Hi everyone,
I’m wondering if I could get some help reproducing the sketch I’ve attached, please?
The cross-section of each beam is rectangular and my main issue is I don’t know how to scale the cross-section of the extrusion along each curve so the maximum thickness is at the center and minimum thickness at either end. I’d appreciate any help with doing this, thanks!
Along with scaling beam cross-sections, do you also want to spread the “rail” curves apart in the middle?
Are the beams flat or higher in the middle, low at the ends?
Hi Joseph, thanks for the quick response. Yes, that is what I would like to happen and would tweening between two rails do this?
Yes, this is something I should have mentioned - I would like the beams higher in the middle and tapering toward the ends. Thanks
Also, In the image the “rail” beams are wider and taller than the others. Is there a way to individually scale these beams to allow for this? Thanks
Yes.
There’s almost always a way. It would be better for all concerned if you provide some geometry yourself, either as a Rhino file or internalized in a Grasshopper file. The rail curves at the very least (one is a mirror of the other), including their arch shape in “front view”. And some detailed guidance about how much wider the rail beams get in the middle? Do all the other beams have the same cross-section?
Thanks Joseph, I’ll get a 3D file uploaded asap.
Here is a sketch using one rail curve as input. There is no attempt to make the rail beams wider or add the cross-wise beams shown in your sketch.
How will you build this? Laminated wood beams?
P.S. Add Cap Holes after Sweep1.
Here is an effort at making the rail beams wider but notice how the gap gets narrower? And the beam ends get wider, not just the middle. You probably don’t want that but this is where things get tricky… Always in the details.
P.S. I just noticed that you want the rail beams taller as well as wider? Because of the pFrames orientation, that means wiring to the ‘X’ input of Scale NU as well as ‘Y’. But oh dear, the rail beams get taller on their tops instead of bottoms, which you probably don’t want either?
This can be fixed by adding “-x” as an expression on the ‘Height’ slider.
Thanks for the replies Joseph & HS_Kim. I’ll dive into these definitions shortly and like you said it is very much about the details! I think the desirable outcome for these definitions is to have as much control over the paremeters as possible.
Some context around what I’m designing - It will be part of a watch as you can see in the sketch I’ve attached. This sketch is purely an illustration of intent with little attention being paid to tolerances, etc. This will most likely be a milled piece but depending on the complexity of the final part, this may change. Thanks, K.
Well yeah, that is always the goal, depending on one’s skill level… Getting from illustrations like these you have posted to actual geometry can be tricky. It looks like the tween “beams” vary in width? (not all the same at the ends?)
Get back to us when you have some Rhino/GH geometry to work with. Cheers.
Hi Joseph,
Hope you had pleasant holidays.
I’ve been working on some geometry for the watch and have made ok progress. I’ve been following a GH tutorial for the design of a stadium to get a better understanding of ways to control the various parameters of the watch. I have, however, passed the point of my working knowledge of GH I would appreciate some help with fixing these issues:
I’ve played with the definition some more and had better results, although it’s still clumsy. I can’t figure out how to properly guide the horizontal beams with the guide rails and also make the rails flush on the underside. The horizontal beams won’t be as thick or tall as the vertical beams. For some reason, the distance between the horizontal beams enlarges as you head toward the ends, is there a reason for this? best, Kieran
GH Help.gh (55.7 KB)
I used “Tween through curves along curves” and this solved the “guide rail” issue. I’m still not sure how to go about matching the curvature of the horizontal and vertical beams…
Happy New Year,
This is becoming a big, complex model now so it’s difficult to give a comprehensive response… I can see two problems with using TweenCrv here, both for the same reason: tween curves are linear in position and length whereas your frame is likely curved in both directions.
My first attempt at making the transverse beams (white group) divides four curves to get 16 points each, so they are the right length (sort of) but not the same shape/position to match the lengthwise beams, which suffer from the same tween effect, causing their tops to follow flat lines from middle to edges.
Even explaining this is difficult, let alone fixing so many issues and details.
The transverse beams do not go straight across so the method I used is pointless.
The rest of your questions about varying beam width are premature. The curvature in both directions must be solved first. As time and attention permit, I’ll try to suggest something better.
P.S. On second though, after glancing at all the work you’ve done in this BIG model, I’m not sure I want to modify it further. At best, I might suggest a different approach, perhaps by starting with an underlying surface curved in both directions.
Thank you for the time you’ve spent helping me, I appriciate it. I understand the issue you’re explaining and I’ll attempt to figure out a suitable solution. I wondered if you help me witht the issue of being able to control the two guide curves so I can better match the sketch I posted earlier, please? Best, Kieran
The length of the GH model is correct and the sketch length is exagerated FYI
The cyan group below) uses rail and station curves I found in your model to create a surface that should be the basis for beams in both directions.
As to shaping the rails, remember that like a boat hull, this is symmetrical so you only need one edge “rail” and can mirror the other one. There is really too much stuff in this model for me to mess with it much further though, sorry. Shape the two rails so the surface has the shape you want first.
Thanks for your help, Joseph. I’ll post any progress I make. best, Kieran
This became absurdly complicated, in part because I used points you already had as much as possible instead of restarting from scratch. In the end, I discarded everything that wasn’t being used. No effort at all to make the beams but you can see where I’m headed by looking at the Isotrim (SubSrf) surface. It’s not quite the shape you want but maybe it will help you get there?
Instead of Isotrim, you can use Iso Curve to get the same division lines to use as the basis for the beams. Staring over to make it simpler and more flexible about the shape is strongly advised.
