Dear DAVID Community,

In 2016, DAVID and this forum have been integrated into HP Inc.
This forum has been read-only since then, and will be switched off by the end of 2019.

To everyone who has contributed here, we thank you for all your work and for the constructive, helpful and friendly atmosphere you have created!
Please continue to discuss on the HP Support Forums.

Compounding Errors

Problems, solutions and discussion about scanning with a video projector and stripe patterns.

Compounding Errors

Postby twiggg » Fri Sep 23, 2016 6:34 am

So I've had my SLS2 for about 8 months and I've used it for a multitude of projects, all mechanical engineering related. I have been so impressed with it's capabilities. However I will say that I've never felt super confident in setting up the scanner, how the parameters affect quality, how to set up the parameters optimally for each object. So here are a few projects I've worked on. All of which using the largest, 240mm, calibration square.

My 1st project was a seat mold for a carbon fiber seat for a F1 style open-wheel racecar.
Carbon Seat Mold

Next was a carbon-fiber front-body for that same car, a real challenge because it was specular with large flat surfaces that were especially hard to align. I ended up using numbers taped on to align, which worked well enough at the time.
Front Body-1
Front Body-2

Next was a engine scan of my daily-driver to help me design cruise-control for my car, it worked spectacularly. I've never had a problem with accuracy in my scans once I got the scanner what felt like dialed in.

This brings me to my most recent project, and why I'm coming to the forum for help! So I'm working on a crazy project with an 86 Toyota MR2, a mid-engine'd "sports" car. My colleague and I want to cram a V6 into it and both being engineers and since I have this scanner I thought it would be the perfect project. So the engine bay is totally empty, a decent amount of room to move the scanner around, and the lighting in the garage is easy to control. So I want essentially a band that is the surface of the engine bay, imagine scanning the inside of a hole. We had a few problems with alignment so I went an got a bunch of 0.5" wood balls and glued them all over the engine bay in pairs, placed so that there is at least 30% overlap between scans for plenty of alignment. The current project is 40 scans, and for each one I have at least 4 spheres every scan for alignment, 2 from the previous scan, 2 new. After I take a scan I do contact pair alignment, then pair-wise alignment until there is no movement of the scan. Then I combine and continue. The scans look perfect to me....until I get back to where I started. By the time I get back to where I started in the bay there is a TON of error (42mm). I thought this was because of alignment but Im not convinced, because I haven't had this kind of accuracy problem before. So my thoughts move towards distortion in the scans. Because the engine bay is a rectangle and the camera scans from the center the working distance is changing between sides, but never such that the camera can't see the full projector pattern. Does anyone have any ideas as to where this error is coming from and way I can try and fix it? If people could just give a brief summary of their process or point me to some threads it would be greatly appreciated.
Posts: 3
Joined: Mon Sep 12, 2016 6:54 am

Re: Compounding Errors

Postby twiggg » Fri Sep 23, 2016 6:38 am

Here's more pictures of the stuff above:
Cruise Control

Engine scans with Cruise control system overlaid.

Overall engine bay dimensions
Overall engine bay

An example of how well aligned every scan is in the project(please tell me if this isnt good!! haha I might just think its good but its crap)
Scan alignment
Posts: 3
Joined: Mon Sep 12, 2016 6:54 am

Re: Compounding Errors

Postby twiggg » Fri Sep 23, 2016 6:39 am

Here's where the error compounded:
Posts: 3
Joined: Mon Sep 12, 2016 6:54 am

Return to Structured Light Scanning

Who is online

Users browsing this forum: No registered users and 1 guest