January 11, 2011 at 22:38 #34556
I’ve been given the task of modelling an old boat which will be undergoing a large reconstruction. It will be added some additional length and therefore I need to model the existing state to see how much and where the existing hull will need to be cut.
The WL and BT lines are given in an att. picture.
I have the WL, BT and sections in 2d drawing in Autocad, but since there is no direct link from it, I’m trying to compose an offset table to enter the data in Delftship, since it seems the most direct way.
However, I’m encountering a problem with the round transom – I’m not sure whether the offset table haven’t predicted for such an option, or have I understood it’s format wrong, but I don’t see how to enter the transom data in it.
I am new to Delftship so I’m hoping that it is the second case 🙂
Can anyone give an advice on this or tell me is there some other way to enter numeric data in Delftship, which offers a little more flexibility in terms of waterlines ending / creating custom curves?
January 11, 2011 at 22:42 #34558
January 12, 2011 at 00:18 #34559
I’m not sure if I got that correctly – you managed to get the Offset Data in Freeship, but the transom is not like you hoped it to be?
if that is the case – what is it like?
(I just tried exporting and importing offset data of a ship I had once made out of data of an old paper sheet with offsets – almost similar transom like yours. the chine of the transom was there, but the transom stern plate was left open. By extruding the edge I could easily close it)
January 12, 2011 at 00:53 #34560
Hello Maximilian, thanks for answering!
Uhmm, I’m not using Freeship, but Delftship 4.03.something – don’t know if that makes any difference.
But that aside, … not actually. What I managed properly is to get the offets of the waterlines and the decklines, but I don’t know how to get the round transom at all.
In other words, I have the centerline, the decklines (which branch at the centerline and go towards the back to the sides of the transom). I don’t know how to create “the curve of the transom” at all.
Upon reading the above, I realize it is confusingly put, but I’m not sure how to describe it better.
Could you perhaps explain how did you model your ship? Maybe I’m doing something fundamentally wrong.
January 12, 2011 at 07:05 #34561
Oh sorry, I put the wrong name, of course I tried that in Delftship 4.27 …
Now as you have the offsets imported into Delftship you should more or less have a hull. Maybe you could post a picture of your current state (perspektive view with control grid (red lines) turned on).
I’m surprised that you seem to have just one curve – the WL – for definig the deadrise. If you stick to that you will get very poor accuracy – why don’t you use more in several heights?
Anyway, I try to describe what I do for a transom stern … Move Control Points to the place where the edge of the transom should be, make lines (named edges) between these points, select and change them to creased edges (knuckle lines), make sure that there are lines inside the transom (if not I select the boundary ie. the chine of the transom and extrude it) move one point inside the transom per point of the transom chine to the CL (0). Now there should be a closed surface – clean it by removing edges (not with delete, that would couse holes again!) to get a clean surface.
One last word: If you need accuracy you should consider importing the profile and plan view of this ship (you can use it as it is in your attachement – just scale it to the right size in Delftship itself) and work with buttocks and waterlines. That’s what I would do.
greetings from Austria,
January 14, 2011 at 14:03 #34565HFParticipant
In my experience it’s a long try to use importing an offsett table to model in Delftship.
I would suggest you two different opinions.
1) Import your linesplan as an background image and use this to image to model the ship
2) create markerlines from your linesplan, import these in to Delftship and use these to model the surface.
In my opinion you should use an separate surface for the transom. How to do that, is something you can find in the Delfship tutorial about recreating linesplans.
January 14, 2011 at 17:32 #34567MarvenKeymaster
That’s indeed the best way to get a good 3d model.
Especially since offsets in general will not describe a round transom accurate enough to produce good results.
January 31, 2011 at 15:55 #34612
Hello Hansfokke, thanks for answering!
First, let me offer my apologies to everyone it took me this long to reply to your answers; I was on a trip and I don’t have Delftship on my travel notebook – only at work at my station).
Yes, I agree with the linesplan approach. It is the approach I usually take when modelling a new hull (new design).
Unfortunatelly, in this case I have a linesplan of a ship which will be undergoing reconstruction soon. It will be lengthened at the stern, and the “new hull” (the new part of the hull) is attached to the old one at a rather difficult place (in terms of faired geometry). So the modelled hull needs to be really well modelled at that segment since from it we will be directly taking measures from cutting plates and frames.
For that reason, at the place in question we made measurements at different sections/frames, and now have x,y,z data for those sections. That’s why in this case I’m trying to push the approach via sections offsets.
Btw, I didn’t understand your second part – what do you mean by “markerlines”? (it is probably a language issue on my side).
February 9, 2011 at 23:30 #34656
I think offsets are a way of getting the accurate (measured) hull shape if you have LOADS of offset points (I would say >10 000). But then you would still have trobule with the round transom – it would be necessary to draw it by hand anyway.
If you have a smaller number of measuring points I would rather draw the model the usual way (linesplan) to have a clearly arranged basis and afterwards compare/correct at each station (exactly where measuring points of the real ship were taken) to the values of each offset point. The ability of delft to define single stations (f.e. in x-direction) gives very good possibities to correct y and z position manually to perfect fit. From an economic point of view correcting the sorrounding area of the attachement place should be sufficient. For bow sections +/- 5 cm accuracy should work fine, near the attachement site reaching an accuracy of +/- 2 mm should be possible.
To point it out clearly: The main reason for doing it this way is that you have to take full control of every single station outline – only having the control net points at an exact measured position does not mean the station outline will be going there, too. In my opinion moving the staion outline to the perfect place is much easier to maintain with a clearly arranged control net.
- You must be logged in to reply to this topic.
Hull modeling from offset table
5 days, 3 hours ago
import excel to delftship and surface became twisted
5 days, 3 hours ago
Tank Modeling DTL file Structure
6 days, 23 hours ago
Importing CSV/TXT coordinates as stations?
1 week ago
starting at a blank page.
3 weeks, 2 days ago
Complete Newbie Needs Help Starting Out
1 month, 2 weeks ago
Corner points appearing in model- need help
1 month, 3 weeks ago
Beginner seeks tutor
2 months, 1 week ago
Lines Plan – Header on pdf
by Anneler Ruedi
3 months, 3 weeks ago
3 months, 3 weeks ago