I have watched countless tutorials but they all have the same problem. In all of them, ribs are oriented on an axis or in some cases in a rotation from a single point.
In this case i have to make a waffle structure that follows the curves like in the picture above and the ribs have to be straight to be cut from a metal sheet.
Is it possible to make a script that would work for different types of design or for every design there will be a lot of changes in the script?
The idea is to make structural elements/ribs in Rhino, laser/plasma cut them from metal and then have a waffle style metal construction for further work.
I see many problems with this⌠The contour approach above has all the frames parallel, square with âWorld XYâ, yet thatâs not what I see in the first image in this thread. (I donât see how the 2nd and 3rd images are relevant at all?)
So I created a similar shape using code from this thread. This is a surface, not an arbitrary brep or SubD, which is important because I found that isocurves make life easier in getting cross-sections (perhaps I gave up too quickly?).
Isocurves are not planar, which is required for Offset Curve, so I fudged them by creating planes from each one (Plane 3Pt) and projecting to them. So itâs possible to extrude the cross-section âribsâ for thickness but not so with the lengthwise ribs.
RSlits might be ideal for notching the two sets of ribs but I couldnât get it to work.
P.S. Thickness of lengthwise ribs was easy using Offset Surface. I started work on cutting notches (including gray group) but that isnât easy so quit - for now.
Rslits works well if you have a clear overlap between the two sets of intersecting entities . For your project, Iâd extend the longitudinal stringers outwards through your skin and then run the Rslits command, then trim the longt. surfaces back again and thicken up.
What youâre trying to create here is yacht structure⌠but with the transverse ribs being perpendicular to a centre spline as well, which whoever has the job of setting this out on site will hate you forâŚ
Iâd suggest making their life easier by simplifying the central spine line to a series of straight lines and then putting your transverse ribs in perpendicular to this control in groups. This breaks your structure down in to units that can be plated neatly (either metal or plywood will have limitations on plate size to nest in to) and are easier to dimensionally control during construction - you have maybe 5 points to set out to define a baseline that measurements are taken perpendicular to for the feet of the ribs.
The longitudinal stringers setout looks lovely, but beware of ending up with either stringers becoming too crowded or panel sizes becoming too large to pass structural requirements.
If youâre cutting it in plywood, beware which way you offset your ribs - if you define the cross section through your master surface as the mid plane, then part of the rib material is going to be intersecting the skin, so you either need to ensure all your thickness is offset âuphillâ if youâre using a 2D laser cutter, or define the chamfers you need to allow the skin to sit right if you are using 3D routing. Itâs amazing how small a gap between the rib and skin will be visible in the fairing of the outer surface.
Unfortunately, thereâs a lot of detail goes in to getting a rib and stringer setup that is accurate, well engineered, has the details that the folks on the shop floor need and makes economic use of materials. Grasshopper gets you maybe 50% of the way there and gives the architect flexibility on design of the external skin until a bit later in to the project, but having confidence in your design and locking off the primary skin and rib location early will make you popular downstream and allow the project to come in on budgetâŚ
Fascinating to see the cross over between architecture and boatbuilding.
Best of luck
Deri
Not quite. A boat is symmetrical with a planar centerline - much easier!
In my experience, ideas happen at a far faster rate than the work to implement/try themâŚ
I got something that is close⌠Enough to see some obvious flaws. The subtle flaws are harder to find.
This became intensely complicated Not a beginner project at all. It takes more than a minute after opening the file to see code and geometry.
Some notes:
I used two clusters. Rib that thickens transverse and lengthwise âribsâ and cBranch that culls the first and last branches of data trees. It has an âRâ output that returns the culled branches so they can be merged later. For several reasons, it seemed reasonable to avoid notching the first and last ribs in both directions.
The color clusters at the far right can show the merged ribs or not, depending on how they are wired. If not, there are two orange groups that can reveal the culled ribs.
Much can be said about this project but itâs late so Iâll just post the code.
Perhaps the most obvious flaw is the lengthwise ribs protruding through the transverse ribs at each end. I fixed that by adding a gray group at the bottom to cut them off and cap them.
Less obvious are the two lengthwise ribs at the bottom on each side (left and right) but I left them alone because a proper fix might entail horizontal footings of some kind?
I tried but donât know for sure how well the notches match where lengthwise and transverse ribs meet? Related but more subtle are the notch cuts where ribs meet at an angle:
I considered adding a Data Dam to keep the very slow red group from being activated when parameters are changed but wasnât sure what to show instead? Many details to consider on a project this complex
Not a beginner project at all. It takes more than a minute after opening the file to see code and geometry.
