Forum Replies Created

Viewing 15 posts - 16 through 30 (of 201 total)
  • Author
  • in reply to: Aft Bulb Hull Modelling #39560

    Welcome on board, Martin!

    Today there’s something wrong with the pictures display (caused by my computer or by the server?).

    Broadly, a bulb is drawn by punching a hole, extruding its edge, then closing it.

    Punching a hole:
    Select a vertice and press the DEL key on your keyboard. If the convenient vertice doesn’t exist, draw it by dividing a polygon.

    Extruding the edges:
    Select the hole’s edges then press the EXTRUDE button on the toolbar. Give the direction and distance by relative coordinates.

    Close the hole:
    Since the hole is probably on the symmetry plane you’ll have to select each point on the edge, one after the other, in a logical order (clockwise or counterclockwise), then press the NEW SURFACE button on the toolbar.
    If the hole is not on the symmetry plane, you can select the vertices forming the hole’s edge and press the CLOSE button on the toolbar.

    I hope it’s clear enough. πŸ˜‰

    in reply to: Thickness and weight in “Design hydrostatics report” #39552

    Telling the truth, I don’t know how to use the LAYERS dialogue box (shame on me πŸ™ )

    BOOOO!!!!! :laugh:
    Using layers is really not difficult.

    You can consider them as transparent sheets you can draw on. You can draw each part of the model on its own sheet (hull, transom, deck, etc. or plank nΒ°1, plank nΒ°2, etc.). This allows you to display only the sheets you need when the model begins growing more and more complex.

    Using layers allows you to use a pretty interesting function: layers intersections. For example I drew a rowboat using rectangular planks (like you get them from the saw mill), I rotated them (to know which was the bottom, the side and the deck), I bend them inward, I asked DelftShip find the layers’ (planks’) intersection. The model was almost complete! :woohoo:

    You can assign these layers their own characteristics, such as color, transparency, and, more interesting, thickness and density. For example, on a little rowboat, you can choose a 8mm thick plank for the bottom, two 5mm thick planks for the sides and a 16mm thick plank for the transom (to attach the out-board motor on). With such different planks, where will be the CG? What will be the final weight? DelftShip can give you the answer, provided you feed it with the right data. B)

    Concerning the planks’ thickness, you may ask DelftShip draw an OFFSET SURFACE base on the hull and desck, using the plywood thickess for offset distance. This will give you the planks’ inner face, and you may use this new surface to draw the bulkheads.

    Just remember:
    – that however it’s spread over many layers, the hull must be only one continuous surface.
    – don’t forget to choose the right layer before begining drawing (just like you must make sure you’re cutting the right plank in the real life!).

    Spend a couple of hours studying the layers’ use may be really helpful. πŸ˜‰

    in reply to: volume properties #39549

    Salud eoc’h, Mergoet, et bienvenu Γ  bord!

    Providence is probably right, It can happen that the Y coordinate along the plane of symmetry may be very close to “0” but not exactly “0”, and this creates a leaking point and the volume is not considered as closed.
    It may happen, too, that the volume’s top is not closed (by a deck), or the topping is not well closed (Note: if your ship has no deck, or if there’s a hatchway, you must close it, even if it’s just with a transparent tarpaulin).

    When you’re in WIREFRAME display, there’s a button enabled to get an other kind of display: “LEAKING POINTS“. It highlights the breaches in your model.

    Ken ar c’hentaΓ±!

    in reply to: Thickness and weight in “Design hydrostatics report” #39543

    I assume it’s because Delftship was conceived first to design ships, before building them. Knowing the weight and center of gravity is useful to balance the ship, and this must be done before cutting the planks.
    That’s why the report is called “Hydrostatics”.

    I assume there are two solutions to your problem for the next versions:
    – either to include the possibility for the user to check boxes to set the report content;
    – either to include a definitely new kind of report called “planks list” or something like this.

    As far as I remember, you have the data you want in the LAYERS dialogue box. Maybe you can make a screenshot of it…

    in reply to: Brand new, please help with leak points #39542

    Let’s beging with the “invisible layer” stuff:
    With Delftship, you draw your model using “layers”. Consider them like transparent plastic sheets in the real life. Using the LAYERS dialogue box you can change the layers’ characteristics, such as toggling them from “visible” to “invisible” (and back, of course!). This is useful with complex models, thus you can focus just on what you’re drawing (for example, the portholes, the oars, the yardarms…).
    When you’re told your drawing on an invisible layer, it means the drawing will be made, but the result will be like the layer you ‘re using: invisible. so, open the LAYERS dialogue box and toggle the concerned layer back to “visible”.

    Concerning the leaking points:
    Delftship doesn’t consider the points below the waterline only, but the whole model. So you must close the hull’s top, either with a deck, either with a fake surface (it’s possible to set it “invisible”). Consider it, in the real life, a little like a temporary tarpaulin used on the shipyard to stop the rain from entering the ship, or the bird to*** in the ship. πŸ˜‰

    Remember: There are two buttons to “mirror” the hull:
    – one mirrors just de display (each subsequent change on the drawn side will be caried into the other side – symmetrical ship);
    – the other one does actually the drawing (none of the subsequent changes on a side will be carried on the other – unsymmetrical ships).
    Beware the confusion! :whistle:

    in reply to: Intersect function #39525

    Hi Ruedi!

    When I use the INTERSECTION function, I noticed that the layers’ selecting order is important. “Deck” then ” Coaming” won’t give exactly the same result than “Coaming” then “Deck”.
    So, I’m used to always use both selections, one after other, to get as many intersection points as possible, then I connect the most convenient points to get the right intersection curve, the I delete the surplus surface.

    An other way to do can be to use the OFFSET SURFACE function to draw a temporary surface parallele to the deck, above it, to draw a vertical cylinder thru both surfaces and to ovalize it (the *temporary* coaming’s *template*). Then you use the INTERSECTION function to draw the coaming edges and connect them drawing one afet other each polygon of the coaming.
    Finaly, you have to punch the hole and delete the temporary surfaces (offset deck and coaming template).

    Using this “temporary surface as template” method can be interesting when you have to draw many times the same thing on many places (portholes for example). Instead of deleting the template, just move it to the new place, ask the new Intersection points, draw the new wafer side, delete new unused surface, move, draw, delete again… and again…

    Is it clear? Does it help?

    in reply to: Font size, new button #39506

    If you can’t find this parameter in Delfship’s options, maybe you can try to use Windows’ parameters to change ALL your applications’ display… including Delftship’s.

    What do you call “transformed” layers?
    Is this a new function? (I’m still using a 3 years old version).
    When using Delftship, you work at editing all layers’ content… For me, an unmodified layer is just an empty layer…

    in reply to: Offset curve #39497

    You posted it in the right place (figure requests’ page).
    Now, you’ve to wait for Martin (forum admin & Delfship creator) to read it… and it may take several weeks. :whistle:

    There are so many things to do and so little time do do them all… πŸ˜‰

    in reply to: Offset curve #39495

    The best thing I can suggest you is to use the OFFSET SURFACE button to create a parallel surface.

    If you need just the surface edge, use the concerned edge of this new surface to EXTRUDE or CREATE A NEW SURFACE, then delete the parallel surface when/if you do no long need it.
    This way you can too draw both faces of a plank, then connect them to get the plank’s side surface, for example. (But be careful: drawing planks’ both sides may affect the weight calculations).

    Unfortunately, the temporary surface will be drawn on the same layer than the original.
    If you need to keep a part of the temporary surface on an other layer, the only way I found (on my old version) is to delete and re-draw each polygon one after the other on the new layer… Time consuming and boring…

    in reply to: Cylinder twisted #39493

    I noted thison my old version, too. It looks like Delftship rounds its sine and cosine calculations… after a slight rotation, meanwhile we feed it with integer numbers.
    One of the problem’s origins may be that the control points are not on the cylinder surface (radius or diameter) but at the intersection of two tengents to the cylinder’s surface. This requires a calculation… which must be rounded.

    I side with you Ruedi, it’s boring πŸ˜‰ … but I’m unsure there’s a solution… except editing the points’ coordinates after calculating them yourself… and when the cylinder must be drawn at an angle… :S

    in reply to: Hull Modelling with stations only #39484

    That’s the most boring part…
    Click’n drag again and again…
    Change the station and the polygons location…
    Click’n drag again and again…

    But finaly, I could get something acceptable. B)

    in reply to: Hull Modelling with stations only #39476

    On Maryak’s example, stations 22 to 32 won’t give you very much information, furthermore, they would look like an unreadable big pack of lines, because they’d all look the same.
    The best is to have all three views (side, top, front/back).

    So, as promised, I’m back after refining my method. B)

    I found a strange bodyplan (large beam and narrow transom) and wished to know what such a ship can look like? So I modeled her just using her stations profiles. I imagine she was around 10 meters long and 3.7 meters wide. I think her transom was rounded (like on Meriak’s example), but creating rounded transoms is an other question, I’ll draw her a flat inclined transom.


    Fifteen screenshots are following

    Martin -our webmaster- is going to kill me for the space my files are using on the forum!

    Don’t feel at loss if my Delftship display doesn’t look like yours: I’m still using a five years old version, but if it works on my old version, it will do on yours, too.

    in reply to: Hull Modelling with stations only #39463

    You’re right, distance between station can change at the very ends of the hull, but when it happens, you can see there’s something odd on the top view (the bow or stern will be pointing out in a aesthetic-displeasing way), then try to divide this station’s interval by two, and if it still doesn’t work well, you’ll have to guess the right distance by trial and error.

    When there are too many stations (I forgot to mention this eventuality), you can simplify the net (use less polygons) and let the software draw more than one station per mesh. I’m used to say about the meshed net “the simpler is the better”, because is easier to draw it and to modify it, too. πŸ˜‰
    If you really can’t get the model’s stations profile fit the background image, you’ll have to add a row of meshes (depends on the curve’s complexity).
    I’ll try to post scrennshots on the next week.

    in reply to: Hull Modelling with stations only #39459

    Hi, Axel!

    Drawing a ship just using the stations and knowing the total length is just what I did yesterday. :woohoo:

    First I started a project for the first sketch, asking Delftship to draw just a rectangle (same length than the ship), I changed its angles’ coordinates to superimose it on the vertical plane of symmetry.
    Then I imported the picture as a background image and centered it according to the plane if symmetry (shown by my rectangle).
    I positioned the top of the rectangle at the ship’s half beam distance.
    I determined the stations intervals (comparing their number and the total length), then I divided the rectangle in as many parts as there are stations.
    I moved the superior angles back to the symmetry plane.
    I moved the mesh angles to their position to determine the edges final position.
    The draft was done. Top view and side view were loooking well.

    Then I divided horizontaly into four the rectangles, which followed more or less the diagonal slicing.

    And now, the most boring: I moved each point, one after other, until I got the model’s stations superimposed to the background image. The problem was that when moving a point you don’t change only the corresponding station profile but the surrounding stations’ profiles too.

    Neither the keel or the rudder were shown on the background image, so I had to guess. :whistle:

    Does it sound clear? :unsure:

    in reply to: Show me your ships! #39452

    Thanks Cruz!

    On the next time I’ll go to the port, I’ll have a glance at the screws’ support, to make sure there’s some kind of mobility along the vertical axis. πŸ˜‰

Viewing 15 posts - 16 through 30 (of 201 total)