Ellipsis

Inventor 2011 Direct Manipulation

Do you roll your eyes when you hear “40% increase in productivity over the previous release”? I did. And I work for Autodesk. I know you guys think I drink all the Autodesk Kool-Aid and then go about repeating the talking points… It doesn’t hurt my feelings, well maybe a little… I’ll get over it.

I’m a proof guy. If you can only tell me and not prove it to me, believe me I’m the first to call BS (bovine slime). So when I heard that the new tools are going to increase productivity, my initial thought was… sure, but do I have learn a new interface, aka the Ribbon? Don’t get me wrong, once I got used to the Ribbon, it was sweet. But the initial dip in productivity hurt man, felt a little like a new application and not one that I had used since R1.

I can’t in good conscious (or in job security concerns) write down my initial reaction after seeing Dynamic Input in the sketch environment, then the Direct Manipulation tools. Flippin Sweet! Is about as close as I can go. These new tools are the best example of anything that I’ve seen that you can legitimately claim immediate increases in productivity. And it’s more than just increases in productivity, the overall user experience is dramatically different and remarkably easy.

About a month ago I put out a video talking about the basics of parametric modeling. You know, the video talking about “practice”. I did it for a couple of reasons one of which was so that I could compare the new tools to the old ones modeling the same part. So, take a look at these two videos, the first one is done with Inventor 2010, the second is Inventor 2011. Watch my mouse, pay attention to where I go for the commands and just the overall ease in which I can model basically the same part. Yes, I updated the part so that it showed 2011 off…but that’s my job, yo.

Inventor 2010 Demo

And here is the same part modeled in Inventor 2011.

So like I said. I'm a show me kind of guy. Comparing these two videos should give you a good reason to upgrade, clearly easier to use and you will save time. Time for longer smoke breaks, coffee breaks, Facebooking... err I mean innovating. 

-Rob

#dp11

Visualization Question

So my last post was a bit feisty... I'll do what Tedeschi has always recommended to me and take both halves of the pill today [grin]. Yeah, right. 

I want talk a little bit about visualization. My tip of the day on Facebook was about adding knurling to the surface of a part as a new color. Pretty cool tip but that got me thinking about how important visualization is to design and even if that impacts your design decisions. 

The Rock Star himself Kevin Richards would slap me upside the head for even asking. In his area of expertise, it is all about the ability to see immediately the impact of light on the body panel of a car for example. But I'm wondering about something a little less obvious, say a chainsaw or maybe a conveyor system. Those items don't necessarily need to look "pretty", no they need to look rugged, durable. Can that be done by incorporating material selection and color scheme into your design? And how important is it to put that into your models?

I'm full of questions today I suppose. I would guess that not every company has the benefit of an industrial designer on staff and many mechanical designers are focused on function, weight, strength, etc... And if we can paint it periwinkle blue and sell more of them cool, but we'll generally stick to penitentiary grey like we always have.

So that's my question to you guys. How much is put into the aesthetic of your designs? Is it even considered when your design isn't a "consumer product"? My inquiring mind wants to know. Oh, and bonus points to the person that identifies the periwinkle blue movie reference.

-Rob

I feel like Braveheart

Remember that crazy look in Braveheart's eye right before he told his friend "I'm gonna go pick a fight"? That's me right about now.

Welcome to my rant. For those of you that know me already, this rant will not come as any surprise to much of anyone, those that are being introduced as part of this blog entry, welcome. I just read an article on how Dassault positions CATIA against AutoCAD for mechanical design. Are you serious? The problem is that this strategy makes sense to customers where AutoCAD is how they think of Autodesk. I’m not giving out any secrets here folks, but I find it interesting that they choose to target AutoCAD customers versus going head to head with our Digital Prototyping solution.

Here is the specific quote that gets me “For years, the default auto design program has been Autodesk’s AutoCAD; the CATIA 3D software represents serious competition. “AutoCAD is a very good tool that is great for architectural design, but with our auto clients we are addressing a much more advanced market from a technological point of view,” says Jonathan Dutton, Dassault’s automotive marketing development manager. 

Let me preface this as to not make my AutoCAD counterparts mad at me – AutoCAD is an awesome product, extremely flexible in its application, and tremendously easy to use. Now having said that… Who the hell is designing cars (start to finish) in AutoCAD? Parts of cars, facility layouts, tooling? Sure, but the whole flippin car? Common man. This entire article is hilarious.

Here’s another funny one: "Autodesk has strengths in exterior design, said David Cole, chairman of the Center for Automotive Research. But along with Siemens PLM software, “Dessault is the big-time player,” Cole said, “and that kind of software is one reason it takes two years to develop a new car today instead of five years.”

So let me get this straight. Most every automaker in the world uses Autodesk Alias and Autodesk Showcase during conceptual design, fact. The article then goes on to say that Autodesk Inventor is good for jigs and fixtures (acknowledging, but a quick attempt to dismiss), then of course AutoCAD is used to fill in gaps along the way and to do facilities layouts, and oh yeah it does some 3D too… but, but folks the bulk of the work is really CATIA….? How can you lay that claim with a straight face? Maybe the bulk of the engineering budget goes to Dassault, but to say that the bulk of the design work is done with CATIA – after the article concedes that Autodesk is present in most all of the manufacturers mentioned…?

Wow. Next time let’s at least make a reasonable comparison. Nice tactic I suppose, but seriously…comparing CATIA to AutoCAD is just weak.

Here’s the article if you are interested in reading it http://industry.bnet.com/auto/10004118/nascar-bmw-3d-design/ You know, now that I look at the image of the BMW inside of CATIA, I can understand why they compare CATIA to AutoCAD. I wouldn’t want to put that image up against Alias, Showcase, or Inventor either.

Out. 

Lights... Camera...

Hmmm… I have a sneaking suspicion that I have named an entry this before, but a quick search of my archives came up with nothing so I will stick with it. In any case, this one is a quick tip that should help out all of you who use Inventor Studio for creating product imagery.

The stock lighting styles that ship with Inventor assume a model of X cubic area or volume. You will more likely than not find that when applying a style to a model that is either significantly smaller or larger than the that default volume that the lights simply do not look “right”, either drowning out your model or highlighting a very small patch of it. Well, before you start moving the light origins there is a quick global change that you can make that may be just what you need.

As you can see from this image, the model is a bit larger than the effectively lighted area defined by the default settings in this lighting style. As I already pointed out, this can be corrected without touching the position of either of these lights by simply adjusting the global scale.

Upon opening the Lighting Styles dialog the last tab, named Position, has a Scale slider. Not surprisingly, by default it is set to 100%.

If this setting is changed, as it is here to 300%, then the position of all of the lighting elements in the scene is changed and the result is more in line with what I wanted.

Much better…

Well, that is it for now, hope this helps.

What's the deal...

What's the deal with digital prototyping?  Well, that is the question I was asked by Carl Alviani of Core 77 while at AU last week.  My response and a short demonstration of the workflow can be seen on thier site, buried in with the rest of the Autodesk University stuff.

Visibility...

While working on this headset assembly last week it occurred to me that a good tip for working with derived parts was staring me in the face. I have a lofted surface that is used to shape the electronics package housing in this assembly. Not wanting to go through the motions of creating a surface like this again for the outer headset housing, it is far easier to use derive to do an associative copy of the face and then offset it for the material thickness of that part.

So, the first thing I need to do is to mark the lofted surface as an exported object. Easy enough… I find the lofted surface in the browser and then from the right click menu I select “Export Object”. I save and am done with this file.

Now, I start a new part, exit Sketch 1 and then select Derive from the palette.

What’s going on? There are no surfaces to select, even though I clearly marked the lofted surface for export. Well, I actually missed a pretty important step back in the original part. I need to make the surface visible, as well as marking it as an exported object.

So… back to the original part, I find the lofted surface in the browser and from the right click menu I select “Visibility”. Save and close the part, start a new one again, only this time when I run the derive command I see that the Lofted Surface is now selectable from the derive menu.

What would seem to be a minor detail can actually cause you much angst if you are not mindful… so please… be careful out there!

Rule Fillet

Well, as promised I will take a look at yet another unique area of functionality available to users of the Plastic Features Technology Preview. Today I will introduce you to the Rule Fillet.

Rule Fillet… intriguing name isn’t it? For those of you who have backgrounds in surface modeling, this name may bring to mind something like this… S(t, u) = p(t) + ur(t), an algebraic description of a type of developable surface. However, in this case, Rule Fillet is exactly what it sounds like. Quite simply, it is a fillet which has as a characteristic a set of rules governing its behavior.

Take a look at the image below… I have applied a Rule Fillet to a Rib feature. In the dialog box you can see that I have selected the entire feature as the source and have selected all of the edges of this feature as targets for the fillet.

I do not however want the top or end faces of the rib feature to have filleted or rounded edges, so I must exclude them from the selection set. If I expand the dialog box to show additional settings, I have access to tools which will allow me to remove both individual faces and edges from the selection set.

The advantage of setting up a fillet rule such as this is that if there are any geometric changes to the feature, e.g., rib elements being added or removed from the feature definition, the fillet will automatically be added or removed as well, without any additional interaction on the part of the designer whatsoever.

A quick edit to the Rib feature, selecting an additional sketched line which was omitted previously updates the Rib network. Fortunately, because of the Rule Fillet, I do not have to restructure the part history, or edit the fillet to include additional edges as the new addition is included in the existing Rib feature and therefore is filleted automatically.

A catchphrase in the software industry for years has been the ability to “capture design intelligence”. It could be argued that features such as Rule Fillet finally achieve this.

Getting Thin...

Well, I’m back after a two week swing through Canada where I helped to deliver the Digital Prototyping message to audiences of Inventor users as well as users of our competing products. After presenting 6 times, in 6 different cites there was one dominant theme… the next generation of tools that Inventor users will have at their disposal are available right now on \\Labs. The new Autodesk Inventor Plastic Features Technology Preview is fundamentally different with respect to basic modeling tasks, different enough that it deserves some serious column time in the coming weeks.

Today I am going to look at what will most likely be the first departure from the typical modeling methodology that you will encounter when using PFTP… hey, I know, not the best acronym ever, but it is significantly easier to enter on the keyboard. Anyway, consider the simple extrude command…

Not quite as simple as it once was when I have additional things to consider with regard to wall thickness. For those of you who do a lot of plastic parts however, THIS is the ideal workflow as the part essentially understands that it is a thin walled part.

In creating this electronics housing I will first need to cap the open top and bottom you see in the image above. In selecting the new “Thin” tab I have access to controls for plastic feature specific characteristics. With this part, I want to close the top and bottom and open the ends.

If you look at the image above you can see that I first deselect the “Open Faces” From and To terminations. Next up I have to now select the faces, or more specifically an edge of the faces that I want to open.

In selecting the “From Profile” option I can now open up the faces that you see in the image above. Ahhh, sweet success… the “finished” start of my electronics housing package.

Next in this series I will look a bit more in depth at additional functionality of the PFTP. Hope you tune in to check it out.

As Clear As Mud...

I am not sure if anyone remembers or not, but one of my first posts last year was on the importance of setting the visibility environment “correctly”. For most of you I am sure wireframe display in Inventor gets used rarely if ever. There are times however when it is imperative, and quite frankly, the only way to achieve certain tasks.

Using Shrinkwrap, available now on \\Labs is a really good example of why it is so important. Take a look at the image below…

I need to select one of the faces buried in the wheel to use the “Close Face” function. Not an attractive proposition when faced with a display that looks like this. Good luck. However… make the following change to the Display Settings in Inventors Application Options…

And behold the results… wireframe that you can actually use!

You guys should listen to me more often… [grin]

Driven Dimensions...

Really busy working on AU dataset(s) this week but I did not want to neglect all of my faithful readers… [grin]. In any case, as I was using a reference dimension to drive another in an equation it occurred to me that this technique that I use so often may not be evident to all of you.

So, that being said lets jump right into it. If you have been using Inventor for any period of time longer than a week odds are you have seen this…

Any time you try to dimension part of a sketch which is already “described” via some combination of dimensions and/or constraints you will see this warning. This is not a bad thing so relax… what you end up with on your sketch will be a driven dimension, but not in the classical sense of the word. No, this is more of a reference dimension. In any case, what you end up with is a dimension whose value is enclosed in parenthesis. It will look very similar to what you see below…

Now, perhaps you are not aware that you can actually use this as a reference in an equation to drive another sketch dimension or feature size/position. In this case, I wanted to add a workplane with an angular offset equal to one half of the included angle of my part. Take a look at this last image…

Here I have selected one of the two workplanes on either end of my part, then a work axis which is located at the intersection of these. I am prompted for an angular offset, so I select the driven dimension added in the previous step and then divide by two… and in my case, multiply the whole thing by negative 1 to get this new plane to offset to the correct side.

Pretty cool huh?