Export Offsets again)

DELFTship forum General Questions Export Offsets again)

Viewing 11 reply threads
  • Author
    Posts
    • #41223
      john truelove
      Participant

      Hi DELFTshipers

      I have two problems

      Problem #1 – exporting offsets

      1. This seems to be a recurring issue going back to 10 or thereabouts, and my perusal of the many posts has not shed any light on my problem
      2. When I exported offsets the first time (Export Offsets 1) I got a useable table of offsets. I do not know if they were correct or not.
      3. I wanted more WLs, so added WL intersections (including at z=0) and repeated (Export Offsets 2). This time I got an offset table dominated by “1.300”, which was wrong.
      4. I re-entered the “intersections” screen (several times and with varying waterlines) but even with the original WLs I now got an offset table dominated by “1.300”.
      5. I exit and reentered DELFTship then reload the model that I am working on – no joy: the offset table is dominated by … 1.300
      6. As far as I can tell, the model is unchanged.
      7. So I pull out the lap top and fiddle with that – same model but an independent file – same result – initially ok but then full of – you guessed it – “1.300”.
      8. I concluded that somehow the models have been corrupted, and put a read only attribute on an older version. This seems to have quaranteened it from corruption ….

      Problem #2 – exporting offsets II

      When the offset table appears “sensible”I am finding impossible offsets – negative half breadths where they just cannot be …

      Both these are enormously frustrating to me (a retired battens and weights Naval Architect having a second childhood) but the models are designs for ships that are to be/might

      be /could be built (and there are others in planning). I had hoped to use ship design software for concept-to-cut steel (or at least concep-to-lofting) and chose DELFTship because

      the free version would give me a long enough trial to get happy with it.

      DELFTship is brilliant, and I very much want to continue with it, but …

      Has anyone got any ideas please.

      (“Ships” under the Australian Navigation Act of 1912 include all vessels not propelled by oars!!)

      John

      (FYI, I do not date from 1912)

      qqq

      Attachments:
    • #42889
      Marven
      Keymaster

      Hi John,

      Re problem 1:
      The recurring 1.300 value looks very strange indeed. Can you please send us the DELFTship model which you have used to create this file so we can use it to debug the software?

      Re problem 2:
      I suspect this issue is related to the first. We’ll investigate this as soon as we have received your file so we can address this issue before the next update.

    • #43532
      John Owles
      Participant

      From time to time I have a go with offsets as I could really use software that can read them.

      So far I have had no joy.

      I tried again today by loading up the attached model “14 footer.fbm”, exporting a table of offsets, “14 footer.txt”, and then importing the resulting table of offsets into to a new blank poject. The result was, again, not good, as can be seen in “14 footer from offsets.fbm”

      The attached files may assist with debugging.

      Here’s hoping. I am very much looking forward to the day when DS is able to work with offsets.

      Attachments:
    • #43534
      John Owles
      Participant

      It seems that fbm files are not allowed, which is a bit tricky when they are crucial to the current excersize.

      I see another post which said that fbm file were allowable. Did that go wrong?

      Please let me know how to upload my example fbm files

    • #43538
      Maarten
      Keymaster

      It seems that fbm files are not allowed, which is a bit tricky when they are crucial to the current excersize.

      I see another post which said that fbm file were allowable. Did that go wrong?

      Please let me know how to upload my example fbm files

      There was an issue with mime type definition- I’ve just tested adding .fbm files- it should work now.

    • #43539
      John Owles
      Participant

      Thank you Maarten,

      I’ll try uploading the examples again. I hope they are helpful.

      Attachments:
      • #43543
        Marven
        Keymaster

        Hi John,

        Thanks for sharing your files with us. The problem with a table of offsets is that it contains only a limited number of points,  the intersection points of stations and waterlines. For the mid ship area this  is mostly just fine, but for the (curved) ends of the vessel it means a lot of information required to build a 3d model is missing, such as where a station (or water line) ends at the stem or keel for example. Traditionally this information was filled in by the loftsman who recreated the linesplan on the lofting floor. This lack of information can sometimes be fixed by adding more stations/waterlines.

        Another problem is that a sheer line with high curvature means that waterlines close to deck level are split in 2 (aft and forward parts). This is not captured correctly by a table of offsets.

        In other words, tables of offsets are often not very suitable to creating 3D models. Other options (such “import surface” or “import chines”) are better suited to the task.

    • #43544
      John Owles
      Participant

      Hi Maarten,

      Thank you for your reply. Lofting is part, from time to time, part of my day job. Particularly for the restoration or re-build of old vessels. In fact I have just lofted a 30ft WWII airborne lifeboat, which was slung under a Lancaster bomber and dropped by parachute to rescue downed airmen off the Dutch Coast or Biscay, etc.

      Lofting from a table of offsets drawn up by a designer of the caliber of Uffa Fox is very easy and straight forward but I do take your point.

      Perhaps the answer is to take a look at the problem from another angle. Perhaps starting with an original table of offsets and working towards the software coding.

      I shall think on the problem, using the ABL as an example, and, if you don’t mind, let you know if I think there is some kind of solution.

      I tend to be like a dog with a bone, I’ll whittle around a problem with the feeling that there is no such thing as can’t, just how.

      John

    • #43587
      John Owles
      Participant

      Hello Maarten,

      I have been away for a couple of weeks and have only just been able to return to this problem of importing a table of offsets and I now think that I may have the answer.

      I have imported the table of offsets, for the Airborne Lifeboat project that I am currently engaged with, into DS and it has made a very reasonable fist of the process, as far as the topsides are concerned. Unfortunately, DS has ignored everything below the waterline entirely.

      As I have said, ‘old school’ lofting is a part of my day job and I beg to differ slightly with your assertion that “The problem with a table of offsets is that it contains only a limited number of points, the intersection points of stations and waterlines. For the mid ship area this is mostly just fine, but for the (curved) ends of the vessel it means a lot of information required to build a 3d model is missing, such as where a station (or water line) ends at the stem or keel for example. Traditionally this information was filled in by the loftsman who recreated the linesplan on the lofting floor. This lack of information can sometimes be fixed by adding more stations/waterlines.”

      This is correct as far as it goes but you failed to consider the buttock co-ordinates, without which the loftsman would, in many cases, struggle to obtain the correct underwater shape of a vessel. Even in the days of working from half models, they often used two models. One sliced horizontally at the waterlines and another sliced vertically at the buttock lines.

      Thus a true table of offsets contains the intersection of points from stations, waterlines, buttocks and diagonals. The diagonals are a final checksum for ultimate accuracy and, for the purpose of this exercise, I don’t think we need concern ourselves with these.

      However, I do think that the problem that DS has interpreting a table of offsets below the waterline, is the lack of buttock lines. Would it be an idea to include buttock lines into the import routine? It may make all the difference.

      I have attached the DS result file 43588, along with the Table of Offsets and a jpg image of the original lines plan that I am working from.

      Best regards

      John

      ABL-MkII_Offsets-DS

      Original Lines Plan

      ABL-MkII_Offsets-to-DS

      Attachments:
    • #43594
      John Owles
      Participant

      From the silence, do I take it that you don’t think my suggestion will work?

      • This reply was modified 1 month, 2 weeks ago by John Owles. Reason: typo
    • #43596
      Marven
      Keymaster

      This is correct as far as it goes but you failed to consider the buttock co-ordinates, without which the loftsman would, in many cases, struggle to obtain the correct underwater shape of a vessel. Even in the days of working from half models, they often used two models. One sliced horizontally at the waterlines and another sliced vertically at the buttock lines.

      Hi John,

       

      You are correct. By introducing buttocks in addition to waterlines and stations you will end up with quite a few triangular and/or 5 sided faces. This will make it very hard to make a fair hull surface, that’s why for most hull shapes this method won’t work.

      I was looking at the Uffa Fox lines plan and noticed it has a canoe stern and (near) vertical bow. Have you considered using just the geometry of the stations by using the “Import surface”  feature?
      This will use a technique called “skinning” to create the hull surface in DELFTship. I expect the results to be a lot more usable.

      • This reply was modified 1 month, 2 weeks ago by Marven.
    • #43598
      John Owles
      Participant

      Thanks Marven, I am not sure that I understand exactly what skinning is or how to go about preparing for the import surface feature but I am willing to have a look at the idea.

      It is a shame when old tech cannot be dovetailed into new tech. It is one of my constant bugbears of the can’t do of modern technology. When tackling some complex piece of engineering from 60 to a hundred years ago, that I am trying to outsource, I am often confronted by  “you do realise Sir, that what was possible back then is quite impossible to do nowadays!”.  I find that I have been under the misapprehension that it was supposed to be the other way round.

      Hey ho, back to the drawing board. Yes I do still have one.

    • #43599
      Marven
      Keymaster

      Hi John,

      For skinning the geometry of the stations would be sufficient.

      Enter the points of the stations (bottom to top, aft to front) in a text file and use the “Import surface” to import it into DELFTship. DELFTship will try to fit a surface through the stations, and in most cases the results will be more pleasing compared to those of the table of offsets.

      See also the enclosed file. The format is described in more detail the manual, but the first line indicates units are in meters. Then for each station the x,y and z coordinates are specified from bottom to top. An empty line is used to indicate the end of a station. EOF marks the end of file.

      • This reply was modified 1 month, 1 week ago by Marven.
      Attachments:
Viewing 11 reply threads
  • You must be logged in to reply to this topic.