Areas of use for this include
Fashion & Footwear
Environmental (DIVA, GECO, Ladybug)
Help converting your parametric model to a BIM model, using Geometry Gym
Creating Custom Grasshopper Plug-in’s
Placing Specific Panels On a Surface
Optimising a Model
Data Conversion - Converting Data to Excel Spreadsheets
Links to Other Software
Converting different types of Geometry, Mesh, NURBS
Using popular renderers including
KeyShot Tutorial Series - To help you learn KeyShot and Rhino.
This is the video transcript, published alongside the video, to further support your Rhino for Windows and KeyShot learning experience.
Hi this is Phil Cook at Simply Rhino. In this video, we are going to take a look at an introduction to rendering out of Rhino [Windows version] with KeyShot. KeyShot is an easy to use, photorealistic renderer, and if your aim is to create great looking images without a steep learning curve, then KeyShot has much to recommend it.
KeyShot is a standalone product and imports data from many common models. So if you use Rhino with, for example, SolidWorks, then KeyShot can render content from both. A useful addition for Rhino users is the plugin that allows for live linking, meaning that the geometry can be updated in Rhino and be pushed to KeyShot on the fly, and we’ll look at this during this video.
The starting point for the exercise is this Rhino model of a wrist watch. All of the components are modelled as water tight solids and each set of components are on individual layers. So for example, the hour, minute and second hands are separated as are the painted areas of the hands which will be rendered in a different material. By default, KeyShot applies materials on a per layer basis. We’re using KeyShot version 6.2 and we’ve installed the KeyShot for Rhino plugin and this gives an additional short menu with the following options: render; export; update and include nerves. I have the latter checked.
Before pushing this out to KeyShot, I’m going to turn on the geometry for the crystal and then go to my KeyShot menu and hit the render button. KeyShot will now open and the geometry appears inside of KeyShot. KeyShot presents us with a real time rendered view and then some options on the left and the right. Briefly on the left we have materials, colours, environments, back plates, textures and an area in which to store our favourites of all of these.
To the right of the preview window we have our scene, and if I expand the scene here, you will see our file here and then all of the various components of the file on their separate layers.
Next to that we have material and here we can see the individual materials that are applied to the components in the model. These are mainly black at the moment because this is the default that’s carried over from Rhino. We then have the ability to edit the environment, lighting, camera and the image parameters.
Let’s first of all look at manipulating the model. The mouse gestures are slightly different in KeyShot to Rhino and to rotate the object I hold down the left mouse button and drag. And to zoom, I use the scroll wheel as in Rhino, but the scroll wheel direction is reversed. If I want to pan, I hold down the middle mouse button or the scroll wheel and drag.
Let’s now look at applying some basic materials. There’s a good visual material library and a good set of standard materials. So let’s go to metal and steel and basic and apply a polish steel to the watch case, and the same material to the basal. Now at the moment these materials look a little lifeless, so let’s got to environment and drag in a basic environment. By default, this environment is used both to light with for the reflections and also for the background. We can change those parameters of course later on. Let’s go back to materials and add some further materials. So again, I’ll add the polished steel to the crown and let’s add a rough steel to the buckles. This material looks a little bit too rough. So let’s go in to the material editor and double click on rough steel and reduce the roughness.
Let’s go back to materials, and this area on the case has a brushed texture. So there’s a couple of brushed textures here. A light brushed texture would be good and we’ll add that to this area. Now I’m going to look at controlling the size and the mapping of this texture a little later on but for now I just want to get the basic material in place.
Let’s go to cloth and leather and just for the moment put a nylon material on to the strap. Again, we’ll look at the texture and the mapping of this a little later on, and of course our crystal here is covering most of what is going on in the face of the watch. So let’s go to our scene, locate the crystal layer and just turn that off.
Now all of the materials for the components of the dial come through as the default. So at the moment we can’t identify these. So although I’m still going to drag and drop the materials, I’m actually going to apply them to the appropriate layer in the model from our parts over here. So I’m going to first of all go to metal an precious metals, and use a platinum here and I’m going to use a polished platinum and I’m going to apply this to the surround for the applied markers and numerals and also to the logo and the boundary of the hour hand, minute hand and second hand. Then as most designers do, I’m going to change my mind and use a polished steel, or actually just a steel for the hands. That’s better. I don’t want those to be quite as bright as the boundaries for the numerals.
Okay, then I’m going to create a material for the self-illuminating paint inside of the hands. So let’s go to paint materials and matt and we’ll apply a matt white paint. I’ll go to the materials over here and locate that material. Double click and get in to the editor. I’m going to make the colour more grey, just so it’s not so bright and I’m going to reduce the roughness a little bit. In my materials over on the left hand side here, I’m going to go to the top level and add a folder. Call this folder watch, and then I can save this material. Actually before I do that, let’s rename it and save this to my new folder. Okay, then I can go back to scene and I can apply this to the infill for the hands. Let’s go back to matt paint and pick the actual full white material now and let’s apply that to the minute track and also to the text on the dial. And also we need a material for the dial itself as well.
Now the dial material, again, I think we’ll edit this. So I want to make this a little bit darker and less rough. Now we might swap this dial material later on but for now this will be good. Let’s save this. So that should be saved in here now. So we’ve got our dial now and our super luminova saved in our watch folder.
Okay so now I can go back to the scene and turn back on the crystal and then give this a material. So to keep things simple to start with I’m going to use a simple glass material, a clear glass and use the basic glass. In order to improve this material slightly, I go to material, edit this material and turn on the two sided value, just slightly increase the refraction on it to make it look a little bit more like crystal. But I don’t want too much reflection on here, just enough that I can get an idea that the crystal is actually domed.
Okay so you’ll see that we’re starting to get a really good basic viable render now. So before we go in and play with any of the textures, let’s just look at a couple of things we can do with the scene. So in environment, we can rotate the actual HDR that’s used to light with and also is used to create the reflections and this is quite nice the way we can preview this on the fly. We can increase or decrease the contrast, brightness and we can choose whether to have the environment in the background or whether to have a flat colour and of course we can change that colour and make this a grey scale and we can choose to have ground reflections to flatten the ground as well.
In lighting here, we can move from our basic lighting preset to product for example, which will give us a little bit better calculated solution. It will take a little longer to calculate but our lighting will be better. We can do things like increase the quality of the shadows. You can see the shadows are a little bit blocky down here, so we can increase the quality of those. All of these things of course increase the quality of the render time but ultimately will give us a nicer image.
Let’s go back to the nylon fabric and look at the texture on this. We can either choose to increase the scale of the texture here and we can choose a mapping style here as well. There you go, and you’ll see if I increase this, you’ll see how much bigger this gets now. Just to speed things up, I’ll go back to the lighting here and reduce this a little here just so we can see the texture. So this has got much more of a texture that we want here. The box mapping would be a more appropriate texture for this because then we’ll get the texture on the side here. It just wants to be slightly smaller than that and that’s looking pretty good. Let’s also look at the texture of the brushed metal here as well. So let’s go back to the material and look at the texture here and let’s try a planer Z map for this and turn this round. I want the brushing to go top to bottom here. I might need to zoom in a little here to just see this. Indeed, I might need to turn up the scale to see which way this texture is going because it’s pretty subtle. That’s better so now I can see that’s going the right way I think. Let’s just turn that up a bit more. There we go, I can see that going lengthways now. So let’s just make sure that’s mapped on the sides as well and turn the value of the scale right down again now.
Okay so let’s just leave that for a moment to regenerate and we should be getting close to a reasonably good render now. So this is looking reasonably interesting now. One of the things that I’m not happy with is the size of this logo here. This is a good opportunity to show the update feature in the Rhino plugin. So let’s go back to Rhino here, hide the crystal here and take a look at the elements of this logo and let’s look at increasing the size of these. So I’m going to scale this 3D and make this quite a bit bigger and I also need to make sure that I adjust the height of this as well. So I’ll have to turn off a few layers. And then I need to move this vertically, turn on project of course here and snap that on to the top of the dial. Show all of the other components now and I’m going to make sure that those objects are selected there and just hit update and go back to KeyShot and you’ll see that my logo has updated. This is also the case if I create new geometry as well.
So if I just put something in here for the sake of argument and update, this will come in to my KeyShot document.
Okay, so this gives us a really nice way of live linking the two programmes together. One of the things that I want to do when I’m designing something like this, which is effectively a piece of jewellery is have a really good handle on the effects of changing materials and the scale of certain items. So the render window here is progressive inside of KeyShot’s, this continually updates and I can either choose to wait for the render to update and use effectively a screen capture, or I can render out to a saved render file. A couple of other things that I can do here inside of camera, is that I can turn on depth of field. I can select a point of focus, click on the model where the focus is and of course reduce the aperture here. Let’s make this about F12 and just get a little bit of blur in to the back of the image. Once you have depth of field enabled, this adds considerably to the time that the real time preview takes.
Okay so now we’ve let the progressive render go for a little while, you’ll see how all the shadows now are smoothed out. We’ve lost that blockiness that was down here. We’ve got nice depth of field going on now, nice smooth blurring here, nice and sharp around here where we set the focus and we’ve got a reasonably good image.
I’m going to take that depth of field off now and let’s look at rendering out an image. So first of all we can pause the real time render and then we can go to the render button here and we can choose either a preset size for the render here or of course set our own values here, and we can choose either to render a region or the entire image. We’ve got options here for the level of quality we want to use. We can choose how many CPU cores to use and we’ve got render queueing in here as well. So let’s hit the render button and you’ll see the render starts to create. Now this is more akin to a normal ray trace render now, where it’s rendering the scene bucket by bucket or piece by piece and we’ll just let this render complete and you can see the quality of the output.
Okay so there we go, we can see the final image now. Let’s just take this up to 100% and you can see this is reasonably good. You can see we’ve got a bit of an error here in the surface. So we can go back to Rhino and fix that but overall the image looks fairly good. So let’s just go back to fix this in Rhino. So uncheck the pause here. Let’s go back to our Rhino model, and there you can actually see the problem here where we actually have a duplicate surface there. So let’s just take this out and take this part and push this back out to KeyShot. Go back to KeyShot. There we go, that looks much better.
So you can see this idea of having the update in KeyShot really, really useful, just being able to go back to Rhino, update the geometry, push that straight back to Rhino again, really good. So let’s just do a few little tweaks on this, just manipulate the view slightly. Just move that refraction slightly as well, get a little bit more interest in the brushed area of the case and move this strong reflection out of the way of the logo there. I just want to see a little bit of light on this crystal here. That looks a little bit better, so let’s render that out and take a look at this.
Okay so there we go, we’ve obviously shortened that sequence a little but we can see now we’re at 100% and you can see this looks pretty good. We’ve fixed this problem with the strap. We might choose to reduce the brightness of the white hands and the numbers and markers slightly but other than that, this is looking pretty good. So there we go, first simple render with KeyShot and Rhino.
Thanks for watching and please do check back for regular video updates at Simply Rhino.
KeyShot is an easy to use, photorealistic renderer, and if your aim is to create great looking images without a steep learning curve, then KeyShot has much to recommend it.
KeyShot is a standalone render product and imports data from many common models. So if you use Rhino with, for example, SOLIDWORKS, then KeyShot can render content from both.
A useful addition for Rhino for Windows users is the plugin that allows for live linking, meaning that the geometry can be updated in Rhino and then be pushed to KeyShot on the fly.
The starting point for the exercise is this Rhino model of a wrist watch. All of the components are modelled as water tight solids and each set of components are on individual layers. So for example, the hour, minute and second hands are separated as are the painted areas of the hands which will be rendered in a different material. By default, KeyShot applies materials on a per layer basis.
We’re using Rhino for Windows and KeyShot version 6.2 plus we’ve installed the KeyShot for Rhino (Windows version) plugin.
If you use Rhino on the Mac platform then be sure to check back with us as we'll be producing a KeyShot for Rhino for Mac tutorial in the near future.
Rhino3d Video Tutorials Transcripts - To further support you as you learn and progress with Rhino we've transcribed each of our video tutorials.
Hi this is Phil from Simply Rhino. In this tutorial, we’re going to take a look at creating production quality surfaces and solids from 2D design intent. We’ve used the example of a ceramic coffee pot as it’s often the case within the ceramics industry that the design progresses quite a way in 2D before any 3D involvement.
The starting point for this exercise is a series of 2D drawings, created in Adobe Illustrator. We’ll look at quickly reconstructing some of the major construction curves in Rhino, before producing a series of surfaces from the minimum of curve input. The emphasis is on creating high quality surfaces relatively quickly before arriving at the final 3D solid model.
This tutorial is in three parts and this is part one.
The starting point for this exercise is a 2D layout presented in Adobe Illustrator. This is a full size drawing containing a plan, a number of elevations and some sectional views. In fact, most if not all the information we need to create an accurate 3D model in Rhino.
If we just zoom in and take a look at the plan and elevations, we can see that the shape is essentially elliptical in plan and also that there is a subtle shape change going on between the curves as shown in the side elevation and as shown in the end elevation. The high point for the curves in the side elevation is somewhere around about here, and it is much higher up, somewhere around here on the end elevation. So clearly this is something that we need to get right when we start to build the surface in Rhino.
Now before I bring the drawing in to Rhino, I’m just going to draw a simple curve in Illustrator, with the aim really of just showing you the similarities and also the differences between how Illustrator and Rhino handle curves.
You’ll see if I click on a node here that as I move a point here, and I’ll just zoom in a little, you’ll see I get these tangent handles that actually move out with this node or control point. And if I pull these tangent handles and move them around, I can change the shape of the curve. So the Illustrator curves are basically controlled by these handles and Illustrator knows to move the appropriate points either side of the centre point on the handle in order to keep the curve smooth.
So let’s save this file and let’s go in to Rhino and open this up.
So I’m not inside of Rhino and I’m going to go to ‘file’ and ‘open’, choose the Illustrator file and bring this in to Rhino. I’ll have a dialogue box come up here and I’m choosing the ‘preserve units’ option which will keep 1mm in Illustrator to 1mm in Rhino and I’m also going to choose this ‘load text’ button here to bring in the text objects. Now the text formatting may get lost as you can see here, and in fact, this isn’t of that much interest to me so I’m going to remove this text, but you’ll see that the text on things like dimensions will come through in the most part correctly.
Now the first thing I need to do is just to double check one of these dimensions. So my dimensions in Rhino are likely to result to more decimal places here, so I’m going to see a slightly different result here in Rhino but you can see that this has come through at full size.
Now before I go on to start looking at the model of the coffee pot, I just wanted to take a quick look at this curve that we drew in Illustrator, and I’m going to pick this curve and turn on the control points for it. You can see here the rows of three control points in a line, which are effectively where the tangency handles were in the handlebar editing tools that we were using in Illustrator on the curve. Now technically, Illustrator draws cubic Beziers which are degree three single span or 4 point curves. So this curve that we have here is not a single curve, it’s actually a number of curves joined together. So if I explode this, you’ll see that we’ve got these small segments here, each one of which has got four control points. So technically a Bezier is similar to a nurbs curve with the exception that it can only have one more control point than the degree and what this means is that with curves that are brought in from Illustrator, when these curves have some complexity to it like this, then this curve is not going to be curvature continuous. It’s only going to be tangent continuous at these ends of the separate curves. And of course, if we turn on the control points in Rhino and we effectively move one of the points that’s on a tangent point here, then of course we’ll expose that fact and we’ll get a kink in the curve. So there’s nothing wrong intrinsically with Bezier’s but just be careful when you start point editing curves that have been built in Illustrator. Very often for the type of work that we’re doing here, nurbs curves are slightly more useful to us.
Okay, so let’s go ahead and delete this curve and then centralise this imported drawing on to our Rhino construction plane and then let’s start to do a little bit of organisation here. So I’m going to change the layer name from default to imported CAD. I’m going to put a numerical prefix in front of that layer as well, so that I can always get my layers back in order if anything untoward happens with them, and I’m also going to have a blank layer right at the top of my layer stack as well. Then I’m going to create another new layer and I’m going to call this elevation’s in place. Now what I’m going to start to do is to move some copies of these elevations and plans in to a common position in 3D space so that they’re all sitting in their appropriate positions. So I’m going to create a new sub-layer here under elevations in place and call this side elevation and make that layer active and I’m going to copy the side elevation here on to that new layer, and then turn off the imported CAD layer, just a bit of extra information there that I don’t want. And I’m going to try and find a midpoint of this bottom edge and centre it around zero, so that’s centred around the axis now on our world construction plane and I just want to check the accuracy of that midpoint there. So I’m going to just snap on to that midpoint there and find a position here, find a position there, check that these dimensions are the same which they are, so that’s good. So it looks as though there’s some symmetry about this model which is good.
So that’s our first object in position, except that of course, this is sitting in the top construction plane at the moment and it would make more sense if this was on the left construction plane. So to move this from its position to the same relative position on for example, the left construction plane, I pick this in the top construction plane and then go to transform, orient and remap to C-Plane. Click on the C-Plane I want to move it to and this moves the object from one C-Plane to another, keeping the relative position to 00 the same. So this is an effective way of bringing in a 2D drawing like we have at the moment and orienting the views in to their correct 3D positions.
So I can now go back to our imported CAD and lets essentially repeat this process let’s say, with this end elevation here. So I’ll create a second sub-layer here and I’ll give this again the numerical prefix and I’ll call it handle elevation. I’ll make this active, grab this drawing here and copy it on to this layer and essentially repeat the process here. Centre that around an axis, let’s just check another element here. Let’s go from 0 with the dimension here to that handle in there, and same again. Okay, that looks good. So again showing me that handle looks symmetrical around 0 and this really needs to be on the rear elevation.
So let’s set view to back, pick this and transform, orient and remap to C-Plane. Okay, so if we, for example, just turn on the side elevation as well as the handle elevation, you’ll now be able to see these, sitting in their respective positions in 3D space.
So now I’m going to go back to the imported CAD and I’m going to repeat the process of orienting these views one by one until they’re all done.
Okay so now we have the various sections in place. Okay, so you can see that everything is just placed where it should be in 3D space. So this makes it pretty easy for us now to start tracing off some curves. Probably the best thing here is probably to start drawing these curves afresh in Rhino. If we look at what’s come through from Illustrator, some of the curves look quite clean, some of the other curves look very dense, very difficult to work with.
So let’s say that we’re going to create some new curves in Rhino here for everything rather than relying on any of the imported curves. So as we sort of pointed out earlier on, the shape of the body of the coffee pot is essentially elliptical and then there is a change in cross-section from the side elevation to the end elevation, where the high point is at the mid here and it’s much lower down here. So it’s a reasonably subtle shape and we want to maintain this elliptical cross-section as well.
So I guess the first thing we need to do really is to actually take a look at the base of this. We’ve got some call out dimensions here at the high point of the curves, but we know that these high points are in slightly different positions. So we might need to start by slightly approximating this and by that I mean that we can actually draw an ellipse of these dimensions here and produce a surface that is close to the side elevation and then amend that surface so that it matches the end elevation.
So I’m going to create another set of layers here. I’m going to call these 2D curves and I’m going to create a sub-layer here and call this body curves. Again, this needs a prefix in front of it and let’s just turn on the side elevation so working on our route view here, turn on project and let’s look at curve fitting a curve over the top of this existing curve here. So I’m going to use an interpolated curve here and I’m going to just change the colour of my layer here so I can see this a little more readily and I’m just going to click on three or four positions there to get that curve. Let’s just extend this curve using the bi-arc option here and pull this up here and pull this down here slightly. Okay, now because I drew an interpolated curve here, I’m not in control of the number of control points in here and obviously where I extended as well, I’m going to get these extra points here as well. So the first thing I need to do is to rebuild this curve and I’m just going to keep this nice and simple. Degree three, four control points. Now this obviously much simpler and easier to work with, and I just want to check the deviation between this and the existing curve. I can do this by going to analyse, curve and deviation and then pick the two curves here and Rhino will tell me that I’m 0.003 of a millimetre out here and if I repeat that command again, with the keep marks option on, it should give me some points where I’m closest and furthest away. So 0.003 of a millimetre, not particularly worried about that, so let’s remove these points and let’s say that this rebuilt curve is close enough for what we need at the moment and I’m just going to trim the bottom off this curve. So just make sure the bottom part of it is the right height and I’m going to leave the top slightly over long here. It doesn’t need to be quite as long as it is.
Okay now when you trim a curve, you don’t change the degree or number of control points. Rhino actually reapproximates this curve when you trim it. So as you trim it remains degree three with four control points.
Okay now the idea of the exercise here is to try and build the shape as quickly as possible without having to rely on building too many curves. So I think that I can use just one side elevation curve here and the ellipse and can then start to build the shape from there. So I’m going to now go to either a perspective or a planned view would do the job here. Let’s do this in perspective, and I’m going to draw my ellipse and I’m just going to check off the size that I need here. So I need it to be 126.1 x 101.1 and of course I need it to be centred around 0. Okay, so I’m going to run the ellipse tool here from centre, type 0 for the centre and for this axis here, it’s going to be 126.1, but I only want half of that dimension, so just type in the maths for that in the command line. That fixes that axis, and then this one is going to be 101.1 over 2 and that is the axis that way.
Okay so we’ve now got some body curves here. Let’s just take a look at how that ellipse compares to the planned view that we had here. So let’s turn off the handle elevation and let’s just have a look in top view here. It’s actually pretty close. Now we’re expecting to modify this slightly, so I’m not expecting it to be exact, but that looks pretty good for the moment. So because I’m going to build a surface here that I’m going to modify, I want to build this quite simply. So let’s create a new set of layers here, and call this surfaces. We’ll call this one, again with a numerical prefix, we’ll call this body surface. And we’ll call this body surface outer because I think we may be building the B-Surface, the inside surface of this.
So the technique that I’m going to use here is a rail revolve. This is a fairly kind of simple way of building a surface where we have a combination between a one rail sweep and a revolve. So I need to take off my project constraint and then rail revolve, pick the profile curve, the rail curve and then define the axis. Pull up vertically to define the axis and this will build my rail revolve surface. Now you’ll see this is quite a nice simple surface and an advantage of this, if we look at the control points here, it will retain the degree two orientation about the elliptical profile and we can check this here by picking a surface and going to what and you can see that it’s degree two in u and degree three in v.
Okay now if we now compare this with first of all the side elevation, that looks reasonably close. In fact, it’s as close as our original curve was. But if we now look at the handle elevation for example, you’ll see that this is where we can see that the shape is changing. Now the idea here is that it is fairly easy to point edit this surface as long as we’re careful. So I’m going to turn on the control points for this surface and I’m going to turn off my Rhino curves that I’ve built and I’m going to lock the handle elevation layer and we could work in Ghost here, it might help us see a little better. What I’m going to do is to pick some rows of points here, let’s just have a look at this in 3D. So it’s important to maintain the elliptical structure of this, that these three points and these three points stay in line. So these need to be moved together. So in other words, what I’m going to do is change the shape just as it appears in this elevation. Let’s go back to wireframe to do this, and what I’m going to do is to scale this. I’m going to use the transform and scale 1D command, and I can snap for example to the middle point here. Pull over here, and then start to pull out those points. Do the same here, and just gradually get this shape correct. Okay, needs a bit of iteration to do this. Just be careful when you’re selecting the points for this type of thing that you don’t select anything that you shouldn’t be selecting. So I’ve been careful to lock layers here. These ones want to come in slightly. So all the time using scale 1D here. So we’re getting closer now. Okay that looks pretty good now, so let’s turn on the shaded view for that and what we should see here is that our shape has stayed the same in our side elevation, but we’ve modified it in the end elevation. And if we’ve been careful with this, then we should see here that we don’t have any disruption of the Zebra’s as we move along here.
Okay, remember to get good feedback from the meshes for the analysis meshes. Go to detailed controls and a quick way of setting the meshes as ever, is to set everything to 0 which overrides the setting. Just use this one value here, maximum distance edge to surface. Set this as something similar to the absolute modelling tolerance to get a really nice tight mesh here and there we go, and get a nice smooth mesh there. So you see, there’s no disruption over where the quad points would be on the ellipsis here. That all looks nice and smooth and again, if we see this within environment map, again that’s nice and smooth.
Okay so that’s our basic outer surface for the coffee pot. Now if we wanted to offset this surface to create the inner surface, rather than to create a solid outer volume and shell inwards, then we have some choices to think about when we offset. So I’m going to create a new layer here. Call this body surface inner, make it active and let’s look at offsetting.
So pick the object, surface, offset surface. Offset will occur in the direction of the arrows, so if you want to flip that direction just hit the flip option here. I’m going to set my distance to three and I’m going to offset. Now when you offset normally, you offset to a tolerance and that means that your inner surface is going to be more complex than the outer surface. Now in this instance, that’s probably not too much of an issue because this is really a B-Surface and in fact this surface wouldn’t even be used to define any tooling in this instance because this part would be slip cast, so that it would only be the A-Surface that was being used to define the tooling. But of course, we need to create the B-Surface, so that we can work out things like the fit of the lid. Let’s just take a look at this and see whether we’ve got any disruption over the tangencies.
Sometimes when we offset a surface that is degree two in one direction and three in the other, we may very occasionally get some disruption around the quad points. We haven’t got any there. So this is a reasonably good way of creating a surface that is not really for aesthetic purposes. It’s just a functional inner shell. However, if we wanted to offset and have a surface that had fewer control points, then we would use the loose option. The loose option really is a control point offset, so it offsets the control points by a fixed distance, giving us effectively the same control point structure and degree on the offset as on the original object. So that’s this loose option here. When this option is enabled and you have a surface that has the quad points in it, then first of all, the surface is going to split around those tangent positions, and it’s likely that if we run the Zebra’s on here that we’re going to see a discontinuity here because those tangent points would be lost. We’re only moving control points and there isn’t enough information in the loose option to maintain the tangencies.
So let’s say that we were offsetting here, a surface of mixed degree, and we wanted to actually maintain those tangencies. Well again, it’s not too difficult to achieve this. What we could do is explode the poly-surface that we get, and essentially discard three of the quadrants. Turn on the control points for the remaining quadrants, and we can use set points – set x, y, z co-ordinates to realign these points, snapping to the end points. So this is what I mean here. So we’re going to go to transform, set x, y, z co-ordinates and I want to align these to the x axis in world space here and just snap back to that end point. And then repeat the process here. Of course make sure that you have the point snap enabled when you do this. And then we’ll now do this for this pair of points in the other direction. So that pair of points now, and I want those to be in y, not x this time.
Okay so these points now if we look at them in top view for example, you can see these are all lined up with each other in x and y now. So they’re all giving us effectively these 90 degree conditions, which should mean now that when we mirror this back out around 0, that we have now a surface where our Zebras are consistent and our continuity has been regained.
Okay, so once we’ve created the offset, let’s just now concentrate on the next part of the design which is going to be the handle. So I’ll just turn on my elevations in place and I’ll look at the side elevation and the handle elevation and just look at these here and it looks as though the handle has got a completely consistent width to it here, and it looks as though the section thickness is constant as well. So this should be a fairly straightforward exercise and something that we can probably use a sweep 1 rail for. So I’m going to create some new curves here on my 2D curves layer, create a new layer for these and these are going to be handle curves. Make this active, and change the layer colour so I can see what I’m doing and make sure that my elevations in place are locked and then I just want to go to analyse and distance, and just look what the perpendicular distance is across here in a couple of places. Okay, so it’s 6.5 millimetres fairly consistently here.
So let’s again do something similar to what we did last time, and that is to use interpolated curve to create a curve here and extend the curve by an arc and then rebuild. You can see that curve is very close to our intent there. I only need to draw the outside curves in this instance. I’ll put the radius part in the smaller radius last of all. So I’ll build the three main curves here. So once again, rebuild and then finally this curve. Once again, these curves are all slightly over long. Okay so those are rebuilt now. I’m just going to see what, if we are reporting a radius back from these from our Illustrator artwork, roughly about 20 millimetre radius here and the same here actually.
So I think what I’m going to do here is just put a fillet curve in here with a radius of 20 and just have a look at this with the elevations in place. That looks about right, but the fillet of course is only tangent continuous at the start and end. So I’m going to take this out now and replace this with a curvature continuous blend. So I’m going to show the elevations in place so I can style the blend and curve fit it to our fillet as much as possible. But of course I want this additional continuity that I’m going to get. So I’m going to set the continuity to curvature which is G2 curvature. I’m going to take the middle point here and I’m just going to pull this back until I get a shape which is closer to the handle. Now of course, we won’t be able to match this exactly at this point now because our curvature continuity is going to give us a slightly different shape, particularly round about here, because what happens is that the radius here is gradually training, going from something that is about 20 here and going to almost an infinite radius here where we’ve got very little curvature going on. So we just want to make this blend look good, not look to stressed at the end of the blends. Okay so we’ll leave those blends a little bit fuller in the middle.
Okay, so let’s look now at the cross-section for the handle. So let’s turn on the elevations in place and look at the handle elevation here and just take a measurement across here, so 20 millimetres across the handle there. So our handle cross-section then is a rectangular section of 20 x 6.5 millimetres. Now the rectangular section isn’t going to look particularly good if we sweep this along the curve. Generally speaking, where you have these flat areas, this is going to look as though it’s almost probably going to be sunken in slightly. So I want to give the surface here a bit of crown on both of the two rectangular sides. So I’ll do that before putting in the corner radii of the blend which seem to be something of the order of 4 millimetres. So those 4 millimetre radii are going to go on after we’ve done the sweep. So let’s just go to the handle curves layer again and I’m going to draw in the basic rectangle, using a rectangle from the centre here, starting this at 0 and this is 20 millimetres wide x 6.5 millimetres high. Now in order to add a bit of crown on to this, I’m going to just take a line here at 0.25 of a millimetre and at this side, slightly less, 0.15 of a millimetre and use that to snap a midpoint of an arc too. So I’m going to use the three point arc here, which is start, end point on arc and repeat the process here.
Now the arc looks pretty good here but I need a bit more in the mass here in this curve, so that when this sweeps around and creates a double curvature surface, I’ve got a little bit more room for blending edges and matching curvature. So I’m going to rebuild this as a freeform curve or a spine, again a degree three, four control points and you can see here when we preview the rebuild that our deviation is extremely negligible, and repeat the process here. So these are both free form curves. So I’m going to mirror this using the x axis prompt here and mirror this one using the y axis prompt. Okay and then I’m just going to draw up a little point here, that’s going to be the point that sits on the rail and I’m going to remove the rectangle and just join these four arcs together.
Okay, now I can think about orienting this on to the handle curve. So for this I’m going to use the orient perpendicular to curve command and I always like to use this command with the curve or object that I’m going to orient in a similar position and I do this for both orient perpendicular to curve and orient on surface. So what I tend to like to do is have the cross-section in this place so it’s lying planer with the active construction plane so that the perpendicular direction is actually the vertical direction in this instance. So I know that what’s vertical is going to be perpendicular when I do the orientation. So transform, orient, perpendicular to curve, pick the objects to orient, the base point and then snap to the curve that I want to orient too. Okay so I’m going to enable to copy option here and I’m also going to enable the rotate option. Move this right to the end of the curve and then because I’ve got my ortho on here, it’s very easy to just rotate to 90 degrees and I should only need to have one cross section here to propagate that correctly along all of these rail curves here.
So let’s create a new layer on our surfaces and call this handle surface and make this active and then look at our sweep. So I’m going to use a one rail sweep and I’m going to use the chain edges option, which means that these various curves don’t need to be joined together. If I have the auto-chain option on, anything that is tangent or above is going to pick in the chain. So I can just pick one curve and all the other curves which are tangent or better, continuous will pick. Then I enter and pick the cross-section curves and enter to get to the curve seam option.
Now the sweep in this case is going to produce a poly-surface which is comprised of four separate surfaces. Each one of these has four curves. So I want the curve seam to be in a corner. So that’s in a good position there. So I’ll enter to get in to preview. I’m going to use the freeform style of sweep here which allows this cross-section to move about the curve, always staying perpendicular to the curve and of course use the do not simplify option here, and then okay to accept that. So I can then just turn off the curves temporarily, and cap the ends of this and then add the fillets. Now we had 4 millimetre approximate radii on the outer edges which means we’re going to need obviously to put a smaller radii on this inside edge and I’m going to look at using a blend rather than a fillet here, just to give me some slightly better continuity. But I’m still going to use a solid command for this. So it’s solid, fillet edge and blend edge. I am going to pick a radius of 4 here, set a radius of 4 and pick an edge. Okay, and let’s enter to get to a preview and just have a look at whether we go for distance between rails or a rolling ball fillet here. It shouldn’t really make any difference here because our angular condition here between the two edges where I’m setting back to fillet is actually consistent. So rolling ball will actually give me something that matches more closely the description that exists in Illustrator.
So that builds okay, so let’s repeat this on the other side, and let’s put a smaller radius, maybe a 2 millimetre radius on the inside edge, again with a blend. Okay, and let’s quickly check this over with the environment map. All we’re checking here is that the blend just looks smooth and that we haven’t really got any little kinks or creases going on where we’ve filleted. So that looks okay and this is how it’s going to fit in to the body. So that’s the handle finished for now.
Thanks for watching and please do check out the second and third parts of this tutorial.
Rhino3d for Mac Tutorial Series - Curated by Simply Rhino to help you get started and learn Rhino on the Mac platform.
This is the video transcript, published alongside the video, to further support your Rhino learning experience.
Hi. This is Sean from Simply Rhino. Welcome to today’s tutorial, in which we’ll continue with our series in order to help you learning Rhino for Mac. In this tutorial, I’m going to be looking further at some Boolean commands.
Okay, now let’s start by just creating a box. So, I’m going to slide up to solid box, corner to corner. I’ve got grid snap on, and I’m just going to make a couple of left clicks here in the top view to define the base and then I’m going to define a height by just dropping down in to the front view here with my cursor. Now let’s change the type of view that I have here to a ghosted view and I’m going to place a sphere on the corner here. So I’m going to slide up to solid, sphere, send to radius, and make sure that I’ve got my end object snap on here. So I’ll make a left click and I’ll slide down to the front view here and snapping to the grid, placing my radius at about nine o clock, therefore the seam of that sphere will be placed in an orthogonal position.
Now, I’d like to just start by showing you this particular command, Boolean two objects. Obviously the limitation on this command is that it will only work with two objects. So if I just execute that. Now top left hand corner Rhino is asking me to select two objects to Boolean and note the delete input is ticked here. I select one object, and another. Now Rhino asks me to click to iterate through Boolean results, click enter to accept. What that means is that if I make a series of left clicks, Rhino is going to cycle through the Boolean possibilities, and you’ll see in the bottom left hand corner here of the interface, intersection. So Rhino is reporting to me what type of Boolean I’m viewing here. So I make another left click, bottom left hand corner, A – B. So that’s the different, as if I was using the Boolean difference command. I make another left click. That’s another difference but it’s the part that’s been removed from the sphere. I make another left click, inverse intersection. That’s interesting because that particular one isn’t available from the drop down menu. I just cycle through. I’ve got a union, intersection, A – B, B – A, inverse selection. That’s probably the most interesting one.
So I’m going to just press enter now and it would have left me with the inverse selection. So let’s just see what that is, an inverse selection. I just drag this object away and you’ll see. It’s removed the intersecting part between those two objects.
Now what I would like to do now is, I’m just going to type in command Z and again, I’m going back to my two original intersecting objects. Now I mentioned in the last video that it’s of benefit if we try and carry out Boolean’s without directly using the Boolean commands. So what I’d like to do is show you a few examples of precisely that. So I’m going to select those two objects and using my gumball here, I’m going to just make a few copies. So what I’m doing is, I’m holding down the ALT key. That’s going to make a copy and again, holding down the ALT key and I’m just dragging that constraining red arrow direction.
So just as a repeat of previous, I’m going to use my… so this is a reminder. I’m going to come down to curve from objects, and I’m going to choose intersection. So, one, two, I press enter. Now that’s left me with a closed curve and as we saw previously, we can use that particular curve to carry out what would have been a Boolean difference command by using the trim command. So if I click on the edge of the cube here and then the top of the sphere, all that leaves me with is to join these two objects together. So now I click on join. So that was carrying out what would perhaps have been a Boolean difference, but using the intersection command followed by trim and join.
Now let’s see how we can carry out something similar here but simply using the split command. So if I execute the split command, Rhino is going to ask me, select objects to split. I choose the sphere as the object to split. I press enter. It asks me for the cutting object, I choose the sphere. I press enter. Now what I’m going to do now, I press enter, I’m going to now carry out the reverse. I’m going to choose my sphere as the object I’m going to split and I’m going to split it with the box. Press enter.
Now, what’s happened here? So I’ve split one object with the other and then the other object with the other. So if I hide this sphere here you’ll see, and if I hide the corner of this cube, you’ll see what I’m left with. If I join these two parts together now, holding down my shift key, I can type in JO, press enter, I’ve carried out a Boolean difference. If I click on my show command here, you see it brings back my sphere, and if I delete these now, these two parts here.
So I’m just showing you how you could carry out a Boolean difference by simply just using the split command. Now how else could we carry out this? We could use the trim command.
If I use trim, if I execute the trim command, select cutting objects. If I say that my cutting object is the sphere here, I press enter. Because I’m in a ghosted view it means that I can click on the edge of the cube here and it will trim that part away. I press enter. Now all I’m left to do is to use this object as my cutting object, press enter and then I click on the edge of the sphere. Press enter, and again, all I need to do is join.
So there you are. I recommend that you carry out these Boolean type commands, whilst exploring trim, split and join and intersection. I hope that was useful to you and I look forward to seeing you again. Thank you.
Rhino3d for Mac Tutorial Series - Curated by Simply Rhino to help you get started and learn Rhino on the Mac platform.
This is the video transcript, published alongside the video, to further support your Rhino learning experience.
Hi, this is Sean from Simply Rhino and in this tutorial, in order to help you learn Rhino for Mac, we are going to be looking at some of Rhino’s Boolean commands.
Okay, now these are the commands that we are going to be looking at. Click on Solid, and I slide down. I’m looking at Boolean Union, Difference, Intersection and Boolean Two Objects.
So let’s start with a box, solid box, corner corner height. Note the grid snap is on. Top left hand corner in bold, Rhino is asking me for first corner of base. I’m going to define a box whose base is in the top view, four units by four units by four units. So right click on perspective. Let’s change that actually to ghosted.
Now I want to place a sphere on the corner of that box. I slide up to solid, sphere and I’m going to choose the centre radius method. In bold, top left hand corner, Rhino is asking me for the centre of sphere. I find the end object snap in the perspective, left click. Now I bring my cursor over to the front view and define a cube say so big. Note that I’m snapping to the grid which means that the seam, this double thickness line here that I see on the sphere, is horizontal. Okay, I’ve inadvertently moved that, command Z on the Mac keyboard is undo.
Okay, now what I’d like to do is make a series of copies of these objects running along the x axis. If I make my selection, and what I’m going to use is transform, array, linear. It’s going to ask me top left there for number of items. I don’t need to swipe across that box that contains two. All I need to do is type in six and that’s going to overwrite that. Now I’m going to define first reference point. I’m going to choose a point here and just observing in the other views, particularly the front view, just to see the spacing between them. Left click to define the second reference point.
Now let’s have a look at the very first Boolean command in that list. Solid, union. Now I chose to use the ghosted display for a reason. Actually, I’ve just pressed escape. You can see the corner of the box inside the sphere there. So let’s just note what occurs here. So solid, union. I choose the box; I choose the sphere. Note that it’s a single selection. I press enter. Now note that the inner part, that the corner of that cube has been omitted and if I select that object and I come over to my properties here, note how Rhino describes it as a closed polysurface, a single closed polysurface.
Now let’s have a look at another Boolean. I’m going to slide up to solid. I’m going to choose difference. Now in this example, you’re going to see that Rhino is going to ask for select surfaces or polysurfaces to subtract from, okay, that’s a key word. If I want to be left with the impression that the sphere makes in the cube, I’m going to need to select the cube first. I select the cube, I press enter. Now Rhino asks me, select with. With being the key word here. I want to subtract with the sphere. I select the sphere, I press enter and you see Rhino leaves the impression that the sphere was making in the cube.
Now let’s have a look at the next one. Let’s come up to solid and intersection. First set, press enter, the cube. Second set, I choose the sphere, I press enter. It leaves me with the intersecting part. Now I’d just like to just talk a little bit about what’s happening here with Booleans because Booleans are an automation of up to five commands in Rhino. Now, one of the commands that is probably the least obvious is this command here. Okay, it’s found under curve, curve from objects, intersection, and I’m going to select my box and my sphere and I press enter. Actually, let’s expand the perspective. Okay, I press enter. Now what the intersection command does, is that it creates an object at the intersection of objects. Press enter. So it’s left me with a curve and if I observe in my object properties here, it’s left me with a closed curve. Now as I said, Boolean commands are an automation of up to five commands and one of them is this particular command.
Now let’s just see how we could manually go about carrying out Booleans. This is important. This is important because Booleans are not always successful and it’s important that we understand how Booleans work and it’s important that we’re able to carry out Booleans without Boolean commands. So let’s just do exactly that.
Now if I wanted a Boolean difference, let’s take advantage of the fact that my curve here, my intersection curve is selected. So if I choose trim now, slide across to the left here and I choose the trim, it’s not going to ask me for a cutting object because my selected object will act as the cutting object. It goes straight in to select object to trim. I click on the edge of the box here and now I click on the edge of the sphere. Now my curve is selected, I press escape a couple of times, then I select the surface and the remaining part of the box here. All I need to do now is run the join command, and there, I’ve created a Boolean difference without using the Boolean difference command and look, over in the right hand side it’s a closed polysurface.
So what I suggest you do is to explore the Boolean commands, but without necessarily using the Boolean commands. The commands that Rhino uses are as I’ve just shown you, intersection, join and trim and maybe split. So have a play with Rhino’s surfaces because what these Boolean commands are illustrating to us are that Rhino is indeed a surface modeller.
So I hope you enjoyed that exercise. My next tutorials will be on layers and layouts. Thank you, good bye.
Rhino3d Video Tutorials Transcripts - To further support you as you learn and progress with Rhino we've transcribed each of our video tutorials.
Hi this is Phil from Simply Rhino and in this video, we’re going to take a look at how Rhino3d can be used in conjunction with a parametric solid modeller, in this case, SOLIDWORKS. The idea is to create the a-surfaces or styling surfaces in Rhino and the engineering detail in SOLIDWORKS. We’ll look at how we can make effective use of file referencing to allow us to make updates and changes to both the Rhino styling surfaces and the manufacturing details contained in SOLIDWORKS without having to do any time consuming remodelling.
Let’s take a look at some of the basics of the Rhino and SOLIDWORKS workflow. Theoretically it’s possible to round trip Rhino and SOLIDWORKS native files but as you can see here, this in practice is not straightforward. SOLIDWORKS 2016 will only read Rhino version 4 files and Rhino version 5 will only read SOLIDWORKS parts and assembly files up to the SOLIDWORKS 2015 format. Thankfully, in the real world however, this isn’t too much of an issue, as the workflow with both step and IJIS files works extremely well and in some cases, better than using native file formats.
Let’s look first at something very simple. In Rhino I’m going to create some basic geometry and then read this in to SOLIDWORKS. So here we have a rectangle that is 50 x 100mm and I’m going to extrude this to a height of 200mm. Then I’m going to save this down as a version 4 file, just call this ‘box’ and then we’re going to open SOLIDWORKS and read the file in to SOLIDWORKS. So I can open the file directly here, by going to file, and open and finding my Rhino file and this will bring the geometry in to SOLIDWORKS and then SOLIDWORKS will perform a check on the geometry and here if we have any faces that are corrupted or edges that don’t join together, we’ll have the opportunity to fix these. Obviously with this simple geometry there’s no problems with it.
Two things to mention that are fairly obvious. First of all, SOLIDWORKS along with other engineering software, uses the idea of Y being the vertical axis rather than Z being the vertical axis. So in real terms, your geometry is rotated about the real world origin through Z to Y. The other thing that is slightly disconcerting for Rhino users is that the 3-Dimensional view in SOLIDWORKS, by default, doesn’t have any perspective applied to it. We can turn on perspectives here in SOLIDWORKS and out geometry looks a little bit more familiar.
The component will come in as an imported reference here and I can change the name of this component should I wish and then for example I could add some features to this. So let’s add some fillets here. I’ll just fillet these long edges here and accept this, and then let’s put a smaller fillet around the base and let’s then for example, shell this out to a thickness of 4mm. Okay, so that’s our geometry. So what we have are parametric features on top of a Rhino reference. I’m going to save this file as ‘SOLIDWORKS part’ and now if I want to actually update the Rhino reference, I can right click on my relevant feature here and I can edit this directly in Rhino. This will open a new instance of Rhino and the relevant file.
Okay, so now we’re in Rhino, let’s make a very simple change to this piece of geometry. Let’s turn on the gumball, use our sub-object selection and just reduce the height of this box by 100mm and then save this back out again over the top of our existing file and then we can go back to SOLIDWORKS and I can right click on my reference and I can refresh and this will update the geometry.
Okay, you’ll see here that a feature here has failed because the reference that I used to create the shell in the first place has moved. So on more complex geometry, it’s a good idea to supress the SOLIDWORKS features that you’ve added when you change the reference and then add them one by one. Here, I can edit the feature and re-insert the face that I want to shell out from and my feature will rebuild again. So we have parametric features on top of a Rhino reference that we are able to change.
Let’s now have a look at moving some geometry via a neutral file format. Here I’ve got a simple box that’s made from planer surfaces with the addition of some constant radius, tangent continuous fillets. There’s a number of steps that it’s worth going through before exporting the geometry to SOLIDWORKS and it’s a good idea to use solids in to SOLIDWORKS. So first of all we need to check that our geometry is watertight and I can do this by using the show edges command. This is found in analyse, edge tools and show edges. Just grab the geometry here and enter, and this will open the edge analysis dialogue. Here I just have naked edges selected with a nice bright colour here and you can see that essentially what’s happening here is that the top face that essentially joins the inner and outer wall of my box here is currently detached. So what I need to do is just join this all together in to a solid. This reports back that it’s a closed polysurface. I can check the edges again but if Rhino reports it’s a closed polysurface then there’s no need to do this.
A couple of other checks that it’s worth doing. First of all, I want to make sure that my geometry all passes check. So I can do this by either selecting bad objects here or by picking the geometry here and running the check command and then because my geometry contains trimmed surfaces here, I can run a command called shrink trimmed surfaces and what this does is that minimises the untrimmed area of the surfaces. So this command is in surface edit tools and it’s called shrink trimmed surface and I can select the whole polysurface in one go here.
Next, to keep the geometry in the same relative orientation once we’ve moved it in to SOLIDWORKS, I need to rotate about the world X axis, through 90 degrees. So I can do this by going in to my right viewport here, picking the geometry and going to transform and rotate, typing in zero for the centre of rotation and then rotating through 90 degrees so my Z effectively becomes the Y axis in world terms.
Okay now we’re ready to export the geometry to SOLIDWORKS so, what I’m going to use here is a step file and give this a name, and I’m going to pick the third of the options here which is AB214 Automotive Design CC2. The options here progressively give me more information in the step file. Okay so now I can go in to SOLIDWORKS and I can open the file, so it’s Tray SDP and you’ll see that the part comes in to SOLIDWORKS. Once again, we have the import diagnostics dialogue box which we’ll say yes to and our geometry passes check here. There’s nothing that needs fixing on it so I’ll okay that.
Now one of the things that we can do here, where we have fairly basic geometry and by basic I’m really talking about geometry that has rolling ball, constant radius fillets, planer surfaces and fairly regular features. If this is the case, we can proceed with feature recognition in SOLIDWORKS and what this does is that it attempts to create parametric features from some of our Rhino surfaces and if I just okay the automatic mode on this, then what should happen here is that SOLIDWORKS will create a plane and then it will create some features relative to that. And so now our original imported piece of geometry is no longer there and we now have a number of SOLIDWORKS features. Now sometimes, these features may not exactly equate to how we built an object but in this case for example, if we look at the fillets here on the outside, these features are in a way parametric so we can change dimensions on here. So let’s reduce this radius here from 20 to 15 for example. Okay, and you’ll see those radii change there.
Now one of the things that you’ll see here is because this object has had features added to it in a reverse sort of way, there is not a connection between the inner and outer fillets here. So we need to find those inner fillets and also edit that feature too and reduce that feature also by 5mm.
So it’s not a complete solution but it does have the ability to parametrises some basic Rhino geometry. However, for the most part, the real value of using Rhino is in actually bringing in geometry that would be very difficult to create in SOLIDWORKS. So generally speaking, we wouldn’t want to use the feature recognition tool. Now let’s take a quick look at exporting via IGIS.
So once again, I’ve got a couple of solid pieces of geometry here. I’m going to rotate these about the world X axis, through 90 degrees Z to Y and then take this geometry out via an IGIS. So export selected, pick IGIS and give this a name. And I’m going to use the SOLIDWORKS solids file type here and then let’s open SOLIDWORKS and open the IGIS file. Okay, and again we’re prompted for import diagnostics and we’ll say no to the feature recognition.
What you’ll see here and this happens with any file type, is that if you have for example two solid bodies in the Rhino file and you opt not to use any feature recognition, then you’ll have two separate references here and this is actually useful in certain circumstances and we’ll look at this when we look at the engine cover model. You’ll also see with the IGIS files, that the IGIS files maintain the object colour that was applied to them inside of Rhino, and as the default object colour is black, then it can make these objects a little hard to read in SOLIDWORKS. So to change this, we just need to go to our display manager tab here and then we can just remove the appearances that actually came in with Rhino and get these back to the default SOLIDWORKS appearance.
Now let’s look at bringing in some more complex real world geometry in to SOLIDWORKS. This is the engine cover shown at the start of the video and this is built from considered slab surfaces that whilst all being double curvature have been optimised to ensure a relatively good surface quality. The transitional surfaces are all G2 curvature continuous blends and creating an object like this is really the preserve of software like Rhino. Whilst we could shell out the moulding and add features such as holes and edge fillets in Rhino, the benefit of doing this in SOLIDWORKS is that of course these features remain parametric. So moving holes or changing the shell thickness should be relatively straight forward. This geometry is watertight, check and all the trimmed surfaces have been shrunk.
It’s a good idea to work on a copy of the file to export or as in this case, to copy the geometry on to a new layer before exporting. With exporting, the general rule is that step works well for simpler objects but that IGIS is better for more complex entities. So in this case, let’s use IGIS. So I now have the engine cover in its correct orientation ready to go in to SOLIDWORKS but before I export this, I’m going to create a small piece of secondary geometry here and this is going to come in as a second referenced object in our SOLIDWORKS model and the idea of this is that this gives us the opportunity to bring in a second object later on in to the SOLIDWORKS file if we need to do so. An example of why we might need to do this would be if we had a piece of geometry that was difficult to create in SOLIDWORKS and we needed it as a reference for a feature that was going to be created in SOLIDWORKS. We could create this in Rhino and export this out and replace the reference of this cube with our new piece of geometry.
So I’m going to take both pieces of geometry now and export them out as an IGIS file and I’ll call this ‘engine cover’. Let’s call it ‘engine cover 01’ and we’ll use SOLIDWORKS solids again and then we’ll open our SOLIDWORKS and open up this file. So our geometry comes in and we’ll be prompted to run import diagnostics and we don’t have any issue with either of the two parts here. We don’t want to proceed with feature recognition, we want to keep these two as references. In the case of this object, our bonnet here, then feature recognition would really fail to do anything here because we have got an object which is built entirely from complex surface.
The first thing I need to do probably is go in to my appearance manager and just remove the black appearance that’s come in from Rhino and then just change the display style here to make this a little easier. Then I’m going to name my imported feature here. So this is the ‘engine cover’ and this here is my ‘spare part’. So not really interested in that spare part at the moment, so I can supress that feature here but it remains in the tree but it is just hidden. So now I can add some features to this geometry.
So we’re going to start off by adding a shell to this and the shell command does something very similar to the Rhino shell command. So I’m going to give this a thickness and then just as in Rhino, I nominate the faces that I want to remove as part of the shell process. Now this area here has been modelled, particularly so that we can replicate what would be a kind of sensible C and C trim in the case of this GRP moulding. So we remove the base and also the back face as well. When we’ve picked all those faces that we want to remove, I’ll enter this and create the shell.
Now typically, in SOLIDWORKS, a command will either work or fail rather than in Rhino, where the command will fail but give you the parts of the geometry that it can make. But just because the command has actually worked in SOLIDWORKS, doesn’t necessarily mean that all the surfaces that are created, for example by the shell command are going to be perfect and to that end it’s worth just having a look at the underside of the shell here and checking the layout of all of these constituent parts.
Typically features like fillet which resolve to a point would be difficult to offset and again very similar to Rhino, features for example, a small fillet that shelled inwards where the original radius was less than the thickness of the shell would potentially cause problems in SOLIDWORKS as well. So it’s worth just checking that geometry out. Then we can go on to add some further features. So I’m going to add some fillets now and we only need to add some small fillets here, really to help the moulder out. We want to avoid sharp edges wherever possible on this, or seen sharp edges so that we don’t have any parts of the moulding that are breaking off when it’s released from the tool.
So I’m going to put a 6mm fillet here on the short corners of the features. So just so I don’t pick the wrong edge here I’m going to disable tangent propagation and I’m just going to pick the six areas here. So you can see that I’m something of a novice user in SOLIDWORKS here but should at least show you the workflow between the two pieces of software.
Okay, so those are our corners done and then I’m just going to look at doing the outside edge. So again, another fillet here. Let’s give this a 2mm radius, turn on the tangent propagation this time and pick this outer edge and all that edge should highlight and run a 2mm fillet all the way round the edge.
Okay, next up I want to go to my top view and zoom in to this area. This is where we’re going to put a hinge to open the engine cover and we want to put some holes in here to accept the hinge. So I’m going to add a sketch here and I’m going to make that sketch relative to the top plane and I’m going to then choose a centre rectangle. So to add the hole detail, first of all I need to create a sketch in SOLIDWORKS and I’ll create a centre rectangle and I need to make this relative to a plane so I use my top plane here and I need to make this a sketch with some numeric input. I’ll find the centre of this plane here and drag outwards and I’ll make this 60mm in the one direction and 100mm in the other dimension and okay that. Then I can exit the sketch and then create the holes as a feature. Make these holes 12mm, position them on this face and add the holes and accept that and then let’s switch back to a 3-Dimensional view and let’s mirror that whole feature to put it on to the other side. So we’ll mirror, and we need to pick a plane about which we want to mirror, so that will be the right plane and then the feature that we want to mirror here is going to be the hole and we’ll accept that.
So you can see now that we can add a number of parametric features in to our model now. So we have a Rhino reference and plenty of parametric features added to that. Now in the real world, what we might have is a more complex part than this and let’s say between 50 to 80 parametric features that are added in to our SOLIDWORKS model and there are times where your customer or your co-worker or your client may request a modification to the geometry that can only be done in Rhino.
So let’s say that we were required to put some scoops in to this bonnet which needed to be curvature continuous and controlled very tightly with the existing geometry. Clearly this is something that we’d struggle to do in SOLIDWORKS and yet we’ve done all this work in terms of adding features and we don’t want to lose that work. So this is where we can actually, using a careful workflow, where we can actually update the reference and re-propagate our features.
So first of all, I’m going to save the file and save this as ‘engine cover 01’ and then I’m going to go back to Rhino. So here’s the same part in Rhino and let’s jump to the modification that we need to make to this. So what we’ve done now is added the scoop and mirrored this over to the other side. So again, in Rhino this is a curvature continuous feature and this again is something that would be best done in Rhino rather than doing it in SOLIDWORKS.
An additional complication here, is that also in the scoop we want to add a particular cut out to this and of course, the shell is actually being done in SOLIDWORKS but we want to actually be able to reference the shape for the cut out inside of Rhino and let’s say the part that we want to produce our cut out with looks like this. Because we’ve got this spare reference in the file, this actually gives us a way out of doing this. So if we look at the same parts in their orientation ready to go in to SOLIDWORKS, what we can do is to save this out as two separate IGIS files. So we’ll take this file here and export it out and we’ll call this one ‘engine cover 02’. And we’ll take the part that we’re going to use as our Boolean cutter and we’ll export this out as ‘cutter’. Okay, and now we can go back to SOLIDWORKS.
So inside of SOLIDWORKS now, the best workflow here to do a save as, to preserve our original file and we’ll save this as ‘engine cover 02’. Now when we’re bringing in complex geometry, there’s a chance that some of these subsequent features may not work. So to avoid having some sort of calamity in SOLIDWORKS, what we can do is we can essentially pick all of the features that we’ve added and right click on them and supress them. So we get back down to our starting Rhino geometry and I’m going to unsurpress the ‘spare part’ reference in here. And first of all I’m going to change my reference for the engine cover. So right click on the engine cover reference here and click on edit feature. This will propagate a warning telling me this is a reference file and I accept that warning and I replace my geometry which was ‘engine cover 1’ IGIS file with ‘engine cover 2’ IGIS file. And the Rhino geometry will now come in and my reference will be replaced and you see I’ve got my new part here. Then my spare part, I can also edit that feature and replace that with the ‘cutter’ and that cutter comes through in position.
Let’s just go to the display manager and remove the black appearance and change the display style. It’s important to understand that although the definition of the vertical axis is different in Rhino and SOLIDWORKS, the absolute position of the parts in 3D space is the same. So that if the relationship between the cutter and the bonnet was correct, then it will be correct in SOLIDWORKS. There is one caveat to this, is that if you are working on a file and bringing in new parts, or new references then be careful about working in perspective mode, because we’ve seen in some circumstances that perspective mode has troubled SOLIDWORKS idea of absolute, 3-Dimensional space. In other words, what’s happened is that we’ve had perspective on, brought in an object and it’s come in at the wrong position. So as long as we’re working the default parallel projection we should be okay with bringing in geometry like this.
I’m going to change the name of this part now to ‘cutter’ and for the moment I can supress that cutter because I’m not going to use it at the moment.
Now what I can do now one by one, is to unsurpress these features. Okay, we should find that all of these will work but it’s a good methodology to turn on these features one by one. So I have the holes back, and then the mirror back. You should see that all of these files add correctly. So now we have our new shape brought in to SOLIDWORKS and we can then turn on the… or unsurpress the cutter and we can either use this as a reference in SOLIDWORKS to create a sketch and to do a cutting operation with that, or we can actually use this geometry to… as a Boolean process. So let’s do the latter to make sure that we’re actually using the Rhino geometry here rather than a repurposed version of it.
So first of all let’s run the mirror command and we want to mirror a body not a feature and for the mirror plane, we want to use the right plane and for the body to mirror, we want to pick our cutter. So our cutter now exists in its opposite hand version. Then we can insert a feature called a combine and we can select the main body here and the subtracted features here, making sure that we’ve got subtract selected here and we can produce our cut. So now we have a component that has a whole set of parametric features and whose shape we can reference and update by changing the original Rhino file.
So in the real world, this gives us a pretty flexible situation where the basic shape can be updated inside of Rhino and the parametric features can be updated inside of SOLIDWORKS and the sets of features will co-exist quite happily.
I hope you’ve enjoyed this video, please look out for further videos from the team at Simply Rhino.
Open a new Rhino file - choose Small Objects Millimetres from the template options.
Import the STL data into Rhino - File > Import (choose STL from the Files of Type dropdown menu) and navigate to your STL file.
If the scanned data needs to be reoriented in 3D space this can be done with the Transform tools for instance Transform > Move or Transform>Rotate. It is important to carry out any reorientation of the model before work begins.
The scanned data should come into Rhino’s Default Layer - Rename this layer 01 Original Scan.
Create a new layer and name it 00 Foam Volume. Move this layer above 01 Original Scan in the layer palette - you can move layers up and down in the list by using the up/down arrows in the layer palette.
Make 00 Foam Volume the Active layer.
Create the foam block using Solid>Box>Corner to Corner, Height. Use a standard foam block size. The aim in creating the foam block now is to create a boundary for any subsequent curve or surface extensions to intersect with.
Hide the layer 00 Foam Volume and Show the layer 01 Original Scan. (Lightbulb in Layer Palette).
Create a new layer and name it 02 Contours. Make this layer Active.
Working in the Top Viewport and with Ortho constraint On, Run Curve > Curve from Objects > Contour to create the contour curves. At the Contour Plane Base Point prompt click on the left hand side of the scanned object and at the Direction Perpendicular to Contour Planes prompt drag the cursor to the right (X axis) and click. Finally at the Distance Between Contours prompt type in 20. The contours will now build.
In some circumstances it may be beneficial to produce contours further apart than 20mm.
The resulting contours are Polylines and will need to be Rebuilt to create smooth splines from which an editable surface can be created.
If any of the polylines have gaps in them then create a new ‘repair’ curve using either a straight line segment or a blend curve (continuity = tangent) to fill in the gap.
Join the polyline segments and repair curve together to form one polyline.
The completed contours should look similar to this:
The small outer ‘orphan’ curves from the perimeter of the scan can be removed along with some of the rear contour curves which are not needed.
Before rebuilding these remaining curves the edges of the contours can be trimmed. As the ultimate aim is to extend the edges of the scanned surfaces vertically, any of the contour curves that double back on themselves can be trimmed at this stage.
On the example file most of the contours on the left hand side of the seat can be trimmed. This can be done approximately by working in Front Viewport and trimming with a horizontal line placed roughly at the high point of the curves.
Hide the layer 01 Original Scan.
Create a new layer and name it 03 Rebuilt Curves make this layer Active.
The curves do not need to be rebuilt individually.
Pick a representative curve and run the Edit > Rebuild command. The Rebuild dialog box will appear. The bracket figures refer to the points and degree of the existing polyline (the point count will likely be around 500 and the degree will be 1) and the figures to the right refer to the new rebuilt curves. Change the Degree from 1 to 3 and set the number of points to around 18 to 24. A rebuilt curve will deviate from the original polyline - you can check this deviation by using the Preview button. Aim for a deviation of around 3mm. As a general rule the more points in the rebuilt curve the closer it will be to the existing polyline but the more complex any surfaces created from it will be - the aim is to reduce complexity whilst getting close to the original polyline.
Do not complete the command!
Pick another curve and repeat the above process to get a good idea of the number of points to use.
Finally pick all the original curves and rebuild them in one go using a degree of 3 and the number of points obtained from the above try-out. The example file uses 18 points.
Important! Use the following options:
Delete Input = Unchecked - Create New Curves on Current Layer = Checked
The rebuilt curves should look similar to this:
The Rebuilt curves now need to be extended so that when a surface is created from them the surface will intersect with the foam block.
Hide the layer 03 Rebuilt Curves
Create a New Layer and name this 04 Extended Curves. Make it Active.
Copy the Rebuilt Curves onto this new Layer. The easiest way to do this is to select the curves on layer 03 Rebuilt Curves and then in the Layer Palette, Right Click on the 04 Extended Curves layer and select Copy Objects to Layer.
Hide the Layer 03 Rebuilt Curves.
The curves are best extended in the orthographic viewports and in the sample file the best view is the Front View.
Depending on the curve type there are two methodologies for extending the curves as shown in the following illustration:
Working in the Front View, draw a line in the Z axis as shown.
Working on one side only create a perpendicular line between the Z axis line and each of the curve ends.
Use the command Curve > Line > Perpendicular to 2 Curves to do this.
It is possibly easiest to do about four or five of these perpendicular lines and then move on to the trimming operation.
Next, working on a curve at a time, Trim the end of the curve with the appropriate perpendicular line using Edit > Trim.
Delete the perpendicular lines after they have been used to trim with.
Repeat the process until all of the curves have been trimmed on one side and then use the same process on the other side.
Turn on the layer 00 Foam Volume and Lock it (padlock symbol in layer palette).
Switch to the Perspective Viewport.
Now working on a curve at a time use Curve > Extend Curve > By Line to extend each curve beyond the perimeter of the foam volume.
Because the curves have been trimmed with a perpendicular line the extension will be tangent continuous and also vertical.
Working in Perspective Viewport, draw a vertical line by using Curve > Line > Vertical to C Plane making sure that the start of the line is snapped to the top of the curve by using the End Object Snap. Make sure the end of the vertical line extends past the perimeter of the foam volume.
Repeat this process for all curve ends that need extending in this manner.
Currently the junction between the vertical line and the rebuilt contour curve will be sharp and if we build a surface from this condition the surface will have a crease in it. To smooth out this sharp corner condition we can use a fillet curve. The dimension of the fillet curve will depend on the job in hand but in the case of the example file about 50mm works well.
Still working in Perspective Viewport run Curve > Fillet Curves and set the following options:
( Radius=50 Join=Yes Trim=Yes ExtendArcsBy=Arc ):
First pick the end of the rebuilt contour curve and then the start of the vertical line (pick both entities near the sharp corner) and the fillet curve will build and join.
Repeat this process as necessary.
Working in Front View, Trim the top of all the extended curves so that they are of uniform height about 5mm or 10mm above the foam volume.
Finally, if the first curve (pommel end) is inside the foam volume then make a Copy of this curve positioned further along the Y axis so that it is outside the foam volume. If this distance is significant you may need to position an interim curve (again copied).
The result should look like this:
We could create a valid surface by lofting our new extended rebuilt curves together but the result could be quite torturous and difficult to edit subsequently. A better methodology is to create a final surface with a clear rectangular isocurve structure and we can do this by creating interpolated Y axis curves.
Note that in comparison to the previous process on the Back Foam we will have less input curves here due to the shape of the scan.
Create a New Layer and name it 05 Y Axis Curves. Make it Active.
Working in Top Viewport create an array of lines in the X axis that are 20mm apart.
To do this turn on the Project Constraint and draw a Line in the Y axis that is snapped to the X axis centre of the foam block. Make sure this line extends past each side of the foam block.
Move the line upwards in the X Axis by 10mm.
Using Transform > Array > Rectangular array copies of this line upwards in the X axis at 20mm.
In the attached file the number of items would be as follows:
X=6, Y=1, Z=1.
Accept the result of the Array by pressing Enter.
Now, still working in top view, Mirror the 6 arrayed lines about the X axis centre of the foam block. We should now have 12 Lines at 20mm apart.
Working in Front Viewport Move these 12 lines either above or below the Foam Block.
Next, Create a New Layer and name it 06 Y Axis Surfaces. Make it Active.
Select the 12 lines and switch to Perspective Viewport.
Extrude these lines in the Z axis making sure the resultant surfaces are slightly taller than the foam block.
To extrude the lines use Surface > Extrude Curve > Straight.
Create a New Layer and name it 07 Intersection Points make it Active.
Turn off Layer 05 Y Axis Curves and 00 Foam Volume.
Create a set of points where the surfaces intersect with the Extended X Axis Curves. To do this run Curve > Curve from Objects > Intersection. This command may take a few minutes to complete.
When the points have been created turn off layer 06 Y Axis Surfaces.
Create points on the ends of the extended rebuilt curves by running these two commands:
Curve > Point Object > Mark Curve Start
Curve Point Object > Mark Curve End
The result should look like this:
Create a New Layer and name it 08 Interpolated Curves Y. Make it Active.
Turn Off Layer 04 Extended Rebuilt.
Working in Top Viewport, run Curve > Free Form > Fit to Points and marquee select a row of points - use the following command line options to create a smooth curve through the points:
Degree=3 CurveType=Interpolated Knots=Uniform Closed=No
Repeat this for each row of points until all the curves have been built.
Turn Off Layer 07 Intersection Points and Turn On Layer 04 Extended Curves X. There is now a network of Curves from which the editable surface can be created.
In this instance we need to create a surface that can be edited easily so, for example, we can raise the pommel area. Surfaces can be directly edited by their control points; however in this instance the surface we will create will be relatively complex and have too many control points to edit it smoothly.
The answer in this case is to use Rhino’s Modelling History. In this case this can be seen as an association between the curves and the surface that is built from them. If the surface is built with History On then the surface will update if the curves are subsequently edited. Since editing the control points on the curves is much simpler than editing the control points on the surface this represents good practise.
We will also create a reference surface without history.
Create a New Layer and name it 09 Reference Surface. Make it the Active Layer.
Select all the curves and run Surface > Curve Network to create the surface.
In the dialog box that pops up select the following options.
Edge Curves = 1
Interior Curves = 1
Preview = On
If the surface looks okay in preview Select OK to accept the surface.
Create a New Layer and name it 10 Adjusted Surface. Make it the Active Layer.
Hide the layer 09 Reference Surface
Turn History On - Note that History will turn itself off after the surface has been created.
Select all the curves and run Surface > Curve Network to create the surface.
In the dialog box that pops up select the following options:
Edge Curves = 1
Interior Curves = 1
Preview = On
Select OK to accept the surface. This surface has a history connection with the curves that created it.
Turn Off Layer 10 Adjusted Surface.
Select the Extended Rebuilt Curves and the Y Axis Curves and Turn Control Points On.
Select the three centre points of each of the curves around the pommel area. Do this in Top View and to verify that you have selected the correct points you can temporarily Hide each curve layer.
The selected points should look like this:
Turn On Layer 10 Adjusted Surface.
Working in either Front or Right Viewport turn the Project constraint On.
Run the Set Points Command Transform > Set Points.
Check ‘Set Z’ only and use the ‘Align to World’ Option and snap the control points so that they are in line with the end of the curves thus raising the surface.
Note that the surface has now been modified smoothly around the pommel and it can continue to be adjusted by using the control points of the curves that created it.
Compare with the reference surface to check that the leg troughs have not been altered.
With the adjustment of the surface complete, the back can be closed off.
Turn Control Points Off and Hide the Layers 04 Extended Rebuilt and 08 Y Axis Curves.
Turn the Project constraint Off.
Draw a straight line between the two back corners of the seat surface.
Run Surface > Planar Curves and build a planar surface using the straight line and the back surface edge.
Join this surface to the seat surface. Note that any remaining history will now be lost.
Create a New Layer and name it 11 Trimmed Model. Make it the Active Layer.
Unlock the Layer 00 Foam Volume.
Copy the foam block and extended surface onto this new layer.
Hide the layers 00 Foam Volume and 10 Extended Surface.
Trim the foam block with the surface and then the surface with the remainder of the foam block.
Sometimes it is better to first create an intersection curve and trim both the block and the surface with the curve.
Join the foam block and trimmed surface together - the command line should report that the result is a Closed Polysurface.
© Simply Rhino Limited 2010 - 2016
Rhino3d Video Tutorials Transcripts - To further support you as you learn and progress with Rhino we've transcribed each of our video tutorials.
Hi I’m Phil from Simply Rhino and in this video I’m going to continue with our engine cover model.
First I’m going to take a look at creating this scoop on the main engine surface.
There are two main considerations here. First we need to maintain a seamless continuity between the main surface and the lead in to the duct. Second, we need to control these edge blends so that they transition from a relatively straightforward edge blend condition and then run into the main engine cover surface. These are both fairly common problems in 3D modelling, even though the specific context might be different. Once we have the scoop completed, we’ll then take a look at mirroring the two halves of the engine cover and then creating the final centre blend. Finally, we’ll make sure that everything is ready to export into SOLIDWORKS.
This is the result we obtained from the modelling we covered in the last video exercise. Now to start to add the duct detail into this top surface, first of all I’m going to extract the top surface, the main top surface here and I’m going to hide the rest of the geometry. I then need to add some curves to describe the sharp boundary of the duct area. These can be either curves that you draw and then project onto the surface, or in this case they are actually created from isocurves. The advantage in this case of using isocurves to create the boundary shape of the duct, is that the shape that’s created will be very close to the overall shape of the surface.
So to create these curves, I went to curve, curve from object and extract isocurve and toggle to produce curves in the other direction. Then once I have these curves, I can trim them up with each other and join them. We need to make sure by just checking in object properties that the curves when they’re joined, produce a closed curve. Once we know that’s the case, we can then go ahead and split our main surface with this closed curve and I’m just going to hide the curve now.
Now of course, because these two surfaces are essentially derived from the same parent surface, then if we go to our analyse and surface and Zebra tools, then of course, the area that we’ve trimmed out will be completely curvature continuous with the surrounding area. And the idea here is that we’re going to leave the front edge of this surface as it is and adjust the rear edge of the surface to create the scoop and the fact that we don’t change the front of the surface means of course, that it’s going to be curvature continuous with our existing surface.
Now the first thing that we need to consider here is that if we look at the control points for our small trimmed surface, they actually describe the overall larger surface, because when you trim a surface you don’t actually change the control point structure. What we can do to make our control points here more manageable and relate more closely to our trimmed surface, is to select the surface and run a command from surface, surface edit tools, called shrink trimmed surface. What this command does is that it re-approximates a new surface that has the same shape in UNV. It has the same number of control points and the same degree in UNV. The idea being that now our control points are relative to our smaller surface, but when we analyse the curvature of this with our Zebra tools, we’ll see that our surface hasn’t changed in terms of its shape and so our two surfaces are still completely curvature continuous.
If we take a look at the control points on our newly shrunk surface, what we know is that these first three rows of control points need to remain exactly where they are in order for us to maintain the continuity at the leading edge of the scoop or duct. So it’s only these three rows of control points that we can move.
Now before we go on to really look in more detail at the shape of the scoop, let’s have a look at the principle behind this. So the principle is that we don’t move the first three rows of points, we leave those exactly where they are and the subsequent points we can move. Now a good way of moving these is to use the move UVN tool which is situated in the transform menu. What the move UVN tool does is move control points, either in the U direction, the V direction or the N direction (the normal direction) relative to a surface. Now normal means locally perpendicular. So each one of these control points will move locally perpendicular to the surface, so this is an ideal tool for either adding a recess or a protrusion to an existing surface. The value that we put into this box here, in to the scale box, is a real world value and if I move a slider from the mid-point to either end of the slider here, I’ll move the control points by 12mm in a normal direction to the surface.
So if I grab these three rows of points here, and move them, you can see that I can start to create a dip in the surface. And very quickly, just to show the principle, I’m now going to grab the last two rows of points and move those downwards and then finally the last row of points and move those downwards again. So you can see now I’ve started to create a surface which of course now changes from our original surface and starts to create the scoop.
Now because we haven’t moved these front three rows of control points, when we look at this tool with Zebra’s, you will see that we have perfect curvature continuity between our existing surface and our new surface at this leading edge and so the principle really is that we don’t need to worry about matching this surface to our existing surface because we know that our control points are already doing this for us.
Now the one downside here of this method, is that we don’t have very many control points to play with here and there’s a large area of this surface that we can’t move. So what we could really do with introducing here are more rows of control points so that we can put a little more shape in to our scoop and that we can have the scoop starting to change direction, a little closer to this leading edge.
So if I just undo my move UVN operations, and get back to where we were a few steps ago, I’m going to look now at introducing some control points into this surface. Now the way that we do this is important. If we just add control points into the surface we are going to change the shape of the surface which of course is then going to change our matching. Likewise, if we rebuild the surface. So what we really need to do here, is to insert knots into the surface which will add control points by implication. So I’m going to go to my edit tools, control points and insert knot. I’m going to pick this surface here and I’m going to add a knot line just behind this first three rows of points and then I’m going to add a knot line just behind the what is now the fourth row of control points. And this is the result that I’ll get now. I’ll get my three control points which are still controlling the matching, have now effectively moved much further forward. So now I can move all of these points and get more shape going on in the scoop.
So again, I’m going to do something similar here. I’m going to go to transform, move UVN, maybe adjust this distance slightly and I’m going to move all of these control points down. The nice thing about this tool is that being slider based, is that you know you can push this back and get back to your original position and I can then change the value here and it’s a nice tool to actually interact with.
So I’m gradually going to push this shape down. Now what I’ve decided to do here is I’m actually going to move these last three rows of control points together. So I introduce a kind of S-Shape into my duct surface here. And the idea about using move UVN here is that the surface that I’ve now created still has a lot in common with the top surface, so that this edge is going to be fairly common to this edge and the fall off as it goes from left to right here, is going to be fairly consistent with this surface. So again, let’s take a look at the continuity here. So that front edge looks okay. Let’s also take a look at this with an environment map. Okay, and that looks quite nice. So you can see that we’ve got a seamless transition here between our two surfaces.
So the Zebra stripes that we use here, really only give us an indication of whether we have curvature continuity across our two surfaces. So to check with more accuracy, we can either go to analyse, curve and turn on the curvature graph for both of these surfaces, or sometimes where we’ve got a simple proposition like just one single edge here, we can go to curve and curve from objects and we can extract an isocurve from both surfaces and make sure these isocurves are coincident. Then we can go to analyse and curve, and geometric continuity and we can test the continuity across those two curved ends. Here we will see that Rhino is reporting these are G2 which are curvature continuous and we could check this in a number of positions along this edge. So once again here, technically what’s doing this for us is that these five control points here, the middle one of which is a coincident point are required for a G2 curvature match. If we wanted to match beyond G2, then it’s the next pair of points that would be required to be involved in the match for a G3 match. But in this sort of situation here, G2 curvature and continuity is fine.
Let’s now take a look at making the side surfaces for the duct. First of all I’m just going to check my distance between this point and this point, which is 36mm and I’m going to have some fairly simple linear sides to the duct but I want to put a ten degree taper on all three sides of the duct.
So how I’m going to look at doing this, is first of all to draw a line. This line I’m going to make slightly longer than my depth of the duct. So let’s make this 50mm long and I’m going to put a point on the top of the line and then I’m going to rotate this about the minus ten degrees that we need here. This is drawn in top view and we’re working in our world top C-Plane which is also shared here by our perspective view. The idea now is I’m going to orientate this curve on to these three edges, doing this in three separate operations and to do this, I’m going to use transform, orient, perpendicular to curve.
So this is the object I want to orient and this is the base point and let’s start first of all with this rear edge and you can see I can orient this on to this edge and make sure the copy option in my command is turned on and I’ll snap this to one end of the rear edge. Okay, and then I’ll repeat this process for the sides. So transform, orient, perpendicular to curve, object to orient, base point and edge. It helps if you can look down the curve when you use this command, just to make sure that your ten degree angle in using this example is facing the correct way. So once more, transform, orient, perpendicular to curve, object to orient, enter, base point and then pick the orientation curve.
Here I need to flip the orientation in X to make this taper inwards and then snap to the end of the edge. Once we’ve achieved the orientation, we can then use a one rail sweep to create the various surfaces. So surface, sweep one rail, do the sides first. This is the rail and this is the cross-section curve. When I use this command, I can use the simple sweep option here to give me some simplified geometry and a cleaner result. Repeat the process over here, rail, cross-section, simple sweep and then do this for the rear.
So next we need to trim these surfaces with each other. So I’ll just remove these curves and to trim these surfaces, I’m going to generate an intersection curve between the surfaces in question to start off with. You’ll see that the surface edges don’t actually run into each other exactly here. So the process that I’m going to use here is something like this. I’m going to go to curve, curve from objects and intersection, generate an intersection between this side and our scoop surface. That gives me a nice intersection curve and then just to check that this curve is running all the way along both surfaces, I’m going to go to curve and extend curve, curve on surface. And then I can trim these portions here.
I can do the same on the other side, so curve, curve from object, intersection and then curve, extend curve, curve on surface and then I can trim and then finally this one here as well. Okay and then we should be able to trim these surfaces now with each other.
So I can select my curves, delete the curves, select everything and join everything together and then I want to check to see if all these edges are closed. So select the polysurface, go to my analysis tools, run show edges. Make sure that only naked edges are selected in here, use a nice bright colour and you can see that all of this is closed up nicely in here.
So we’ve now got our kind of sharp version of our scoop or duct. So now I can start to add in the blends. So first of all I’m going to put in two larger corner blends. So I’m going to go to solid, fillet edge and blend edge. I’m going to use a radius here of 50mm and I’m going to pick this edge and this edge and then enter to get into the preview and the rail type that I’m going to use here is distance between rails. So this takes a perpendicular measurement across these two edges here, and then I’m going to enter and let that build. Again, you can check this page with the environment map just to make sure everything looks okay and then now for the more difficult part.
So as previous, I’m going to use the solid blend edge tool for this top rail and this lower rail here and as in the previous video, the blend will build correctly up to the point where the edges of the blend run into each other and then they will cross into each other. That’s fine because it means that the rear part of this blend will build correctly and the difficult part we have for manual control over.
So let’s build our starting blends. So solid, fillet edge, blend edge. I’m going to choose a radius of 10mm here and I’m going to pick the entire top edge, and then the entire lower edge. Enter, and look at the preview we get here. You can see that the outer edges of these rails possibly are not too wrong here, but Rhino isn’t really able to trim up the interim part here, and this edge over on this side, again the outer edges are almost probably where they need to, but again, nothing is built in the middle here. But this is a fairly easy solution now for us to remedy manually, so I’ll Enter to build the blend, and what we have here is a series of surfaces that have not trimmed my sharp sided poly surface at all. And what I need to do first of all, is to split these surfaces about the point at which they intersect each other. So, I’m going to go to Surface, surface edit tools, and split at iso curve and I’m going to just make sure I find the correct point here, turn shrink on here and split this piece of the blend and then split the upper part of the blend again with shrink on at the coincident position. Now I’m not going to throw away the front of the blend because I may need those later on, so I’ll just hide these. And, I’m going to do a similar thing over the other side here. So surface, surface edit tools, split at iso curve, just zoom in, just make sure we drop right on that intersection, make sure we enable the shrink option and split, and then same with this surface, and then hide these front parts. OK, so, if I just extract these portions for a moment, we can then hide this part and let's just trim away the side surfaces now with these blend parts. So I’ll run Trim, I’ll type in CRV so I can only pick edges, and I’ll start to trim away these sides. Ok, so each time I’m picking the curve option here, and just carefully trimming away these surfaces. And then the final piece here, ok and then we can join these together, and we can check that these parts are joined together correctly here, that looks good, and then just take a quick look at these surfaces, make sure everything looks alright there, that looks fine. Ok, so now let’s show back in the rest of the poly surface that we need here. And let’s look at what we can do with this front. What we need to avoid with these fillets as they come towards this front edge, is running them to a point. It’s perfectly valid to have a fillet condition here to run into a point, but in this case the geometry is not going to look quite right for us, but more important than that, because we have essentially one or two three-sided surfaces that are going to converge to a point, that’s going to give us a big problem when we offset to give us our B-surface when we shell this out. It’s going to give us a condition that will fail, so we want to create here a condition that not only looks good, but is also going to have enough integrity to be able to shell later on.
Now, if we look at, for example, the one side of what Rhino actually built for us, you’ll see that there is a fair amount of consistency about these rails here. And, if we continued this edge, the front edge of the scoop here, along here, you can see that we can get two four-sided surfaces here that start with the two blend shapes, and then, blend into the main engine cover surface at the front. So, we can actually use most of these edges that Rhino has built for us here. So let’s just look at this side first of all. I’m going to just explode this surface so this part and this part are separate, and I’m going to extract and iso curve here from this main surface, and snap that to this leading edge. Remember we created this leading edge with an iso curve, so this is helpful to us now. Now let’s bring in the other surfaces here as well, and you’ll see we have a similar sort of situation here where the consistency of these outer edges of these fillets is actually quite good. So, I should now be able to run trim and again type in the curve filter so I can only select curves and edges, and I’m going to trim the scoop surface with those edge curves, and then I’m going to duplicate these two edge curves. Curve, curve from objects, duplicate edge, there. Ok, and now I’m going to hide these surface parts here, I won’t delete them just yet in case I need to backtrack at any stage. This here can be trimmed up like so, and here this edge is a little short so I’ll extend curve, curve on surface here, and then I can trim these parts up with each other. Now I should be able to trim this outer blend edge from my main surface. So this is a combination of curves and surface edges here. So I’m going to run, trim, use the curve filter and then pick these edges, and these front curves, ok and trim these out.
Having trimmed our top surface with the outer boundary curves, we can now join everything together and then select any remaining curves and delete those, and check using our edge analysis, that all our edges around here are joined up correctly, and also as we go along it’s worth checking with an environment map and also Zebras as we go to make sure we’re not storing up any problems. So, just want to look at the continuity here, need to zoom into these areas to see Zebra’s, and we may need to adjust meshes as well.
When we created the corner of the main surface in the previous video, we introduced a blend curve from this point here to a midway point on this edge. And we created two four-sided surfaces and then matched to each other. We can do something similar here and before I do that I need to make sure that this short edge here and this short edge here is split about its midpoint. This should have been the case from when we trimmed, but just to make sure that indeed the case we can go to Surface, and edge tools and split edge, we can pick this edge here and split it about the midpoint, re run the tool and do the same over here. Ok, now if we were to the adjustable blend curve, we can either blend curves or edges, but we can’t blend one of each, so because it would make more sense to use a curve for this position, what I’m actually going to do is to generate an iso curve here and here that snap to the split edge, So I’m going to go to Curve, curve from objects, extract iso curve and snap to the end here and the end there. So now we can look at creating the adjustable curve blend. So let’s go first of all from this curve here to the lower edge of this top rail here. So let’s go to my adjustable curve blend tool, pick this curve, and then this edge and we’ll blend between there.
Now, because of the way that the geometry is working here, we’re going to see a big difference in this curve between blending from this edge and this edge. So let’s repeat the adjustable curve blend, and blend this front curve to this edge. And here we’ll see that we generate two quite different looking curves, and probably what we want is somewhere that’s between these two curves. And there is a method here that we can use that will give us that result. So, I’m going to delete these two curves and I’m going to create a curve that’s an average of this edge here and this edge here. So to do that I’m going to go to Curve and tween curves, and I’m going to choose this edge and this edge, move my cursor to the left edge and click, and you’ll see that I can generate a curve now which is an average of those two edges. And if I go back to my adjustable curve blend here, we might see that this gives us a slightly better looking result. Now, we need to look at this curve carefully and make sure, for example, that this curve isn’t dipping below this scoop surface here. So, let’s just take a look in right view here and confirm that and that looks ok for the moment.
Now, with this process there may be an element of iteration, so we may need to redo this process a few times, but for now I’ll use the similar process on the second side so, tween curves, and then adjustable curve blend, and take a look at this curve as well and we’ll accept that, and then we can start to build some surfaces. As previous we’re going to start by using a sweep to rail here. So, rail, rail, cross-section, cross-section and I’m going to match curvature at point A. And I’m going to take a look at the result that we get here. So that looks OK with the environment map and let’s have a look at this with the Zebra’s, what we can see if we go in close to this front edge here is we’re slightly out of continuity there, and let’s have a look at this point here, and again, slightly out here.
So, before we start to get too worried about that little lack of continuity, one thing that will help the Zebra’s is actually joining these two surfaces together. So let’s do that and see what this front edge looks like now, and you’ll see that looks a lot better. So, for the moment I’m going to leave the surfaces as they are, we should just actually check, I guess, that these are actually joined up properly, Ok that’s fine, and let’s build in our other surface here. And let’s again use a sweep to rail. So, edge and edge as the two rails, and then these two edges as the two cross-sections. Match, curvature across both, let’s join them, check first of all that they join up, that’s ok, and now let’s have a look at the analysis. Ok, so shape wise that doesn’t look too bad there, and let’s take a look at the Zebra’s, just look right across that front edge, that looks ok. So, let’s take a look at doing something similar on the other side. So, surface, sweep two rails, first rail, second rail, cross-section, cross-section, and let’s match for curvature on that top edge. Take a look at this shape that we get here, ok, that looks ok, and take a look at the front edge with the Zebra’s, just pull right into that edge, it’s not too bad, and let’s build another sweep in here. I want the edge here and not the curve, edge here, edge and edge, match for curvature, cross both of the rails, join, check the join and repeat the work with the environment map. Ok, so we should see that we should get a pretty good result here. Now, in some circumstances you may find that as we get towards the front here, that this edge becomes slightly problematic. So you might need to consider building a four-sided surface here, matching to three sides, splitting the edge of that surface and then, in other words, giving a little more room where this is one single surface, might help the shape of the front of the blend in some instances. So, once we have achieved this we can show in our other parts of the geometry, and join back together. Once we’ve joined the surfaces together, we just want to make sure that our result is a closed polysurface, and also we can just run the environment map again just to make sure that we’re happy with the various blends that we’ve created.
Now it’s time to look at adding any additional detail, and then to mirror the other half of the engine cover. So I’m going to extract this centre planar surface here, and delete that, and then just run the mirror tool. Here, the copy option, fairly obviously is on, and I’m just going to hit the y axis option here to mirror about a line of symmetry. I can then join the two halves together, and take a look at the result again with an environment map. Ok, so, for this centre blend here, I could use the solid blend edge tool, but there’s possibly an easier way to do this. I just want to check the distance across the perpendicular here, 19.999 so it looks as though I used a 20mm blend across this edge. And I’m going to go for something similar across the middle here. Actually, I’ll probably use something slightly sharper across the middle, and let’s look at a pretty easy way that we can do this. Going to take a line here, with the both sides option, start the line at Zero and pull the line along the centre of our engine cover here. And then I’m going to offset this curve with the both sides option on, by a distance of 8mm. so I’ll have a 16mm overall blend across the centre. And then I’m going to pick the two outside lines here and just trim away my engine cover. So, we’ve not got a gap here between the two halves of our engine cover and I’m going to use the blend surface tool to create a blend across these two front surfaces, and across the two top surfaces, and then this area here I’m going to blend separately, because I really need to control the shape of this blend correctly across here. And then, for the underside and the back, these are just fairly simple plainer surfaces.
Now, I want to make sure that I’ve got enough shape to the blend here. So, I don’t want this blend to be too flat. So, I’m going to run the blend surface tool here, first of all across this edge here, and I’m going to lock the sliders, I’m just going to push up the blend a bit here. Just make a note of this blend factor that I’m using here, and let’s just have a look at this with the environment map. I want to make sure that I can see a reasonably sharp blend across this centre, ok, that looks ok. And then I’m going to use a blend with the same factor on the front edge. So once again, lock the sliders, and use the same factor here, and then also take a look at this, ok, that looks ok. Now, you can see that this blend here is built slightly short and as a result, what I’m going to do is extend all of the edges of the blends just slightly here and then I’m going to trim them off. So let’s extend this by a couple of Millimetres, and this one, this one and this back edge here. First of all let’s do the easy parts here, let’s just trim with a line this piece of blend here, and do the same at the front. OK, and then to get the shape of this little nose piece right here, I’m just going to use a technique that we’ve used before in this exercise, I’m going to use the adjustable curve blend, blend across here, with the curvature continue as blend, and do the same here, and then I’m going to take this and pull it back on to the blend surface and then trim away the blend with it, and then likewise do the same here.
Ok, if we blended this all in one go, so we blended this, this and this part here, then we wouldn’t get this curved shape going across here, I think this is going to be important to explain the shape here when we see the final result. So now we need to look at just creating the surface that’s going to fill in this little nose area here. Now, I could use a two rail sweep here, but the issue will be in actually getting this edge and this edge, which are the cross-section edges, to actually join into a watertight condition. So I’m going to look at using a network surface here.
Now, the disadvantage of the network surface, is that it’s going to be slightly complex. But because my geometry is reasonably good here, and my boundary shapes all have a lot of similarity to each other then I think the result should be ok here. So, going to use curve network, pick the four boundary curves or edges, match for curvature on each edge and the important issue here is to make sure that this value here and the tolerance is the same as your absolute modelling tolerance that you’re working to. So, here we’re using the default small objects millimetres tolerance of .001 of a millimetre, and that’s the value I need to set in here. If I slacken this value off then my surface will be less complicated, but there’s less chances that my edges will actually join into a watertight condition.
So, I’ll preview that surface, I’ll ok this, I’ll join this together, and now we can take a look at what this looks like. Ok, so, that looks ok, and let’s take a look at this with the Zebra stripes. Let’s firstly take a look, make sure that this is joined up into a watertight condition. Yep, that’s ok, and let’s take a look at the Zebra’s across the edges here. OK, so that’s pretty good, ok I can’t see any disruption along these edges here. So, all we need to do now is to create our plainer surfaces to close off the volume. So, just draw a little line across here, and then use surface from planar curves, here join this together and I can just cap to finish off the lower part. Ok, so, we’ve got a closed polysurface now, so, just a few things now that we need to do to get the geometry ready for SOLIDWORKS. First thing is that, any co-planar surfaces like this are possible issues for solid work, so we can clean these up very easily. Pick the geometry, go to our solid tools, and right click on this tool here, and this merge all co-planar faces. And any coplanar faces here will be merged. Ok, so you can see that this face now is once face, as is this one here. The other tool that I need to run here is, Shrink trimmed surface. And I can run this on a poly surface here and it will shrink the individual surfaces. So, surface, surface edit tools, and shrink trimmed surface. Ok, and here it’s going to shrink 75 surfaces and 44 are already shrunk. The solid modellers will work much better with trimmed surfaces, the disadvantage of shrinking is that if you need to un-trim and re-work any of the blends, then you may have reduced the size of your base untrimmed surface by too much to do this. So you would generally do this on a copy of the geometry.
So, one last thing that we should do now is to pick our geometry and run check, just to make sure our poly surface doesn’t contain any bad surfaces, and, that’s all fine, and so we’re now pretty much ready to move downstream in this case to SOLIDWORKS. So I hope you found this video useful and thanks for watching.
Hi, I'm Phil from Simply Rhino and in this video series, we're going to take a look at creating this engine cover with the aim of eventually creating a solid model that we can export from Rhino into SolidWorks to add solid parametric features. The initial aim is to create some reasonably good surfaces that define the overall form and individual transitions with a good deal of clarity. In this first video, we're going to quickly examine laying out 2D and 3D curves to create the basic slabs that we see here, before moving on to look at creating the main blends and in particular this slightly difficult front corner transition.
Going back to the overall geometry, the shape around this front corner is less than ideal for a smooth result, and we've done this purposely so that we can look at the methodology we can use in these real-world situations where we need to build smooth transitions on geometry that doesn’t really fulfill an ideal. Let’s take a look at how we can lay out some curves in Rhino that are going to define the boundaries of the main slab surfaces that we're going to work with. Whether we merely draw these curves in Rhino or we trace them over an imported bitmap reference, the starting point would generally be some 2D curves that we can then arrange in the appropriate 3D space. Therefore, we can fairly easily create the three-dimensional boundary of the engine cover. Here, we're only creating half of the engine cover because of the fact, fairly obviously, that it's symmetrical about its centre. And this idea of laying out the curves and techniques that we can employ to do this is something that we cover in quite a lot of detail in our intermediate advanced class.
So once we've got our 2D curves in position, we can gradually start to add some of the interior curves. And again, it would make sense for these to be created initially as two-dimensional curves. Eventually, these curves would need to be optimized, or rebuilt. And here, all of the curves are going to be built as degree five curves with six control points. And as well as giving these curves the same number of control points, we're also going to employ another technique here.
If we were to look at this curve here and this curve, these two curves represent essentially the same curve, which is this curve at the highest point of the engine cover here. And we're seeing it's either in its right view or in its front view depending on which curve we're looking at. Now, if we turn on the control points here, you'll see that these control points are aligned along the world Y axis. And this is something we can do to make it easier to build good three-dimensional curves.
Generally speaking, to build these curves, one technique might be to build one of these curves and then rather than to draw it again in the right elevation position, merely it's a case of copying the curve, orienting it about the correct construction plane, and then adjusting the control points -- in this case vertically -- to arrive at the correct shape. And when we do this, we can use tools like the curvature graph to assist us with the shape.
So, gradually, we can lay out some two-dimensional curves. And we can also gradually start to take some of these curves and push them onto another layer, where we're going to create 3D curves. So of course we know our boundary curves here that are existing, are going to be part of our final 3D layout. So we can copy these curves onto this new layer.
And let’s copy the high point curve, which is this curve and this curve, that we've got expressed as two separate two-dimensional curves. Let’s copy this over to our layer two here. Let’s just change the object color of these boundary curves. And let’s look at a technique that we can use to quickly generate the three-dimensional curve from these pair of two-dimensional curves. And this is to use the command called curve from two views.
Okay, so in this case, it’ll generate the three-dimensional curve for us. Okay, now this command works by essentially projecting the two curves until they hit each other in 3D space. And as a result, in the curve that we generate here is much more complex than the original curve and it isn't something that we can use directly in this case. But we can use it as a guide and to check that the position of our clean curve is correct.
So the way that we can produce a clean version of this curve is to first of all run a command called ‘extract points’ from this curve. And I'm going to lock these points. And then I'm going to delete that two-dimensional curve. I'm then going to turn on the control points for this curve, and working in my right view, I'm going to turn on the project constraint and then take each one of these points and move them up to snap them on to my reference points.
Okay. And we just have the curve from two views here, just to act as a guide. So once we're happy with the shape of that, we can delete the curve from two views. We can unlock our points and delete those as well. And we can gradually start to create some curves. So we can use curve from two views as a quick way of creating the 3D curve and then we can maybe go back through some 2-D iterations until we get the desired result. And eventually, we'll arrive at a set of curves that describes our layout.
So each one of the curved curves here, so these curves here, each one of these is a degree five curve with six control points. And to build the surfaces from these, we can predominately I think use sweeps. So we can use Sweep 2 Rails. And it's important when we use this command that we use this simple sweep option here. This disables any matching in the command, but what it does is ensure that where geometry permits, we'll have the same degree and control-point layout as the contributing curves.
And so we can just keep running our sweep 2 rail command here with the simple option until we've built our slab surfaces. Okay, so that looks okay. And then we can just create some planar surfaces to fill in the rest of this. It's very often a little easier if we can make things into a solid at this stage. So join this together. And then cap to close off the bottom.
Okay, now we've got a closed polysurface. And we can then move onto looking at the blended conditions, particularly around this front end. Now, it's a good idea at this stage to maybe start to sketch out what we want the affiliate or the blend conditions to look like here. And we can do this quite crudely, but just working something out on a bit of paper very often is a much quicker way than trying to work something out completely on screen.
So here’s a quick sketch that’s just been created by tracing over a print of the sharp solid that we've created previously. And you can see that the idea here is that we want to maintain a constant size of blend around this portion here and here and then have a larger blend radius at this corner. And then we want to create a transition here. Now of course how we actually create this transition is the difficult point, but what we can see here is the clear intent about what we want to do along the main portion of these blends.
This kind of mimics the way we're going to work in Rhino. We're going to build the known parts of the blends first of all and then finish up with the transition. Once we have the slab surfaces and the sketch reference, we can start to add the fillet edge. Here, I have a closed polysurface, or solid, and I'm going to use the solid blend edge tool. Because of the fact that we have four edges that all converge to a point and that one of these edges will have a substantially larger blend than the others, then the geometry isn't ideal. This will result in a solid blend-edge tool failing to complete, but we can use the surfaces that the tool does create as a starting point to build the troublesome corner manually.
So let’s start by going to solid, filet edge, and blend edge. I'm going to choose a 20mm radius for these three edges and then change the radius to 80mm for this edge. Then I'm going to enter to get into the preview mode. Now, the rail type that I'm using here is distance between rails. This ensures that if you take a perpendicular line from the setback of the blend to the other side of the blend, this distance will be 20mm. This keeps the blend looking the same as the angular condition changes from face to face.
You’ll see here that in the preview that the corner doesn’t build, and really we wouldn't expect it to build given this type of geometry. So what we're going to do is accept the result that we get and then to work into this manually. So let’s take a look at what parts of the surfaces created by the solid blend edge tool we can keep. Most of the smaller blends are in fact okay. And if we look at these two surfaces here, they're fine up to the point at which their edges converge here. These two blends are in fact poly-surfaces, and you can see that the blend has run into trouble where we've encountered the corner.
And if we explode these poly-surfaces, we'll see that we can just remove this front portion here. And I'm really interested, as I say, in these two blends, up to the point at which the edges converge. Now the isocurve direction here should be perpendicular to the blend edges. And therefore, we can use the isocurve direction to split with. So I'm going to go to my surface menu and use split isocurve and pick the first of these surfaces and split with an isocurve. I'm going to enable the shrink option and snap right onto the intersection here.
When the shrink option is enabled, both of the surfaces that I obtain by splitting, will effectively be untrimmed. And this idea of having an untrimmed edge here is very important in case we need to run a match surf command to it. So, I'll now repeat the split isocurve command on the second of the blends here and I'm going to need to reinstate the shrink option here. This option isn't sticky, so be wary of this. And split again.
Next up, I want to take a look at this larger blend. We won't really be able to use anything of the blend that’s been created with the solid blend edge tool, but we can at least use this to generate a reference for the start of the setback for the blend. Before I do that, I'm going to take my closed poly-surface and explode this and I'm going to hide the portions of this that are not immediately relevant, so these three planar surfaces.
And I'm going to look first of all at this front surface here. All the surfaces that we created for our main slabs are untrimmed surfaces, and so the isocurve direction describes really well this shape of the surface. So what that means is I can use that isocurve direction to create the setbacks with. So I'm going to go to surface, surface edit tools, and split at isocurve. Pick the front surface and enter. I'm going to enable the shrink option and I'm going to snap this isoline back onto the start of the blend and I can then remove the small piece of surface. Now, the blend edge on the side surface wants to run up to this point here on this smaller blend.
So, this blend here now is redundant and I can again use split at isocurve. Make sure that I enable the shrink option and snap to this edge and then delete. And then finally, I can go again to surface edit tools and split at isocurve and I can split this surface again where it intersects the front edge. Okay, now I can start to trim out some of the surfaces with the edges of the blend. So to do this, I'll use the ‘trim’ tool. And at the ‘select cutting objects’ prompt, I'm going to type in ‘crv’ and enter. This limits my selection now to either surface edges or curves and it means I can now pick the edge of this blend surface, enter, and then trim away the top part of the surface here.
And I can do something similar with the side and this intermediate surface here. So trim, ‘crv,’ enter, pick these two edges, enter and trim. And then trim, ‘crv’, and trim away this part of the curve. So next thing we need to do is to take a look at this large blend here. And we could either create some curves here and do a sweep, but it's probably easier to use blend surface here. So I'm going to go to surface, blend surface, I'm going to pick the one edge, enter, and then the next edge, and create a curvature continuous blend. Now I'm going to look at this in top view just to look at the shape of the blend, and I'm going to lock the sliders and I'm just going to push out the blend shape of it so it's just not quite as flat.
What I want to make sure doesn’t happen is I don't want to get a - a sort of an S-shape here, so I'll pull this slider back slightly. But I just want it to be slightly bigger than one here. Okay, so I may need to look at this with an environment map. Just to take a look at this shape to see how it looks. Okay, so assuming that we are happy with the shape of the blend, then one thing we may need to do here is just to make sure that the bottom of this blend is actually pulling down to the same depth as these two flat edges of these surfaces here. So to check this, what we can do is just extend this surface here by a small amount and then trim it back with a - with a curve. Or a line.
Okay. We do this because eventually, our aim here is to make a solid from this and get the solid into SolidWorks. Okay, so that’s our corner blend. Now, we may need to come back and relook at this corner blend, but for now, we'll accept the corner blend condition here. And the next job now is to create this inside corner here.
And we can do this with the adjustable curve blend tool, by blending this edge and this edge together. We want to set the continuity here to curvature at both ends and again, we just want to take a look at this in the top view here just to take a look at the shape of the curve and make sure it sort of is empathetic with the shape of the top of the blend. That looks okay to me. Okay, so the curve that’s created by adjustable curve blend will be more or less touching this top surface here. But it may be slightly off the surface. So what I'm going to do is go to curve, curve from objects, and pull back, and I'm going to use the pullback command to pull this curve onto this surface here.
What the pullback command does is use the surface normal direction to suck the curve onto the surface. And with this command, I generally have ‘delete input’ option saying yes, because it's very hard to see sometimes which is the pulled curve and which is the original curve. Because in this case, the movement of the curve is going to be very, very small. Okay, so that curve is now on the surface. Now, the downside of this is that we previously had a G2 continuity, against this edge and this edge. And we will probably now only be tangent. So I'm going to check this by going to analyze, curve, geometric continuity and picking the two edges. And indeed, both of these now are only tangent.
However, in this case, that tangent continuity is good enough for me to build this corner reasonably well. So now we have that curve, I'm going to trim again and use the CRV input. Pick this surface edge that curve, and that surface edge and enter and trim away. And I'm going to select the curve and delete it. So I can now join this geometry together and take a look at this with the environment map tool. Okay, and just want to have a look at the just sort of overall shape that we've got here. And we can take a look at this with the zebra stripes as well. Just to check that our blends have built correctly. Okay, so everything looks okay from that point of view.
So, we've got a reasonably good starting point now. And our last job is to build this troublesome corner. Now, this may take some iteration. And one of the first things that we'll notice here is that we actually have five edges here. Now of course NURBS is a four-sided typology and we really need to work with four-sided surfaces here. But there is a fairly sort of tried and tested method that we can use here. And there are a few variations on this theme, but let’s look at just one of them.
So what we need to do is to create a curve that is going to establish a four-sided surface for us. Now, perhaps the main driving factor in this shape is this outside edge here and how it comes ‘round and wraps into this front blend here. So I'm going to go my ‘adjustable curve blend’ tool and I'm going to blend this surface edge with this surface edge here, to create a rail that we can use to sweep along.
And again, I'm going to go into my top view here, and if necessary, just pull this out slightly here. Now, we're not actually going to use this edge of the surface in the final analysis, but we still need to get the shape of this correct. Okay, so we want to make this look fairly consistent when we look at it from top view. So I want to make this width from the curve to this edge look fairly consistent. So we'll accept that result. And then we're going to build a sweep. So we're going to use sweep 2 rails and we're going to sweep against this rail and this rail and use that and that as the cross-sections. And, at point B here, we're going to match for curvature continuity.
So, I'm now going to join the sweep into the rest of the shape and take a look at this with the environment map. And just take a look at that sort of shape that we're getting here. So the overall shape here, a bit difficult to see with that environment map. Let’s try this one. The overall look of this now looks pretty good. This corner, top rail of this corner looks fairly consistent as we come around here. This shape looks quite good, it has a - a good deal of commonality with its bottom shape, which is what we want.
Of course, as we explained earlier, this is far from ideal, this corner transition, and we've really done this to try and prove that we can build a reasonable corner out of a geometry condition that’s not ideal. So, the remaining part now is to fill in this area. Now, again, if we think about the shaper here going back to the sketch, this area of course is also blending ‘round to here.
Now what we really don't want to lose is the high point of this surface here. So what we could do is take this edge and bring it round to the middle of this edge across here to, again, give us a four-sided surface. So to do this, we're going to go to curve from objects and extract isocurve. And I'm going to take an isocurve, going to toggle the direction first of all. I'm going to take an isocurve out of this long filet here and snap to the midpoint. And then I'm going to go to my adjustable curve blend and run an adjustable curve blend around here.
And again, I'm going to take a look at this in top view. And we want to kind of make sure here that our adjustable curve blend comes just inside this sort of middle of our existing sweep that we just created. So we'll try this to start off with. Now, this curve of course is going to be at a - at a different height to our existing surface. So I'm going to extract this sweep here and I'm going to trim it with the curve. So I'm going to trim in top view here.
Okay, and delete the curve. Now just have a look at this just to make sure we haven’t got anything too bad here. And you’ll see what we're trying to do now is again to run a sweep along here and the surface change here from this small dip to a small crest in the other direction isn't too big, so we should be able to hold continuity to the two long edges quite well here. So I'm going to join the geometry together here. Just check my edges here just to make sure that this and this are closed up. And then I'm going to create a two-rail sweep using the long edges as rails, and make sure that I'm curvature matched across A and B. and let’s take a look at this now with the environment map, so let’s join it together, analyze, surface, and environment map. Let’s just remove this curve.
Okay, so that’s not too bad. So you can see that the shape looks pretty good there. In the environment map options, the fluorescent tube is a good type of environment to look at this type of surface, where we can see the shape of the blend that we've created really well with this. It's very good for showing up defects as well.
Okay, so we can build a reasonably proficient blend fairly simply here. And all that we need to do now is to show the surfaces that we've got hidden. I can remove that back and then I can trim away the corner here, join my geometry together and then cap the back of this. Okay, so we now have a closed poly-surface and we can now either move this downstream into SolidWorks or we can add more detail to this.
In the second video in this series, we'll take a look at creating this scoop detail in the engine cover. This includes another less than straightforward blend condition and also how we can maintain a seamless transition at the leading edge of the scoop, where the scoop meets the existing engine cover surface.
Thanks for watching and I hope this video has proved useful.