Best practice with cloud data import

[temo]

Hello

So just a quick question: I've some table of half breaths for a boat hull that I have put into Excel in a three column style(x, y and Z). I want to import the coordinates into Solidworks as points, so I pasted all the data into a comma dilibert text file, and it seems to import fine into solidworks. But the question are, what's the best way to go from here, as it's only half breaths, all the points need to be mirrored about the length of the hull. Would it maybe be better to do the job in Excel, just copy all the points and manually give all the needed point in the Y-direction a negative sign before importing. Would there be any pros or cons on how I can work with the point cloud in SW, I need to add some outer skin, some decking and some interior later?

Terje

[Frederick_Law]

I'll model half in SW and mirror.
Not sure if all those extra points will slow down SW or not.

[temo]

No, not me either, never used point cloud data before

[bnemec]

Is it truly random points as in scanned data? Or do you have groups of Y-Z (width and height) points for planes along the X (length)? As in longerons (or bulkheads? not sure what they're called in ship building). If the data is grouped then you might try several section sketches then connect those profiles with loft(s)?

I've tried this with airplane fuselage cross-sections and turns out ok except for the ends and transitions got tough for me. ...and I wasn't going to try to actually build from it either.

[temo]

It's like you say, bulkheads, not random points.
So it's like 5x10 points for the main hull and about the same for each flot as it's a trimaran, and it will of course doble when you mirror, either the points or the body parts.

I think maybe it's so few point that it should be doable?

[bnemec]

It's like you say, bulkheads, not random points.
So it's like 5x10 points for the main hull and about the same for each flot as it's a trimaran, and it will of course doble when you mirror, either the points or the body parts.

I think maybe it's so few point that it should be doable?

There are others here that know more about this workflow than I, but I would try this:
1) lay out planes, offset from front plane (Front plane would be a datum, in planes it's often the firewall or whatever point the weight and balance is figured from, I'd imagine similar for boats.) one plane for each bulkhead that you have a cross section for.
2) start a sketch on the plane and place sketch points according to your spread sheet, since there aren't that many points I'd just dimension them each by hand. Might be worth importing from excel if you like that type of thing, depends on your needs.
3) connect the points with a sketch, spline if need be? This is where I'm not sure the best.
4) repeat steps 2 and 3 for all the bulkheads for a hull.
5) if needed add guide curves front to back.
6) loft a surface through the bulk head sketches and along the guide curves.

Don't mirror the points. When mirroring the most stable and efficient is to mirror body(s) so get the lofted surface watertight and stitched to a solid body sooner than later IMO. Also, if the right hull is a mirror of the left then I'd get one half of one side as a solid, then mirror it. Then mirror that about center plane to get the other hull.

[temo]

Thanks, yes I think you have a point here, I have all the points in an matrise so it should not be to much work to key in the points, especially as the plane would represent the Z-coordinate.

I've always wondered about this, in what order to use mirror, from simple lines in the sketch to features and solid bodies, I'll try your approach.