Invalid Feature Vectors

  • Filter
  • Time
  • Show
Clear All
new posts

  • Invalid Feature Vectors

    Good morning,

    We were running a linear open scan. After the scan had completed, a box popped up asking us if we wanted to increase the find nominals number for this scan. This has happened before, and if we make the number bigger, the scan completes. We are not scan experts, but we think that means that the particular part we are scanning, is more irregular than most, and we have to increase that number so the machine will think it's ok to continue. Hopefully that doesn't affect the report, or the integrity of the scan. This particular time, we increased that number to ten mm, and it still didn't like it. We pushed cancel, and it finished the scan, and reported it. However, it just showed the scan itself, and not the "side rails", on the graphics. The reported number was in tolerance, just barely. ( This feature has a profile callout of six mm ). I tried opening the dialog box and changing the value there. The next time after it ran that scan, it stopped and threw up this alarm_ "All feature vectors are invalid for Vector Best Fit alignment A1" I deleted the report, deleted the scan, remade the scan and the report, and tried it again. The same invalid feature vectors alarm popped up. I'm at a complete loss. We've used this program to scan about thirty parts, and this is the first time this has happened. The only thing that changed initially was the part. And it doesn't visually look like it should be the problem.

    Any suggestions?


  • #2
    Do you have another part to try? Most helpful would be if you have one handy you have already run previously. This could help to determine if the issue you are experiencing today is due to the part or an issue with Pc-Demon.

    sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery


    • #3
      You need to verify the surface vectors of the part and " fix " them as needed. Sometime a " surface" may be backwards. Hard to explain and easier to show but is the surface is wrong, when you generate the scan you are trying to scan " in material " and not on the outside. You should do this on all surfaces of your CAD model before you start programming as step 1 after you import your model and create a single feature.


      • #4
        Ok, we're going to try another part, we have lots of those. As for verifying the surface vectors, not sure how to do that. This is an iges file, and when we import those the box asks us if we want to fix surface vectors, and we always say yes. Is there an additional step? Would love to know how to do that if there is. Not sure if that's it though, because we've successfully measured about thirty of these parts. There are two of these surfaces, identical to each other, and separated by about an inch. They both seem to report normal. That is, where there is a void in the surface, the scan graphics show a "dip" in the right direction.

        I'll check another part now, Thanks everyone for the quick response.


        • #5

          It shouldn't matter that we are reporting the profiles in Xact, and some other features in Legacy should it? just wondering


          • #6
            Another part worked, I think that's what happened. However I did redo the scan and the report again so I don't know. I'll check one later today, that I've already done.


            Related Topics