Calculations vs. 3D CAD – A historical curiosity

KTM_Bike_Complete_2c

**Guest post by Al Dean**

 

When you consider the level of intelligence, automation and frankly, jaw dropping amazingness of most of today’s 3D design and engineering tools, it’s perhaps a curious thing to note that most systems don’t provide much in the way of support for one of the most fundamental parts of the design and engineering process – mathematics and calculations.

 

We’re all familiar with the ability to create parameterised, intelligent models. The ability to take geometry dimensions and drive other dimensions by linking them together is one of the basics of every modern 3D design system. Parametric modelling is, I think it’s fair to say, a de facto standard in term of working practices.

 

Over the years, the sophistication of those dimensional links has increased. From basic mathematics (D1 = D2-D3) to more complex equations, look-up tables, the ability to have state based or logic operators. But essentially, we’ve got a set of tools that allow you to link one set of dimensions to another.

 

The curiosity is that this doesn’t really reflect the very earliest stages of engineering development. That point where ideas are worked out, where the real thinking is done and yes, there’s a whole lot of calculations. It’s looking up empirical methods, it’s using established theory and methods to work out if what you want to achieve can actually be achieved. And as I’m sure most of you will know, most of this doesn’t really lend itself to physical dimensional constraints as you would place on a digital model. What we’re talking about here is documenting the design process and the intent behind your product decisions.

 

So ask yourself this. Why is this the case? I’ve been giving it some thought for a while and I’m stumped if I can come up with a good reason. Perhaps it’s because the CAD industry revolves around the concept of geometry creation and that most vendors don’t consider those tricky and challenging parts before their own tools get involved. Is it perhaps that in many industries, the CAD work has, until recently, been something that’s done separately from the engineering conceptualisation and calculation phase?

 

If you look at the range of mathematic or calculation applications out there (There’s PTC Mathcad, of course, as well as MatLab, Maple and of course, everyone’s favourite, Excel), there’s very little in the way of linking these to the 2D or 3D geometry design systems. Yes, Excel integration is commonplace and many 3D design tools include some form of basic formula + parametric capability, but it often does not go beyond that.

 

PTC has, of course, got a head start on this since its acquisition of Mathsoft, just under 10 years ago. The latest release brings some interesting tools that allow you to create those links between the calculation worksheets within PTC Mathcad and your geometry models in PTC Creo.

 

The mechanism to achieve this link is interesting as it’s not just pulling values from PTC Mathcad calculations to drive geometry parameters – it works the other way too. Outputs from PTC Creo models can be fed back into the worksheet and used as part of further calculations.
There’s also an interesting move in how that data is stored. A traditional workflow would mean that you have two data entities – the geometry set and the worksheet. PTC Creo now allows you to embed PTC Mathcad Prime 3.1 worksheets directly into the geometry data.

That, combined with the fact that anyone else can use that worksheet using the free for life PTC Mathcad Express, makes it a powerful combination.

 

There’s a lot of talk of intelligent product models being the future goal. If you look at much of the messaging around a systems-based approach or even pushing into the Internet of Things, there’s much talk of integrating real world parameters and data into the design process – but these are over shadowing the potential to do something just as fundamentally key and perhaps more longstanding – the ability to have intelligent models that are driven, in many respects, by the fundamental calculations that drive their design and engineering.

 

This entry was posted in Education, Mathcad, The Inside Scoop, Uncategorized and tagged , , , , . Bookmark the permalink.

2 thoughts on “Calculations vs. 3D CAD – A historical curiosity”

  1. Fred Kohlhepp says:

    I agree with the basic premise of this blog–that calculation does not always drive geometry in design, at least not as soon and as rigorously as it might. I would note that Mathcad (before PTC) could embed both Visio and Smartsketch objects in a sheet and pass information both ways, to and from that object.

    Embedding Mathcad sheets in Creo models is a worthwhile step forward. It is, however, hampered by the limits placed on the calculation by the new version of Mathcad (Prime) and the inability of Prime to communicate effectively with other versions of Mathcad (even Prime 3.0). (One wonders if an “advanced” calculation, not available in Express, is used how the coupling will work when a full version of Prime is not available, and whether subsequent versions of Prime will be able to read the earlier link.)

    This process will also be impeded by the natural inertia of the existing design process. Much design is currently developed by aesthetics and geometric constraints, then “thrown over the wall” to the analysts to determine if it is acceptable.

  2. Raiko says:

    I’m still of two minds whether the integration of Mathcad is worthwile. Though I love the idea that I do not have to enter digits twice in order to create a geometry I have some caveats.

    #1 integration is not a virtue in itself unless it speeds up or facilitates a process. Timewise the savings are, in my eperience, marginal. The transfer of data is the real deal.

    #2 integration poses severe problems when it comes to understanding/interpreting a computation. The calculations are complex enough, but having to deal with data crossing from one system into another is another level of complexity. It needs to be VERY transparent in order to enable others to understand what was done and for what purpose. BTW, if Creo needs to be opened in order to check Mathcad calculations I’d refuse to use it. Opening a huge programm to check some equations is a no go in my book.

    #3 human bottle neck: Like it or not humans do need time to compute and speeding up data transfer does not enhance quality.

    #4 Compatibility: what is the value of an integrated worksheet that can’t be opened a revision later (be it Creo or Prime) and consequently not scriutinized by others? will there be a downward compatibility between Creo, Prime and PDM?

    #5 Documentation needs. Again, integration does not serve any purpose for another issue. Besides a CAD model engineering departments are obligtaed to document their work in a fashion that makes it traceable. It won’t be enough to store a Creo model with an integrated prime sheet.

    Raiko

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s