From: Arthur W. Brooks
Sent: Wednesday, August 09, 2006 10:38 AM
To: Steve Raftopoulos
Cc: Bradley E. Nelson; Hutch Neilson; 'Mike Cole'; Thomas G. Brown;
Wayne T. Reiersen; Michael E. Viola
Subject: RE: VVSA1 Data after best fit translation

Attachments: VVSA1_MTM.pdf; VVSA1_pppl.pdf

Steve,

Re: "08_07_06_VVSA1_SHELL_BEST FIT.xls"

Attached are plots of all the scan points for VVSA1 Shell relative to our design targets. From a FW encroachment standpoint, VVSA1 is still acceptable and perhaps even slightly better than the MTM scan data showed. 

In plots 1 and 2 (expanded scale) the measured data was sorted by its distance to the FW and compared to a comparable set of design data  (where each point is a projection of the measured data on the design VV). The design data is also sorted (independently) by its distance to the FW, so both curves are monotonicaly increasing. This gives a comparion of the distribution of distances to the FW over the entire surface,  but not a point to point comparison. In plots 3 and 4 (expanded scale), the design data sort is maintained and the measured data is put in the same order (by point id) to give a point to point comparison. The data in both sets of plots is the same, just viewed differently. 

While there are many points out of tolerance (plots 3 and 4) inward toward the FW (789 of 9278 or 8.5% for PPL scan vs 816 of 8450 or 9.7% for MTM scan) for VVSA2, the distance to the FW surface where it is tightest has not been reduced. Also, over the full surface area, the number of points/places where there is not adequate room for a full first wall thickness (plots 1 and 2) is comparable to the design.

Art