Why we build our surface models from scratch

Well, it’s time to dust off this blog and get it going again!  It’s been way too long.  Let’s just say life happens sometimes, but I’m glad to be back at this again.  So, to kick off the first post, I’m going to talk about something that we spend a lot of time talking to customers about when we build surfaces from laser scanned data.  We’re often asked, “Why do you use Rhino3D to do reverse engineering work, instead of one of the purpose made reverse engineering software platforms?” The short answer comes down to one word – quality. While these purpose made software platforms do quite a good job on mechanical types of assemblies, where they fall short is in the area of freeform surface modeling. Since the vast majority of the work we do falls under the category of freeform surface modeling (that is complex shapes that cannot easily be described with straight lines or arcs), to get a high quality model that we are proud to deliver to our customers, we build them “from scratch” using Rhino3D. Take for example this laser scanned aircraft nose:

NoseLaserScanned

This is acceptable laser scanned data, certainly good enough for the purposes of building a surface model off of. Another vendor took this data, and using purpose made reverse engineering software created this:

Nose_NotSSDC

What our customer needed was a watertight (gap free) exterior surface model of the macro shape, minus all the bits like antennas and other small details, so that CFD analysis could be run on it. What you see above is a collection of surfaces that are patched together to form this nose, except that they are not actually watertight. The alignment of the “isocurves” – those lines you see along the surface, does not make any sense in the context of the actual object. Areas like the transition from the nose to the windshield exhibit some really strange surface defects that give it a lumpy look, defects which are not in the laser scanned data. In short, this surface model was useless to the customer, because it could not be imported into his CFD software, did not accurately capture the essence of the shape, and was such a collection of small patch surfaces that making it watertight would have been extremely time consuming. Here is the surface model we created “from scratch” from the very same laser scanned data:

Nose_SSDC

The vast majority of the nose is one single surface, and the entire thing is watertight, so that it can be imported into the customer’s CFD package without any further work.  Using the laser scanned data and photos of the actual aircraft, we were able to give the windshield and windows the proper shape.  There are no strange lumpy defects, and the isocurves are aligned in a way that actually make sense.  In the overall scheme of things, an aircraft nose is really not a complicated shape.  It’s quite telling that a reverse engineering software package does such a poor job of making what is frankly not the most taxing of tasks. From what we’ve seen, this is the rule and not the exception when it comes to shapes like these with reverse engineering software.  Did it take longer to do this from scratch?  Probably, we can’t know for sure, since we didn’t make the first surface model.  But, we’re not exactly comparing apples to apples here, since the original surface model was not actually usable by the customer, so it’s like saying “we could do this faster if we deliver something that’s totally useless and causes you nothing but frustration.”  We take pride in our work and we like to make our customer’s lives easier, and so we take the time to do it right, and build our models from scratch.

Published with permission from School Street Design Company Blog. Source.