Scan 3d

[mange34]

I have a problem with scaning compared to a 3D part .

I want to scan the light blue surface as you can see on the pix i have instert.
then i want to evaluate the scan with best fit.

i have come so far that i have measure the part but when i see the results of the scaning it dosen´t look right .

it´s when i scan over the small radius it dosen´t work. it looks as i have the ball compensation in just two axels.


i use SP-25 and PC-dmis 3,7 MR3

:

[slesholdofthreep]

I am by no means a scanning pro, and others can confirm or deny this, but I believe you are correct about the comp only in 2 axis. Maybe convert scan to point and ouput rawhit data and use a cad package to add spherical comp.

[Matthew D. Hoedeman]

I am by no means a scanning pro, and others can confirm or deny this, but I believe you are correct about the comp only in 2 axis. Maybe convert scan to point and ouput rawhit data and use a cad package to add spherical comp.

The comp will depend on WHAT probe you have. I only have the TP2 and it is strictly 2-D scanning capable, unless you turn off the probe comp.

Now, about your scans, are you generating them from the CAD data or are they 'learned'?

If you generate them from CAD data, the points will have the nominal 3-D cad vector and will comp along those. If it is a learned scan, it will be only 2-D, unless that probe you are using is a scanning probe and even then I am not 100% sure about it. I am only 100% sure about the TP2. You CAN get true, 3D comp, IF YOU USE surface points instead of VECTOR points in the scan window and then ALSO use 3 surface sample this. Of course, thei measn you will be taking 4 hits for every 1 you collect. SUCKS!

[slesholdofthreep]

Matt, I was hoping you'd chime in. Good info

[Matthew D. Hoedeman]

Matt, I was hoping you'd chime in. Good info

Yeah, MOST versions of Pcdmis do a great job on the 2-D scans and all suck at 3-D, unless you want to take 4 this for each point. I am SO [email protected] GLAD we got a laser camera, no more attempts at 3-D surface scan for me! YIPPIE! In fact, I used it just 2 days ago on this little bitty part. 450,000 scan point in about 45 minutes.

[slesholdofthreep]

Whats the are the specs of the laser camera? Was it an easy retrofit?

[Matthew D. Hoedeman]

Whats the are the specs of the laser camera? Was it an easy retrofit?

It's a stand-alone camera, doesn't use Pcdmis at all, all the data goes straight into Geomagic where you make your surfaces.

It is here:
http://www.konicaminolta.com/instrum...d9i/index.html

[robehor4]

I have a problem with scaning compared to a 3D part .

I want to scan the light blue surface as you can see on the pix i have instert.
then i want to evaluate the scan with best fit.

i have come so far that i have measure the part but when i see the results of the scaning it dosen´t look right .

it´s when i scan over the small radius it dosen´t work. it looks as i have the ball compensation in just two axels.


i use SP-25 and PC-dmis 3,7 MR3

:

What type of scan are you doing.

I would suggest 3 rotary scans about the z axis.

[mange34]

Thx for all support.

mathew you wounder what kind of probe i have and i have a SP-25 scan probe and i generate against a 3-d part ..

and i have problem to genererat to .. sometimes it works and sometimes don´t.

Have anyone notised that it can be any diffrent between iges.step and direct cad ??