Thin wall feature

[[He...]]

Hello gentlemen,

I have a thin wall feature and could not have the probe hit on the right place. Could you please walk me through how to put the formular on there?

Thank you. 🙏   

[[Ma...]]

You would need to measure plane above, use it in alignment and then scan circle in that alignment. That circle will scan along that plane.

Of course if that part is not warped or not machined on one go, then you would need to try more approaches.

[[He...]]

I did use the plane above in my alignment but for that thin wall the probe keeps missing round scan. So, I thought the formular could help. 

[[Ma...]]

Do your probe fall of the part or it's hitting that torus in the part?

If you have problem with warped part, then i would try to get 4-6 points individually with help of touching point on that plane.
Otherwise i would try curves ( to obtain a few point to construct circle )

[[He...]]

Yes it' hitting that torus then went off the part, thank you.

[[Je...]]

Please sign in to view this username.

, if you have curve, I would recommend trying the "edge curve" strategy that is built into the curve feature window.

It can measure that edge based on a set of guide points on the opposing face.

[[Iv...]]

Please sign in to view this username.

 Because you mentioned that the probe keeps missing around the scan, I believe you may have too much warping, as

Please sign in to view this username.

 hinted at. If this is the case, using an edge curve may have similar results since I believe the software will continue to evaluate measuring height as a fixed height. However, in your case, the measuring height is variable enough that each measuring location should be evaluated separately. Also, combined with the thinness in your geometry, it leaves very little room for error. 

I'm not sure if it was permissible to use touch points instead of scanning. But if so, I believe the process below will work for you. 

• For easier math, I would create a theoretical sub-alignment relative to the circle. You could make this a real sub-alignment with an actual circle to better locate the following features. 

• Create a point that is relative to the sub-alignment. The point should probe at the theoretical edge of the circle and have the correct nominals for that circle. 

• Then use the Projection button (previously called the None button) and set up an edge measurement. This will require Calypso to take a point on that top plane before proceeding with the actual measurement. You can then control the height og the probing point here

• Set up a pattern around that circle.

• Recall the points into a circle

Evaluating each location separately should take care of that air-probing issue. It is difficult to confirm with that picture you posted, but if that edge measurement is the smallest ID in that bore, you could consider using a cylinder probe. 

[[He...]]

Please sign in to view this quote.

Hi Jeff, thank you for that and I'll keep post. 

[[He...]]

Please sign in to view this quote.

Hi @Ivan, thank you for that and I'll keep post. 

[[Je...]]

Please sign in to view this username.

, I like your strategy of manually programming locator points to assist navigation.

I wanted to mention that Edge Curve is perfect for warped and thin parts.  I've used it on sheet metal and other wavy surfaces.

It's intelligent enough to adjust probe depth based on individual embedded points in the edge scan.  In essence, it's the automatic version of the strategy you suggested.

Please sign in to view this username.

, I'm interested to hear how this turns out for you.  There's definitely more than one way to solve the problem, which speaks to how versatile Calypso is at solving problems that other metrology software cannot.

 

[[He...]]

Please sign in to view this quote.

Herer is the strategy that you told me. please review if I miss anything?

 

 

[[He...]]

The OD 1.410 

[[He...]]

Please sign in to view this quote.

 

Please sign in to view this quote.

Hi Jeff, I did the strategy that you told me use Edge Curve to scan the thin OD but the probe not event touch the part/miss touch. Please let me know if I miss anything. Thank you. 

[[Je...]]

Hi

Please sign in to view this username.

.  Were you able to assign the scan path for the guide surface (reference scan) that assists your curve feature of interest:






Both the D2 and D1 settings here are critical.  Typically, you want to set d1 setting to be half of the nominal material thickness.

 

Here's a topic that discusses this:

 

 

[[Je...]]

And here's another example of an edge curve.

D1 is set to "0" because I intended to scan the feature of interest at its nominal location (no offset).  It worked terrific for this part, which was warped severely.