Forum Replies Created
I’m looking at the January 2012 issue of The Naval Architect (published by Rina.org).
On page 44, it says:
“DELFTship launches version 4.27
The Netherlands-based DELFTship has launched the latest version of its hull form modeling programme.”
The latest version of DELFTship sees the release of it (sic) latest feature, the calculation of hydrostatics in waves, along with improvements such as automatic update routine.
More in the article.
Why are they talking 4.27? That is an earlier number than is currently in the downloads section.
Any word on what we can expect in 5.x? Is it just GUI cosmetics, or should I start saving up some cash to re-invest in DELFTship?
That’s what I do with the stations and waterlines, so that I can have shell stiffeners longitudinal, stringers… Buttocks come in handy if it is useful to have long’l girders.
In my case, i use Punch ViaCAD, and i set up major and intermediate sizes of T lines to then rail sweep along those stations, waterlines, and buttocks. However, if my surfaces are not well-faired, some stiffeners once solidified (from surfaces made from the dxf 3d meshes) may not be perfectly flush. But, for my purposes, it’s fractions of a millimeters most of the time, and if built in the real world, then welding probably would fill in the gaps.
Boy, I sure wish that, and a frame-generator/distance from stem-to-stern, too. And, a simple tank generator, too (although not as comprehensive as the extension sold for DS Pro)…
Maybe the next release will have that, and more, such as an ability to fence/grop select and such?
Have a look here:
You may have to join to see the pictures.
Just as a courtesy, my company’s firewall or maybe Google in the middle displayed a warning:
Associated with spam or possibly compromised.
To influence which pages are blocked:
Contact your administrator about security settings on your network.
Provide feedback to [name of the protection system we use].”
Well, you can always try to turn your boat 180 degrees around. The algorithms used however are based upon a specific type of ship. You cannot used the Delft series, which is based upon sailing yachts, to predict resistance for planing motr boats for example. And of course the methods are also based on ships sailing with the pointy end in front;)
In short, it’s possible to fool the software by rotating your vessel, but the results will be useless.
Hmmm… That raises a question.
I had not until recently and then not again until reading this realized that calculations are (in my mind, crippled) in that someone, somewhere, deciced that Naval Architecture shall measure from the right, pointy end on the right, and that “tradition” flowed henceforth through time.
Given the computer age, I personally, strongly feel that modeling software in any industry should ask the user the USER’s prefered model orgin and then calculate from there. One Nav Arch I asked a couple of years ago had no specific on why it is that way, but posited that rudders were drawn first and therefore the architect or drafter started from the left and rolled the drawing scrolls up from the left.
Regretfully (wryly/snidely), I learned to draw ships (space ships) from the left to right, bow to the left, sine it seemed logical to them that since we typically use rulers and measuring aids numbered lowest at left, the art would follow function. Some real-world naval architects follow suit, but “tradition” has crept its head into the computer world and I suppose there is “resistance” to most firms providing calcs that work with the USER orientation, not implacable tradition.
So, now, I that while I could (and did) rotate my models 180, because my paper drawings are that way and because transferring information would be less insane, all the resistance calcs are worthless (not just because my hull is far, far larger than yachts, and later it finally was strongly impressed upon me that Delft series results are not applicable to frigate-type hulls), or rather, meaningless. (This makes me wonder and tempted to rotate the model again, bow to right, and running calcs on both. I thought I had checked just before committing myself to doing so, back in 08, but now I cannot recall, hehehhe…)
Thing is is that I don’t recall the manual stating the bow must be on the right. The calcs didn’t display a preamble of such a requirement, either. However, I am not a Naval Architect, and naturally would not expect to be constrained by programming or tradition. I generally thought it was something traditional, not that it was convenient for titleblocks to be read easier.
Have you told DS to “SHOW LEAK POINTS ONLY”? The view shold be easier to troubleshoot. Then, seek out any points that may lie just outside of Y=0. Some of the user-submitted models strangely have just ONE oddball point (sometimes near the bow) that looks fine in elevation/buttocks view but maybe lost in visual clutter but at Y=0.025.
If that fails, look for concurrently-located or co-located points. They may or may not be the same type, so do not rely solely on the colors in preferences to sleuth/suss them out.
And, check the Layers module to see whether or not you’ve inadvertently turned on for hydros calculations any layer which should not be so.
Also, make sure that — in the case it is not an issue with the hull — you don’t have any geometry in the body of the hull as placeholders for engines, decks, etc. Sometimes, I inadvertently turn back on the meshes standing in for weigths. Being busy with moving points, I often don’t catch it when the SAC (Section Area Curves) just drop off of the display when I do something that nullfies the live hydros calcs presentations.
If you do not have the SAC (the calculator button on the tool bar) on, it might be a good idea to turn it on so long as the values displayed don’t land inconveniently on points you’re editing. If so, reorient the view to get them out of the way. If you have screen space, go to the Window menu and click on New Window go avoid re-tiling your current window setup. If you re-tile them, they shift around and bow is now in a new location, and plan might move around. So, with new window, you can get a full screen of the model and put the view in whatever mode you wish while editing. Or, resize it and let it overlap other window views such that you can keep an eye on the SAC or the values in the bigger New Window. If they vanish, you moved a point of the hull away from Y=0 or turned on geometry that is included in the hydros but negates the hydros engine calculations.
Digest this in small, manageable parts unless you need a cure for insomnia.
Turn on mesh.
If desired, turn on Stations, Waterlines, Diagonals, and Buttocks.
Before you export, turn on the mesh and interactively use the various shading/gausing/stress view modes to look for inconsistencies in your fairing.
If you have DS Pro:
Turn on the “Show Control Curves” feature
Turn on the “Show Curvature Plot of Intersection Lines” feature
For managing the control curves, select a few at a time and hit F7, keeping a KEEN eye on each viewport. Train your eye to notice movement of larger items on a vp while you may still focus on another vp.
If your hand is unsteady, be sure to save periodically so that you needn’t undo-undo-undo endlessly.
Be sure to check out the Undo History and how it can help you via its graphical history view.
Export the model as a DXF 3D Mesh file.
Export the model as a DXF 3D Polyline file.
Into AutoCAD, import your … wait, in ACAD, you just OPEN the model. You’ll see the very gridded mesh of your hull surface. Depending on your display and shade mode settings in ACAD, you might become disoriented. Depending on your scroll/zoom/UCS settings, you may have lots of flicker — all depending on how many stations, waterlines, buttocks, and diagonals you have in your model.
Creating shell and deck plating in ACAD to obtain solids.
How *I* do it
I break down the hull into creased areas (not ideal or necessary for a small craft) because I’m designing 560′ ship. There i is no way I want one, large mesh to be my model. This may in theory be something to rave about. Some shops and ship builders like it because it’s simple and greater control can be had over a model with as few control points to herd as possible. I break it down for two major reasons:
— model management, so that memory consumption can be dramatically reduced (I don’t always want to zoom through 177mx19mx30m of steel and structure and ladders, even though the model pans and zooms fairly fast. The eye clutter can be annoying)
— compartment mass and CGs calculations (I want to know mass and centers info on a compartment basis, not just a whole hull basis, and I can get either by selecting items of interest and looking for moments and other info in a very neat and tidy panel or I can export it to a neat CSV file with next to no pre-formatting and faffing about, compared to ACAD, which at least in 2008 has moments and other data of different types on one or more or the same line, frustrating efforts at fast manipulation of exported mass props… at work, that irriated the bejeezes out of me, and is another reason I use Punch! ViaCAD)
But, I don’t have high-end, steep-learning-curve software. I have to force Delftship to give me what I need rather than use it how it is designed.
Ages ago (back in 08), i could not figure out why my color-coded sideshell pieces were not snapping to the stations and creased lines between control points as I hoped they would. My whole, entire rationale behind how I do it is to use the same by-hand faired model as the basis of my chopped-up model.
So now, I use creases at my watertight bulkhead zones. This forces me to keep most of my control points as vertical and horizontal as possible so that any sub-creasing is predictable. Nicely, it even allows me to have some “removable plate” in the event I want to have actual see-through areas rather than do it in CAD.
Traditionally, and for fairing reasons, control points may be placed as buttocks controllers. They might never be horizontal. I used to try to use them as guides for decks, but I realized I could export the waterlines for that. Then, I realized I could add another wl serving as plate thickness. I could later extrude it from centerline outboard to the surface of my sideshell.
I then tell my CAD app of preference (Punch! ViaCAD Pro v7) to convert the mesh into a surface. I then THICKEN the surface. Unlike ACAD, which refuses to allow the user to input a mix of Imperial and Metric units in a drawing file without changing the file settings, ViaCAD allows me to specify any coordinatate in either measurement unit. I can enter the length in fractions or decimal points of a parsec and draft in meters and the beam in feet. It’s my problem if fractions and decimals pose a cutting problem, but it is very generous and useful a feature to not be constrained to pick one over the other.
ViaCAD does not allow thickening of a mesh, nor drawing to or referencing from the mesh in a number of cases. But, after converting it to a surface, I can do many things unless I have too much complex or compound curvature. This is where it can be handy to use creases, and not just for hard chines and hull edges.
Set a crease some minimal distance ABOVE the waterline so that during expected heeling or rolling or pitching, the wetted surface of your model is as faired as you can make it by hand. Above the crease, you can design however you want/need to.
Another good reason for creasing is that as long as your fairing is up to build standards, you can thicken various parts of the hull to differing amounts for balancing, strengthening, and weight/mass distribution. Why have one whole model (whether 10′, 30′ or 500′ or 1800′) all thickened to one value? Real craft and ships are not that way. So, i set my layer thicknesses accordingly. Same for the decks. same for the bulkheads. So, I force DS to be my pre-designer tool and my layer prep. Each compartment has a color and it is different from its neighbor so that I can visually keep track of where I am. Anyway, by breaking down the model into parts, in the case of large vessels, memory consumption and CPU work can be dramatically reduced because smaller parts are drawn, calculated upon, thickened, and so on. I’ve had 60MB models evolve from a 243 kb DS file.
Your choice of CAD tool is not for me to change. But, if you want to know another reason I use VCP for modeling my ships: the colors of the hull sideshell and deck plating are more enjoyable and easier to set and work with. Thickening a surface into a solid is so simple it’s almost childsplay as long as the surface is not overly compound and so long as any extrusions are based on carefully faired lines having no kinks or reversals on close inspection. I don’t believe you get a real solid in ACAD 08, not even in 12 or 13. IIRC, you get lines representing a solid, and you get shading, showing you. You can get some CGs and mass props, but setting materials and such is something that tormented me in ACAD 08. Even TurboCAD 14 or soe (from 2007/2008) tormented me, and nothing tech support suggested was ever satisfactory or really useful. When I tried ViaCAD 2D/3D, I was hooked, and so hooked I upgraded later to Professional.
If you have money or access to use a legit copy, you may indeed be well off using Rhino with Parametrics, or MaxSurf, or several others. Visit popular boat design site for more help that is able to help with design/engineering/theory/etc questions.
Hopefully, I didn’t make you fall to sleep, hehehehe.
I am not fully verse in AutoCAD.
As for PLATES…
If by plates you mean shell plating, you need to ask yourself whether you want smoothly curving, unbroken hull mesh surfacing or if you want to have the hull “broken” in convenient places where the waterlines and stations intersect.
As for one big half-breadth hull surface, you can turn on the mesh and then export it. Depnding upon the CAD program or hull modeler you’re using outside of Delftship, you may or may not be able do much with the mesh. Since I’ve already mentioned that I am using Punch! ViaCAD Pro, I can tell you how I use the mesh. I export the mesh model to DXF 3D Mesh. Then, in VCP, I import that model. In my early forays into DelftShip, I realized i did NOT want one huge hull, since I am doing detail design and I want to account for weights of EVERYTHING I draw — not just the equipment, ducts, pipes, and so on, but the specific region of the hull and its bulkheads and decks, too.
So, I created waterlines and stations not for (well, not JUST for) hydros reasons, but to “break up” my hull. By creating stations at convenient intervals (matching my paper drawing in general, but adjusted when I found out the 3D model gave me more space than my pencil lead at 1″=20″ did), I found that I could assign mesh areas to specific layers. So, I created layers that were named for the hull compartment. I color-coded each so that in and out of DS, I could more easily see where I was in the model space.
When I imported those into ViaCAD, I realized I could thicken each compartment’s plate more realistically. Instead of going on some “average plate thickness” notion, I could by compartment thicken the plate. But, later, when I CREASED waterlines, I realized that I ended up with individual rectangular plating that I could more gradually thicken or thin in a more refined way to realistically deal with thinning the plate as it got higher or more near an extremity. I could thicken plates in zones needing more protection.
Now, where some may leave that to a hydrodynmics and weight specialist, this is MY model. *I* am all of the departments. So, I am not going to just use a hand-wave and declare or use rules of thumb or wrack my brain trying to calculate weight of x square meters of y type of steel of z thickness. I just tell Delftship the density information so that I can sanity check it against what ViaCAD later will calculate.
In ViaCAD, I assigne the MESH to a sublayer of its compartment, namely MESH. I copy and then transform that mesh into a SURFACE. I assign the surface to a layer in that compartment, and I then create a layer called Plate/Solid/etc as convenient, and then thicken the sideshell and assign it to its appropriate layer. I then go into “properties” and tell VCP to make it 1020 steel and then VCP shows me the CGs and moments of that steel. Later, when I add stiffening, decks, cutouts, ladders, partitions, pipes, and bunks, and so on, I can over time see know with more confidence what my hull region is WEIGHING, not just displacing.
CREASES AT WATERLINES
If you’ve created your points and made the general hull (or if you open one of the demo hulls), create waterlines and stations at convenient intervals just for experimentation.
Crease the waterlines. Note that creasing WILL slightly modify the hydros results. Creasing will likely force you to make more manual adjustments, but the positive side effect is you may end up with more planar plating if you re reasonably careful. Un-creasing can, however, re-introduce multiple problems since DS is not one of those $10,000-per-drafter-seat programs that will automatically re-fair within a tight tolerance close to what you had prior to fairing. DS WILL automatically fair for you, but YOU probably will want a shape of your own, not one of the interated farings.
Now, if you export those creased areas, you can play around and realise the things creasing can do for you besides making single and double chines.
You will probably spend WEEKS interactively iteratign your Delftship to CAD app drawings, looking for excessive sheer, camber, and twists, and reversals that you might not notice in DS if you’re not using the histogram tools correctly. If you’re creating stealthy corvetts or just want to exploit all that buzz around “steath shaping”, then creases can be your friend very quickly.
Remember, though, the manual process can be fun if time is no issue. But, it CAN take a good deal of time if you’re very demanding and perfectionist.
Are you trying to create shell stringers/sideshell stiffeners? That is, are you trying to create T or L stiffeners, such as web/flange plate?
I don’t think you can do that in Delftship.
What I did was to create my hull and fair it. Then, I added waterlines and stations, at various intervals where I thought I later on *might* need or want to add stiffeners.
After much fairing, i exported the DXF 3D Polylines and the DXF mesh. I then opened my CAD application and imported those files.
I assigned various waterlines and stations to conveniently named layers.
I color-coded things to make it easier to work.
Do you have a CAD application that can work with solids or create solids? I’m not talking about meshes. If you do not, then consider something you can afford. You have many choice, ranging from Rhino to Autocad, and Rhino with various add-ons.
If you find them to be too expensive (as I have) and are not generating cashflow from your drawings (as is my case), then you may want to try Punch! ViaCAD Pro. It is less than $300, and it has some pretty good 3D modeling tools. It is what I use. I like the look of the solids when I properly construct them. And, I like the reasonably-laid out and exportable CGs and Moments information that is calculated for individual and multiple solids in the model.
To create stringers, I sweep the profile or the web and the flange separately along a waterline.
Quickly, to generate stiffeners…
— Pick a waterline that will be your stringer’s path
— Find the stern (in my case i have transom sterns)intersection
— Draw a line that will be the web
— Draw a line that will be the flange
— Use one of the rail sweep tools you find works with the ending profile you’re wanting
— Thicken the surface using the “Thicken” tool. Before leaving the Thicken mode, zoom in to make sure it is where you want it. Use the “Ctrl” key to “throw” or shift the thickness to above, below, or exactly on the waterline.
If you like it, keep it. If not, then look at the angle at which you created the web and the flange. Nicely, the sweep tool will follow the line/curve even though it bends 90 degrees in the case of a transom stern of waterlines.
If you need to break those waterlines because later you want stringers interrupted by the bulkheads you might install, just break the solid with the trim or another tool. But, for that solid, some associative information may be lost. If you have to re-thicken the stiffener, it should still be possible later, but you might in some cases want to re-sweep the surface and re-thicken the solid.
Swept surfaces are associative to the curve used, so if you activate control points and move them around, you can force the surface to be reshaped. But, don’t get crazy — dramatic displacement of points could lock up the drawing due to calculations, since in 3D you might think you’ve moved a point only an inch or some number of millimeters, but if you are incorrectly using snaps, you might actually be moving a point by meters or kilometers in the drawing.
If you are creating sideshell stiffeners/panel stiffeners/bulkhead stiffeners, you can use the same or similar techniques.
* But, keep in mind that you want an app that allows you to export the solids information as some sort of dxf or delimited (comma, tab, etc) format. And, you want this format to appear STRUCTURED, not in some time-wasting mixed format that forces you to copy, paste, splice, and eliminate stuff that won’t neatly align into columnar format. Why? Well, you might later on want to run reports in a spreadsheet or a database app and don’t want to become a scripting expert just to untangle an insane or obtuse layout that is set up for visual use rather than analytical use.
* Clarification: I wanted to clarify that I am not criticizing DELFTSHIP’s hydros outputs and CGs/Moments info. I’m speaing of a well-known, very expensive, CAD app that requires some amount of scripting to get some decent formatting out of the mass properties information.
Thanks for that thourough description 🙂
Also, while in the layers area, consider color-coding your layers. Be sure to assign material mass/density information. If you are using Freeship/Hydronship AND DELFTship, be sure to note that they shift decimals differently than the other, and you could have wildly wrong density/weight results if you do not make the layer-by-layer changes. This may hamper or frustrate moving files back and forth. FS/HS also has a different idea about where the baseline is, as far as hydros calcs are concerned. Consequently, you’ll need to select your exported model and translate it vertically and do the same each time you import any supplemental/redundant layers — depending on your CAD app of choice. I use Punch! ViaCAD because it has very nice 3D design tools and a crisp, no-need-to-mode-switch 3D view. I hate and loathe the idea that if doing 3D work in a CAD app I must for some reason switch between drawing mode and “display” mode. ViaCAD shows my colors and surfaces and solids very beautifully, even if they are not engineering perfect.
As far as layers go, compared to FS/HS, DelftShip has a very decidedly easier to use layer display and editing facility. It was a very nice change from its previous design. Martjin did a nice, heartwarming job of making things easier.
As for creating stiffeners, since I’m using ViaCAD, this’ll be useful if you too are on ViaCAD or ViaCAD Pro, rather.
— Import your stations into VCP.
— Turn off any layers not needed
— Create in the Stations layer two sublayers: one for stbd, one for port.
— Assign as appropriate; repeat for waterlines and for other layers
— Create layers according to how you work.
In my case, after a couple or more years of experimenting (due to ViaCAD’s current inability to move layers out of their branches) I’ve decided to create master layers by compartment, then sublayers within by deck, then layers for sideshell stiffeners, stringers, girders, and longitudinals. Other layers are created as necessary, say for ventilation and so on. Some things demand to be completely on their own layer, so they are easy to manage than tracking down in deeply nested layers. Also, you may want to create construction layers within each compartment layer so that for visual reasons you can “declutter” your “picture” of your progress. It can be maddening to have unrelated construction lines and aiding surfaces in the way. But, one very, VERY nice feature of ViaCAD Pro is its cursor, named “LogiCursor”. When you attempt to select something, if it is coincident (by model orientation on screen) with some other geometry, a small “ambiguity” popup appears and presents to you the items which you can click on, and their geometry creation number in case you have multiple geometries/pieces same or similarly named.
Another very VERY nice feature of VCP is that even while the other layers are turned off, you can use an assigned shortcut key to “unhide” other layers. Press your shortcut key, then, with the layers dialog box open, right click on a layer of interest, and then select “Select all items in this layer”, and HWALLAH! Those items appear. So, this again is a(nother) strong reason to create things by compartment in a subdivided model with the whole model in one file (if you have the RAM & CPU; my machine is a lowly 4GB RAM laptop with a 2GHz CPU, and I am running it in a virtual machine called Sun xVM (formerly VirtualBox). Incidentally, I give win 7 ONLY 1.7 or so GB of RAM, and I personally feel the graphics in VCP in VirtualBox look better than in native Win7. My model pans, zooms, and redraws reasonably fast. Note, however, that when using the scroll wheel, the model moves jiggly if your cursor is over nested lines/curves. I think this is because the cursor (LogiCursor) is hunting for lines to populate the ‘ambiguity popup’ list. I just move the cursor around, tactically, to get the view I want then zoom in where I know the is not ratsnest of lines or curves to slow me down.
Select as the current layer that which is in the hull area you will do the work. It’ll save time and headaches to always do the work on the desired layer, or some generic “staging ground” that is not going to tangle or conflict with layers unrelated to your intermediate work.
When creating sideshell stiffeners, I select a station of interest. I create a line for the web, either horizontally or at some 20 degree angle, but of appropriate length for that region of the hull or compartment. Some might be 200mm, some 300mm.
I then use one of the rail sweep tools with one of the several sweep methods a given rail sweep tool may offer. I also then create a line for the flange, maybe 80mm or 100mm or 200mm, depending on examples i find on the Intertubes, since I don’t have high-end specialty software. This sweep act creates the surface based on a web construction line. If your angle of choice is decent, the swept surface emanating inboard will be approximately equidistant and reasonably good for modeling, but maybe not for construction of very large ships as they will use their own cutting and nesting tools. For a yacht or something under 100′, you are talking about fractions of a millimeter, nothing to sink the boat.
(When convenient or expedient, I take of copy of the “T” web & flange, and just translate-copy it to each station where I’ve decided there will be sideshell stiffeners.)
The rail sweep work of each of these lines relative to the station in question is a one-at-a-time affair. (Unfortunately, no scripting, no array sweeping.) So, in my case, having 15 compartments, with some 5-8 s/shell stfnrs, i have a lot of work. But, I don’t want to always mirror, so my work may be additional. Fortunately, VCP has a Mirror copy, so if I change the line lengths of the web on starboard, my port mirror changes.
(Rant On: Oh, I detest bow-to-the-right, so in DS/FS/HS i painstakingly mirror copy everything and z-axis rotate the hull, and then destroy the port side. Rulers (other than arch/engineering rulers) read from left to right. Most of us read from left to right. Most of us pull tape measures from left to right, and other than in drafting class, my brain goes NUTS stacking up numbers from the right side of something going to the left. I understand that propellers were drawn first, and on the left of the roll of paper in the olden days, but sometimes “tradition” is a major PITA — particularly when computers don’t care but humans DO… (rant off)
Similarly, do the same for the stringers, using waterlines. Nicely, since I don’t break waterlines, I can start a T stiffener at the stern on the CL, then tell VCP to sweep it along that curve. Once the curve ends, the sweep ends. I can use a cutting plane to lop off excess material once i’ve drawn all the stations and stringers i need.
I also thicken the surfaces. While still in the sweep mode for a given surface, i use the “Ctrl” key when I need to “flip” or throw the thickness to the opposide side of the swept surface.
Periodically, I use the materials properties facility to assign (in my case) SAE steel type to get the massprops, CGs, and moments. VERY, VERY BEAUTIFULLY, VCP gives two types of mass props info, depending on whether you select just ONE solid or whether you select multiple solids. Actually, it displays MOMENTS according to whether you’ve selected a single solid or multiple solids.
Now, note that all of my model is in ONE HUGE drawing. WHY? I want to be sure that my CGs and moments info stay consistent. Yes, I could break down the hull across files to reduce file sizes, save times, and opening times, but layer mangement could become a royal PITA if I ad-hoc wanted to rename a layer. I’d rather go through the pain in one file rather than 15 or more.
Plus, if you want to place + marks at each CG, VCP will do that. You could put them all on a single layer, then show ONLY those CG + marks and visualize your model that way, if that tickles you. Some people I know do each compartment in an ACAD file, and don’t care that the dwg has x=0 for each compartment. I cannot STAND that. If I take a measurement or imagine that I will receive a laser scan, i want the scan refs to very closely track the model’s notion of the BHD ref along X. It just makes future calcs less tedious and appear more thought out.
Also, if I want to see the whole model, it is there. I’ll later on save a copy, then progressively delete whole layers and resave, progressively resizing the drawing file downward. That wholeness of the model also means my mass properties information is in one place. The mass information can be exported, and it is far more spreadsheet and .dbf ready (either text file that you can easily turn to relational database model) than that I suffered under in ACAD 2008. I havent’ bothered to see if ADesk bothered to revamp their massprops tool and its output, but it was maddenign for me since I don’t do scripts/macros, and I don’t suffer crappy text layout that demands the use of a programmer when mass information should be IMMEDIATELY useful on export, that is columnar and row data need to not be intermixed with formless information that has to be manually stripped or reformatted…
(An aside: For a CAD app not aimed at naval architects, VCP does rather well for me (I’m not a naval architecty by ANY stretch of the imagination, but i CAN create GAs that worry some people. In Tokyo in 2004, some Mitsubishi people who saw my drawings didn’t know what to make of me when I tried to show my PAPER drawings, and our meeting was very short. All the while, one (a senior person) kept saying “NASHHUNAL SEEKRETS”. And, they refused to look at my drawings, eyes averting. Maybe they thought I was attempting to set them up for idea theft or something. Others who were engineers in my hostel kept asking me what Naval Architecture program I graduated from. That despite scrawls and bad linework, hehehe… end of aside)
Anyway, ViaCAD Pro will set you back some USD $250 prior to any coupons or 3rd party reseller offers. VC 2D/3D is about $99, but lacks some 3D tools that are in VCP. Sure, Rhino/Orcad is a well known (but costly to me) path but not all of us are designing for build, but PRETENDING to design for build. Anyone who got a hair to build my ships would import them and re-rationalize the bits they like by using their own shop tools or yard tools to avoid going crazy with my naming conventions and layers names.
If anything, VCP makes for good pre-vis in color. I get mesmerized looking at shell and deck plating work when it reaches as stage of panning and zooming and mass information.
You may also want to do this:
– Create stations at places where you want watertight bulkheads
– Create stations at appropriate (if you go by Class rules for a given region/flagging) or convenient to you (fictional or notional designs) distances so that you later have them in place. Know that sometimes, either DS or your CAD program may “twist/kink” up some lines but you won’t know that until you try sweeping a line down a one-rail curve in your CAD app. Out of some 320 stations in my models, I sometimes have 3 or 4 stubborn, wanton, craze-inducing lines that don’t behave, and even regenerating them in the CAD app does no good. It sometimes takes convertion the line into another type.
– Create two faces of the bhd by creating two stations where bulkheads exist, the distance being the plate thickness of the bhd. This could later help you to avoid certain trimming issues that might be specific to a particular CAD vendor
– Create waterlines at locations of your deck plating — assuming you’re designing MODERN vessels in which “design for manufacture” is your rule, and thus your deck plating is horizontal, not sloped
— Turn on the wls and stns
— in Layers, create names for the watertight subdivisions (major watertight compartments) in your hull. Turn on meshes and assign them as convenient to you. This way, later in your CAD app, you can vastly more conveniently get out of the way any piece of the hull that is confounding your CAD activities as opposed to your (DS-based) HYDROS activities.
As for the bhds, when you export them, and when you look at a profile in D/S, the bhds will be easier to locate by eye.
Now, as for those meshes and waterlines, turn off the meshes. Now, select each waterline of interest, and CREASE that/those waterlines. What you end up with is plates or panels at each waterline and intersecting station. This might be nice if you assume you are designing a hull made of flat pieces rather than curved pieces. If designing a type of stealth ship, or just a very cheaply built ship, flat panels used extensively, as you probably guess, require probably no rolling, meaning all that attendant jiggery and press equipment is not part of the construction process, lowering some costs, probably significantly. But, that is beside the main benefit here:
You can arrive at better weight estimation/calculations if your hull is more conveniently broken down in logical places. Why try to estimate weight of a compartment if the sideshell and decks and stringers in that compartment are running the full length of your DS model? Also, the stations then are pretty much already broken down by deck and waterline, meaning you can build sub-assemblies in CAD if that is your fancy. But, the pain in this is that rather than having a limited number of stations from gunwhale to center of keel at bottom, you have that times whatever number of waterlines you choose to crease, +/- other modified areas.
Note: It can be quite time-consuming. The biggest most demoralizing part of doing it my way is that CAD and hydro vendors have very few if no two-way plugins, which otherwise could vastly speed up incremental changes or fixes when errors are discovered in finer-grained work within a CAD app. It is a major, major bummer that DS, Freeship/Hydronship, and ViaCAD and other CAD vendors don’t get together. If they did, a lot of vexing work could be streamlined, and poorer ship designers could avoid paying thousands of dollars per seat for CAD-design modules that they’ll hardly use regularly, but must pay for to get SOME utility out of the CAD app. (Imagine free-form designing the general hull in CAD, then importing to DS the various preliminary stuff, and assigning by layer the functions of those pieces and having the hydros further enhanced by the presence of those tiny little inclusions. Overall, it might not make much of a difference in this regard.)
Hopefully, this is useful information.
Obtain a hull plan/bodyplan/etc.
Open — and save as save as a new name — one of the tanker hull models that is close to the Titanic’s size (dims, not displaceemnt necessarily) and play with moving the nodes/points around (up/down, fore/aft, and athwartship).
Occasionally, look at the various modes and at the reports and learn to by eye gauge what your actions do to the digital model and how the reports indicate a more efficient and a less efficient hull. Add or remove sideshell so that it begins to resemble the hull. For the superstructure, you’ll need to interactively use the manual and the software.
At some point, reverse engineer the model akin to “looking under the hood”.
The nice thing about it not truly merging is that it remains possible to adjust the endpoints of lines of a surface or mesh when turning off other layers that share that coordinate.
That unfortunate thought crossed my mind as i hit send.
Maybe this will work: create in Delftship from the box tool one or more planes acting as undisturbed surfaces.
Alternatively, instead of using the box tool, use 4 points and then select each and create a face.
However you create the water surface, turn on the waterlines and turn off the hydros for the surfaces. If you’re experimental, create a number of stations and waterlines that you can use to select points to elevate or displace to simulate waves in a cad pkg, or maybe in Matlab.
I know how you feel. I too would like to have interactive graphing of various displacements in one view rather than export, synchronise, and redo to arive at a manual optimsation. Too bad the baseline version was mired with a modal mindset rather than non-modal .
Yes of course, and that works just fine except for the need to export the data as well, which doesn’t work.
But I’m trying to come up with an alternative solution.
This is an ideal off the top of my head:
Export the model as .dxf 3D Mesh and 3D Polylines to 3D CAD software. In my case, i use Punch! ViaCAD Pro v7. Append each name with its contents type, such as “… DXF 3D Mesh.dxf” and “…. DXF 3D pl.dxf”.
Import individually singly each dxf into your CAD app and suitably adjust the layer names if necessary.
Find the sounding pipes or overboard discharges, and so on, that are related to downflooding points. Select them all and turn on their points if that is necessary to select, move, or lock them.
Lock the end points that are above the waterline or at the waterline. If possible, make them further non-selectable after locking them. If you cannot do that, then I have an idea in a few steps past this paragraph.
Now, select the hull that will be trimmed, rolled, heeled, etc.
Adjust the x, y, and z angles as desired.
If this idea works, then you should see the locked lines “grow” in some direction in accordance with the displaced positions of the model’s movement.
Idea: if that doesn’t/won’t work, then orient the model as desired and then draw rays vertically from the submerged points you’re tracking or exporting to your database. Note the distances and positions.
This obviously is tedious, but it may work. It really will depend on the CAD package you’re using.
Alternatively, to avoid dealing with a massive select/de-select mess or nightmare:
— leave the hull in place and rotate the water (a big square or circle or trapezoidal shape),
— fire out the downflood and discharges positions the rays up to or through the water.
— Segment the lines and one-by-one select the lines that remain, and then get their x2 coordinate points.
It’s a PITA doing it this way, but this is likely how I’m going to do it.
I hope this is useful or helps you arrive at some more workable ideas.
2 weeks, 5 days ago
Hull modeling from offset table
1 month ago
import excel to delftship and surface became twisted
1 month ago
Tank Modeling DTL file Structure
1 month ago
Importing CSV/TXT coordinates as stations?
1 month ago
starting at a blank page.
1 month, 2 weeks ago
Complete Newbie Needs Help Starting Out
2 months, 2 weeks ago
Corner points appearing in model- need help
2 months, 3 weeks ago
Beginner seeks tutor
3 months ago
Lines Plan – Header on pdf
by Anneler Ruedi
4 months, 2 weeks ago