<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
<font face="Arial, sans-serif" size="2">
<div>Hi All, </div>
<div> </div>
<div>I met with a bunch of my NIST colleagues yesterday afternoon to
discuss technical support for the tool wear demo at our upcoming June
meeting. We have some interesting options. </div>
<div> </div>
<div>At the Wednesday telecon, Larry Maggiano mentioned that one of his
colleagues talked with NIST's Eric Stanfield about using one of our
measurement systems to measure actual tool wear. I'm working with Eric
to figure out how to do this with minimal turnaround.
</div>
<div> </div>
<div>We have a dynamometer and can hook it up to the machine tool in a
couple of modes. One mode is to mount the part atop the sensor and
continually measure forces. Another is to mount a sacrificial piece of
the same metal as we're cutting atop the sensor,
placed off to the side, and periodically take cuts while measuring
forces for a short period. </div>
<div> </div>
<div>The dynamometer is an "AC" device, meaning that the signal is
proportional to the change in force, and a time integration is done in
the support electronics to turn the change in force over time into a
force measurement. The sensor is thus good for high-frequency,
short-term measurements. This argues for the periodic sacrificial cut
mode. Another argument for the periodic mode is that it's easier to set
up. The part can be clamped onto the table as usual, without regard to
the sensor cables etc. </div>
<div> </div>
<div>I propose that we set up the demo so that we cut the boxy part as
usual, and include periodic breaks from cutting, say after each setup,
during which a simple well-defined cut is done while we measure forces.
</div>
<div> </div>
<div>Regardless of the mode, we still want to incorporate the force
measurements into the software somehow. How do we plan on doing this?
One way is to simply compare what forces are expected, based on a model
(from Boeing?), and congratulate ourselves when
they match. Another way is to bring the measured forces into a model
for a learning-type application. </div>
<div> </div>
<div>I think using the dyno will be simple, but coming up with a
software interface to bring the force data into ST-Machine will be the
hard part. NIST can write the DAQ software to get the forces and log
them out into whatever format is convenient. </div>
<div> </div>
<div>What do you all think? </div>
<div> </div>
<div> </div>
<div>--Fred</div>
<div> </div>
<div><font face="Arial, sans-serif">Frederick M. Proctor</font></div>
<div><font face="Arial, sans-serif">Group Leader, Control Systems Group</font></div>
</font>
</body>
</html>