Hello, I have struggled to reverse engineer impellers for quite some time now. I am hoping to get some modeling advice from some of the users here. How we currently reverse engineer impellers is by using a Faro ScanArm to probe the existing impeller and then use the scan data as a reference. We constantly struggle to model the impeller vane. We have yet to find a consistent modelling technique that works for each impeller. Here is the current one I am working on and the current approach.

Scan Data

Loft Profiles

They are ellipses that are constrained by reference sketches

Guide Curves

Solid

As you can see the loft is not smooth. For our application, the vane needs to be extremely smooth. If someone could give me advice for the current approach or thoughts for a new approach it would be greatly appreciated.

I have attached a file for demonstration purposes. The file contains scan data from the inside vane surface only.

Impeller Demo.SLDPRT - Google Drive

Thanks,

Dustin