other ones. However, every time that you will drag/paste something into your analysis, the Model tree will collapse, which leads to the next parts: 2) "Model" section of Nastran Model Tree collapsing (suggested here) The "Model" section of the Nastran Model Tree collapses when edits are made within or from it. Anytime you drag anything from there up to your analysis, it will collapse. If you delete a single item, it will collapse (not if you delete multiple, interestingly). The same is true when creating multiple idealizations. Between each idealization creation, the Model section collapses and must be re-expanded to create an additional group. 3) Inability to use keyboard shortcuts to copy and paste parts This is quite simple, but why is there no option to ctrl + c to copy things from the model tree, and ctrl + v onto the analysis? Or ctrl + n to create a new item of whatever we have selected in the model tree. 4) Missing undo function Not really sure why there is no undo/redo button/function in Nastran, because it is REALLY, REALLY needed. I cannot tell you how many times I've cursed at myself for my hand slipping and cancelling/deleting something and having to waste time setting it up again. The lack of ctrl + z / ctrl + y is sorely missed. 5) Duplicating analyses (and/or subcases) duplicates every single Model tree item (brought up many times, such as here, here and here) Like others have stated before, when I am duplicating an analysis, in 90% of the cases I just need to change something small in the analysis, without wanting to run the risk of messing up my current analysis or losing its results. This means that I don't want every single item in there duplicated, as this just floods the model tree. it should just copy your analysis 1:1 without duplicating, or give you a pop-up asking what items (connectors, mesh refinements, loads etc.) you want to transfer over OR duplicate. And, if this "duplicated" analysis is later deleted, ask whether the idealizations, loads, etc that uniquely belong (thus are NOT shared with other analyses) to said analysis should be deleted. This should really be the same for subcases. New loads, constraints etc. should NOT be created!! This suggestion was accepted in 2019, and in the 2024 version is yet to be implemented (not sure about 2025 or 2026). 6) Missing mesh control carryover in Model tree Carrying over from the point just made about copying/duplicating analyses, it is quite sad that there is no mesh control carryover to the Nastran Model Tree, as this means that just about everything except meshing settings and mesh controls are carried over if you create a new (not duplicated) analysis, which sometimes can be quite cumbersome if you have a lot of mesh refinements (I had 5+ on a model the other day that I needed to remake one by one several times). Aside from that, it sometimes could be neat to just store different mesh control functions and just be able to swap them in and out easily without having to remake them every time. 7) Toggling visibility of component/part does not toggle its meshing This would really be a great addition, although I'm not sure of its feasibility (would maybe have to somehow link the part to its nodes, and make those disappear/appear). Anyhow, I cannot be the only person that has thought "oh I'd like to see what the hidden mesh/hidden face looks like behind this other component, let me toggle its visibility to see it", just to catch yourself staring at your reflection in your monitor as the mesh is still there. This would be great for inspection of the part pre-analysis, and obviously would also be great to have during the result processing, but I am aware that I am stretching it here. 8) Missing bearing loads to shell elements edges (suggested here) Currently, it is only possible to add a bearing load to shell surfaces, but it still isn't possible to apply bearing loads to circular edges on shells, which is a shame. 9) Change default direction of bearing loads Whenever you first want to define bearing loads, they default to "normal to surface", when it really should default to to component direction. It does not make sense that it defaults to normal direction, and can be an easy oversight that will lead to wrong results. 10) Bolted connection not needing a split face for input (suggested here) I've noticed that bolted connections have evolved quite a lot in Nastran over the years, which is a welcome addition. To improve them further, I'd suggest that selecting an edge should become the norm, without needing for a split face or any of that stuff. Chances are, if the edge I am selecting is 9mm in diameter, then I will be using an M8 bolt and nut (program puts in 9mm bolt, which is also fine), and the bolt and nut head sizes should (and are) already be known, so there should be no need for a split face. Additionally, the bolted connection should have an input for washer diameters, at which point the washers themselves can pretty much substitute the need for a split surface, as then that washer diameter can be used to generate, or "imagine" a split on the face at the location. 11) Inventor unusable whilst generating Nastran file/meshing, adding queue mesh (suggested here) When Nastran is meshing or generating a Nastran file, Inventor becomes unusable (quite literally disables any input until the task at hand is completed). It's already not a particularly smooth workflow when you have to do it once, but when you have to mesh & generate Nastran files for 4 separate files (like I had to do today) to run them through Nastran Editor, all whilst your Inventor is unusable, it really is a shame. On a more basic level, it would be great to still be able to use Inventor whilst these things are working away in the background.To make this workflow even smoother, I would suggest to add a function that allows the mesh generation to be queued with the start of the analysis/generation of the Nastran file, then run the analysis right away. So this means that instead of directly generating a mesh when you adjust the mesh settings, there could be a checkbox saying "queue mesh generation" or "mesh before analysis start" This is both useful for when users want to use the Inventor Editor to then run the analysis, as well as just not having to do two inputs (which can be quite some time apart) 12) Reports missing many features (suggested here, here and here) Unfortunately, currently the report feature of Nastran is only useful to get a nice set of screenshots of the results of the model. However, the lack of report customization is a shame, because I'm sure many users would find this a really important or useful feature if it was further developed. Features that are requested (in order of importance, to me): Generating reports in other file formats (which are not not HTML) Being able to toggle on/off sections of the report (look at picture below from another thread) Being able to choose which graphs/pictures are included in the report Currently this is a big one for me, because I can only ever use ~40-60% of the screenshots that Nastran outputs, considering some of them are irrelevant, empty or not useful to my report If multiple subcases are present in the analysis, generate separate sections for each subcase, with their respective loads and constraints present in that section Add material names from the Nastran/Inventor library into the report. Why should they be called ID1 and ID2? Generate tables which show the placement and type of constraints/loads Bonus: Midsurface is missing features of the split face (suggested here) I recognise that this one is likely more cumbersome to implement than others, which is why I'm putting it as a bonus. Own experience: This issue also occurs when there are split faces on the solids, and then the Solid model is converted to a Surface model inside Inventor (not in the inCAD environment). This suggests that it is a feature (or quirk?) of Inventor. Currently the only way to fix this is to do a full body split, then make sure that the (mid)surfaces are either held together by continuous meshing, or otherwise offset bonded contacts. It would be nice if the splits would be included in the midsurface (maybe with a checkbox that could be toggled in the midsurface command). From thread: Faces of a solid are often split so that loads and constraints can be applied to a portion of the face. If the solid is converted to a Midsurface in the In-CAD environment, so that it can be properly modelled with shell elements, the splits do not appear on the midsurface. In the example below, feature 1 is split on the top and bottom faces, feature 2 is on the bottom face only, and feature 3 is on the top face only. None of these three splits appear in the midsurface created by In-CAD (bottom half of the figure). The midsurface is just one face. I would love to see in the comments what people think about these features. If you'd like to post a comment below showing support for a specific feature, or to make a comment on it, please include the number of the suggestion!
Show More
