From: Arthur W. Brooks
Sent: Thursday, May 15, 2008 10:54 AM
To: Mark Smith
Cc: Thomas G. Brown
Subject: Simple replacement for invkinematics2a.f routine

Attachments: amat.051408; transpts.f; transpts.inp

Mark,

As I mentioned yesterday, the invkinematics2a.f routine was written to do more than 
just generate the trace of points for the lasers, cg and support points. As a result 
the code is much more complicated than needed if you are planning on recreating it 
in MATLAB. The attached code "transpts.f" is fairly simple and uses the transformation 
matrices in amat.051408 to do a simple multiplication of points in the 
transpts.inp (read from FORTRAN standard input *).

I've done a spot check of the transformation matrices with ProE 
(Analysis -> Measure ... Transform from ACS0 (top Assembly) to CS0 (Coil Part). 
Just note that the amat.051408 assumes SI units (meters) for the translation 
component (4th column). The code handles the conversion internally 
(read inches -> converts to m for matrix multiplication and then converts back 
to inches-> outputs inches).

Note also that the amat.051408 is independent of the mechanism and servo tables 
used to impart the motion. It just represents the 'state', not how to get there.

The other the routines stay the same (lasertrace2.f & genpts_ibl.f)

I hope this is useful.

Art