Scanning a steam train with fricking laser beams. / Nov 22nd 2012

tl;dr We scanned a real steam train with fricking laser beams and what emerged was better than we could have expected.

This is part 2 of a series of blog posts. You can read part 1 here, where we talk a bit about why we’re doing this project to record a steam locomotive called Winifred and the historical context.


As part of a generation who grew up with Tron and the digitisation scene I’ve had a feeling of awe and almost overblown expectation of scanning real things with lasers to make digital objects. This was only heightened by recent “cultural objects” such as the Radiohead “House of Cards” video. It currently feels as if any 3D object is digital shape and mesh data waiting to exist. The speed, quality and price of white light and laser scanners appear to be adhering to something akin to Moore’s law if not an accelerated version. 

Still from Radiohead "House of Cards" by Aaron Koblin

Still from Radiohead “House of Cards” by Aaron Koblin.

Having started to think about creating model kits through 3D printing, one of the next considerations was how to get complex collections of geometries into the kit production pipeline. Laser scanning is a clear and obvious route for this, however the price and ease of acquisition of the data was an unknown. Given Winifred’s historical importance there was now a clear test case for this where the data had an intrinsic value to the railway preservation scene as a historical record and as artefact in its own right.

Researching laser scanning firms shows an industry emerging strongly, with the usual strata of quality and attention to detail at one end and buzzword chasing charlatans at the other. In my research one firm stuck out as very suitable for the project of scanning Winifred. That was a firm, Digital Surveys, who had a specialism in measuring and surveying “as built” installations such as ones in the petrochemical and offshore oil and gas sectors. These would have similarly complex challenges in terms of pipework and occluded areas. In addition they had the in house skill to interpret complicated point cloud data into complicated 3D models of engineering prototypes.


After almost 2 months of intermittent planning, Ben from Digital Surveys came to Bala with his Faro laser scanner and collection of white reference spheres.


He took a test scan and seeing Winifred in three dimensions emerge out of the workshop on the screen felt like the digitisation moment. It was the precise moment that Winifred became a digital object; a shape which could be “reproduced” if not in function, hitstory, totality, and at a lower resolution, but at will. 


Throughout the day and in sometimes cramped conditions Ben took a total of 15 scans.


A couple of weeks later we were sent this video of the point cloud. I can’t even begin to describe how it makes me feel. It’s so much better than we could have ever expected and expectations were already fairly high due to the cultural reference points described above. 

Winifred Point Cloud Animation from Digital Surveys on Vimeo.

Tomorrow we’ll talk about the next phase of the journey deeper into the geometry.

We’d like to thank Ben and the team at Digital Surveys for all their work and also Dan Crow for all the company and help he gave on the day of recording, helping to document what was done. Thanks as ever to Julian Birley, Winifred’s keeper and the ever helpful team at the Bala Lake Railway.

Powered by Tumblr