help with boat design thicken/general improvements
If you are having Thicken (or Offset, or Shell) errors, the first thing to look at is surface curvature. A most common failure of these kinds of surface offset failures is areas of high curvature. A Gaussian curvature analysis can usually help point those out. Here is an example, from the surface before the Thicken in your model:
You can see that there are some areas of high curvature near the bow. So, my first guess is that these are likely to be problem areas if you get a Thicken failure.
Second, I notice that you make heavy use of Boundary Patch here. It is not bad, by itself, to use patches, but this feature has been known to generate some problematic surfaces at times, especially around the boundaries. This can require tolerance when stitching, that can lead to downstream failures such as in Thicken. I did not inspect your Stitch features, but do you recall having to increase the tolerance when stitching? If so, that is never a great sign. There are lots of threads here on boat design. I am in no way an expert, but I've noticed that most people use Loft for this. Fun fact: The Loft feature gets its name from boat design: Lofting a Boat . It might be worth exploring that method - loft tends to produce high quality surfaces. Some of your patches can be done with Loft (any 4-sided patch can be replaced with a loft with rails), but you can also loft the entire hull all in one go.
You mention "rails" failures, which I usually associate with Loft, but I do not see any Loft features in your design. Can you clarify what you meant here?
Jeff Strater
Engineering Director
Not going to happen. First time forming, second time reporting. There's rules to follow when it comes to Fusion. Most of the time is switching directives and changing the approach. From basic model, to just fast modeling.
I can certainly understand the frustration. You're talking about sometimes waiting for the computation to complete. And, find out it's an error. I get confused too. Sometimes it's a shape of a square. And, the computer takes an hour to produce an error for a chamfer. For example.
But, I can just dive deep why your model will not work here. It's not too complex in the design. I get the history and the highlights of the design.
The design is now, just dirty.
First while cutting corners with the program.
The two rigid lines on the outside are mirrored. And separated. Then they also bow and then bend.
The last thing to worry about. They are lower than the rigid lines on the inside.
This happens to be a problem for that exact tool. Thicken or offset. Because the outside point down, will become an inside large portion corner facing up. And, the exact opposite will happen on the inside. One going down other going up. They collide. You get a hole, instead of thickness. That's what's not happening.
The last part. I found. Is really not going to be fixed. If I split the body in the middle. You got this part in the front that has inverted sides. It's like an S shape. Same, thing happens here. They collide in opposite directions, for a hole.
Going back to your model. I would go back before that Reverse Normal. I have no idea who told you to do that there. But, I actually would fill the back bracket with reverse normal. You're doing the wrong side once again. That's why the entire inline is also like a letter S.
And, After a terrible trial. I have to say. I wouldn't fix this for abduction. Instead just do something else that works. Like this. for example. Because. Let's face it. I don't really see industrial potential design here. And, why not actually add proper and may I add working webbings to the 3D print. You can hollow this as this as 60mm tested.
@raptorhelicopter wrote:
I don't quite understand everything you are saying but I'll try to figure it out.
Me neither and the part ! do understand is technically incorrect!
Listen to @jeff_strater's advice!
1,. Your rails in several places to not align with the profiles.
2. Use loft, instead of patch.
If you go through the timeline of the attached model you might be bale to determine how this was built, but I am sure you'll have more questions!
Okay maybe I can try to explain this problem with accordance to NASA's comprehensive anthology 624.
Here I got a hat.
If I turn this hat inside out. I find seems on the inside. Which is appropriate most of the time.
I'm trying to explain what would happen if try to turn a crowded cylinder inside out using computer logic.
Here's an example image. The yellow and turquoise walls must switch sides inside out. And, one of them either must gets larger to stretch around the other, or one of them must get smaller to fit inside the other.
The computer logic will just switch them in size. But, even the computer will add stress points. Like here in red. And, just the same. Either of the two, They either flip inside out, or don't move at all. Just, get yourself some playdough or paper mache to play with before you attempt building logic.
Let me maybe dive in a little further why the boat is not going to happen. Because as much as you try to make it perfect you seem to misunderstand where your model surface cuts itself. And, where it stops being perfect.
@Anonymous wrote:
Let me maybe dive in a little further why the boat is not going to happen.
Maybe you could download the model I had attached and see how it already has happened. At least digitally 😉
@TheCADWhisperer wrote:
@raptorhelicopter wrote:
Sorry, I have zero idea what you are on about buddy.
That makes two of us.
Three 😁
@Anonymous I am going to ask you again to maintain professional decorum and to refrain from insults and personal attacks. We will not tolerate that on this forum!