3D Scanning

[toc]

Does anyone have experience of using 3D scanners integrated to CAD to inspect sheetmetal components. We have a Faro arm CMM which I believe you can attach a laser scanner to. How fast is point cloud scanning - is it practical to use it to check first off pieces in a busy fabrication shop. I understand to get reasonable quality the digital scanners are very expensive.

 

[scotland6]

Try

http://www.3dscanners.com/

for some possible assistance.

 

[jdsmi]

We use a Faro CMM for inspecting vacuum formed plastic parts (comparing the part to the CAD model) and have been very successful. Inspections that used to take days are now done in a couple hours.

 

[ProEpro]

JDSMI: do you use a laser scanner atached to your Faro?

TOC: Do a carefull evalutaion of the scanner you look at. Some of them are better able to handle shinny parts like sheet metal than others. You will get the best results by painting the parts.

expect to spend ~100,000$ on the project. (If you want a model maker laser for your Faro you need to upgrade to a Faro with more axis.)

Faro sells an inspection package that you can experiment with. Depending on your parts it maybe enough.
ProEpro

http://www.whitelightdesign.com

 

[jdsmi]

No, we don't have a Lazer attachment but we're currently looking into it. We're also evaluating Romer.

 

[raycmm]

That's a good point to evaluate romer.
They have a laser scanner called g-scan.
They are in fact the only ones to offer their own laser scanner with their arms.

You can see it on

http://www.romer.fr

In north america they are sold by AXILA

http://www.axila.com

 

[jdsmi]

We just took a look at the Romer product using the Perceptron Lazer probe and Geomagic Inspection software.
Very impressive.

 

[raycmm]

How many axes on the arm you have seen ?

 

[jdsmi]

It had 7 vs. Faro's 6.

 

[ProEpro]

The biggest difference between a Faro and a scanner is that you can get hard points from a faro by touching things. with other scanenrs like an ATOS you only get non contact measurements. Contact methods are better for things like edges and holes. Noncontact are faster for surfaces and lots of data points.

Do not by anything until youhave used it yourself on your parts in your shop!
Do not by anything until youhave used it yourself on your parts in your shop!

This thigns are very tricky andmany thigns abotu the parts or the enviroment can give you bad data. The same thing goes forthe software. Make sure you can make it do everythign you want it to do.
ProEpro

http://www.whitelightdesign.com

 

[ScanOr]

We recently worked with a company based out of the National Research Council in Ottawa Canada. They have the best image data I had ever seen and were extremely informative.

You may wish to give them a call. The company name was XYZ RGB (because of a special single tube RGB laser I think) for inspection and color applications.

You can get their contact info of their site at

http://www.xyzrgb.com

p.s. their system was also mounted to a CMM.

- Dale

 

[reveng1]

I have just had a demo of the g-scan (see previous message). I have been working on clouds of points for several years now (for reverse engineering). It was really very good: very few bad points, high accuracy, and very easy to use. I had some doubts on this technology (scanner on an arm) since the scans from their competitors were really worse . The main reason seems to be that axila is actually the first and only company to manufacture the arms and the scanners too (see

http://www.axila.com

). Their competitors have to interface their arms on third party scanners, and they seem to be in trouble.
I have also seem scans coming from gom that were perfect too. But no demo.

 

[ProEpro]

You should look at a Minolta. I have seen thier data recently and it is better than from an arm mounted scaner. It is much more accurate.


ProEpro

http://www.whitelightdesign.com