• This forum is the machine-generated translation of www.cad3d.it/forum1 - the Italian design community. Several terms are not translated correctly.

loft o rete di whores?

  • Thread starter Thread starter Panawood
  • Start date Start date

Panawood

Guest
Good evening, everyone.
I find myself having to faithfully redesign the hull of a 700 vessel and possess the original water lines and the various sections.
according to you the best choice is loft or curve network?
I have already done some experiments with the network of curves, but the extreme variety of forms does not allow me to achieve reliable results.
Hello jacomo
 
Panawood said:
Good evening, everyone.
I find myself having to faithfully redesign the hull of a 700 vessel and possess the original water lines and the various sections.
according to you the best choice is loft or curve network?
I have already done some experiments with the network of curves, but the extreme variety of forms does not allow me to achieve reliable results.
Hello jacomo
Click to expand...
I'm not an expert yet. I'm trying to give you my help.
I think I have understood that the lines that you own form a network, so try with the srfretecrv command trying to get between the curves and the surfaces a good continuity.
regarding the loft for this case I recommend the strict style mode, since the surface faithfully follows the curve path.
If you were able to post an image, he judged to understand the position of the curves could be of help.
Hi.
 
or take a systematic series model (I start from a npl) and move the vertices to sew it on the curves you want.
in this it helps you to have dynamic sections.

You lose us a little time if you don't have your hand, but the result is egregious
 

Attachments

  • Caicco.webp
    Caicco.webp
    62.7 KB · Views: 30
Good morning, first of all thanks stone and fulvio.
I suppose "npl" is for cloud of points. Is there a method to import the numerical table with the spatial coordinates or does it not work like this?
I attach the drawings I have available to which I have already drawn the lines in dwg.
for the numerical table should I create it by transcripting the coordinates one by one?
precise that my intent is to create a solid formed from the outer shell and from the inner one to "slice" subsequently so as to regain the original ossuage complete of the corners of fourth-bone.
Hello jacomoplano01 .webp
plano02.webp
 
no, npl is a systematic series of a certain type of carene, but it is not important.

If you take a round hull model already made, you can move the nurbs control points so as to sew it on your sections. If you want me to put you the model of my starting hull, but I can't do it before tonight.
 
Thank you very much, whenever you want and here you can.
You think you can manipulate a sample model like yours and adapt it to a sailing round ship with that curved bow?
Hello jacomo
 
Panawood said:
You think you can manipulate a sample model like yours and adapt it to a sailing round ship with that curved bow?
Hello jacomo
Click to expand...
the image I posted before is a Turkish caicco, round hull, wood keel. I made it from the model I'll post you tonight
 
but it is beautiful, it seems to work with the rubber.
And that doesn't even tear.
Now, if I need 27 horizontal and 15 vertical sections, should I add them to the model, highlight its control points and then sew them on my lines?
Is it okay if the points get them with the multiple sections command?
 
attention...do not add isoparametrics if they are not necessary. project curves over the surface, then move a little all the tops so as to overlay the sections to those you want to get.

Besides, do not move one point at a time, select them to groups and scale them in the various directions, you can shape the surface much better.

At least, this is my way of working, but it is not said to be the best.
 
hi fulvio, after a bit of tests (tante) I arrived at this result.
But what does not satisfy me is precision.
That attached is a grade 5, but I think it's too much.
in addition I can not combine the two surfaces, hull and bow (the stern is not there yet)
 

Attachments

the carena has several bugs, it is not a great deal of work. probably leaves from an unperfect construction plan. do not try to go too close to the sections.
couldn't you use my file again?
the bow section because it does that strange curve?
the bow surface is not in tangency, it makes a maximum angle of about 5° at half height. It's complicated to do it in two pieces. I have made very clean surfaces, or a surface so complex even if you put it in tangence will never be "nice".

If you tell me what you need the model, I try on the weekend to do something about it.. .
 
Hi, the two surfaces you see are the two npls I added lines, which you sent me.
Surely the construction plan is not perfect, I retalied it from documents of the era.
the purpose is to obtain a solid to be dissected later and then to reconstruct the original ossuage.
It is the duc de duras, a merchant of the 18th century, converted into a warship.
I should reconstruct it with a maximum deviation of 2mm (which in 1:48 scale are about 10 cm in reality) from the original planes and I would keep the tolerance to "lisciate" the surface.
 
By the way, the upper part of the bow is so because I extended the height lines to have a upper edge of the wall, straight.
Actually the upper part of the bow will be eliminated about 230 mm. up
 
As promised, I put a little hand on the weekend.

I used a single surface like the one I posted earlier and did not add isoparametrics. the surface is clean enough, but it does not approach the surfaces very much.
parts from this (which was the first operation to do). now the next steps are:

- without adding isoparametrics, move the points so as to approach as much as possible to the sections.
- in making the previous point, make sure the trend of the points is regular*
- add isoparametrics only where necessary, to approach even more to the sections
- do not strive to approach the sections that have a "strange" pattern compared to others. can be a relief mistake.

(*)
For example, vertical or horizontal point groups should not do strange zig zag. often moving for groups vertical ones, horizontal groups make strange jokes, and vice versa.

I hope I was useful/clear. I have to run away...
 

Attachments

bye,
First of all, I thank you for the intersection (at the first occasion I will offer you a pint of rhum)
I tried to "tortutrate" your model, but I can't get a model in tolerance, in some points it is even 4 mm.
At this point I wonder if it is best to demand a perfectly smooth surface but out of quota or a bit of a bugged surface but in tolerance.
By the way, is the deviation indicated in the "analyze" menu expressed in measurement units?
 
depends on what you have to do. In general, I would strongly recommend that you have a "nice" surface, clean and quiet, even if you break away a bit from a hand-made building plan (although well I think) and then recalcate.
the indicated deviation is expressed in the set measuring unit, but what is it? I wasn't with the distance you said.

However, do not exaggerate with the insertion of new isoparametrics, and play with points. you should be able to get closer to the sections. a 4mm discord, if it is an aesthetic model, it is more than good. rather after made the model, make a series of sections and water lines to see if the surface is really beautiful
 

Forum statistics

Threads
44,997
Messages
339,767
Members
4
Latest member
ibt

Members online

No members online now.
Total: 0 (members: 0, guests: 0)
ciao
Back
Top