Faro Arm, Gage R&R

sdeters

Registered
We are doing a Gage r&r with the Faro arm probe and laser scanner quantum S, and Polyworks software. We are struggling badly. We have 10 die-cast housings we are running this study on.
We are all over the place with results, and I am trying narrow down where the issue is.
For my first question here, has anyone done a Gage r&r with a Faro arm, with polyworks, and has had good results?
 
Last edited:

Miner

Forum Moderator
Leader
Admin
Is the problem with repeatability, reproducibility or both? Posting your study will help.
 

Miner

Forum Moderator
Leader
Admin
I think you need a minimum of 5 posts before you can upload a file.
 

sdeters

Registered
Oh ah I finally can attach the study, Cool. Please review this study below.
I am learning this, and these studies take a long time.

I am hoping we can figure out where the error is happening and address it.

Thanks for anyfeedback you could provide.

Shane
 

Attachments

  • FILTERED - GR&R - ANOVA1.xlsx
    1.5 MB · Views: 303
Last edited:

Miner

Forum Moderator
Leader
Admin
In all of the characteristics, repeatability is the biggest issue. Several do have some operator to operator issues, but those are usually easily solved through standardizing the method and training the operators to the method.

For repeatability, the issues are more difficult to identify. The following are the more common issues:
  • The measurement device itself is ill-suited for the measurement
  • Low rigidity of measurement device or fixturing
  • Inadequate clamping of part being measured
  • Wear or poor maintenance of the measurement device
  • Cleanliness of part, fixture or measurement device
  • For CMMs:
    • selection of probe diameter,
    • probe length,
    • cosine error,
    • # of hits used to identify a feature
  • Variation in form of the part being measured
I am sure there may be additional issues associated with the laser scanner and software, but I am not familiar with those.
 

sdeters

Registered
Ok Thanks for the feedback, we will check out all of these topics.
I now have a place to start, which kind of was in the back of my mind anyways.

I will need to start another thread on your # of hits used to identify a feature or create a feature. Because two of our alignment datums are smaller "cones" tapered core features. I think once we figure out how to probe these features with the three R's our data will be a lot better.
 
Top Bottom