The ``gold standard'' employed in this experiment (which we will refer
to as calibration bar) was constructed as follows. Holes 15 mm
deep and spaced 10 mm apart were drilled into an aluminum bar by a
Matsuura MC-510V numerically controlled milling
machine
. The diameter of
each hole was carefully milled to the same diameter than the probe tip
such that after the probe was pressed into the hole, no play was
detected. The hole was sufficiently deep that the probe tip could
descend until an increase of the probe width limited further travel.
This milling process, which is accurate to 
0.005 mm, ensured
that the probe tip would descend the same amount for each hole.
In order to allow the probe attached to the FARO arm to also be localized by the OPTOTRAK, four LEDs mounted onto a small aluminum plate were attached to the probe. This was achieved by inserting the probe in a hole, whose diameter matched the shaft of the probe, reamed into the plate and secured to it by threaded fasteners.
The proprietary program running on the PC that converts the voltages from the FARO arm into numerical signals (see Chapter 1) does not provide the facility to write the probe holder position information directly into a text file. This task was therefore performed by modifying the Viewing Wand software, and required that the entire Viewing Wand system be moved to the location where the experiment took place. The coordinates returned by the FARO arm throughout this experiment were thus furnished by the ISG Viewing Wand to which the arm was interfaced. Note that the role of the Viewing Wand was limited to storing the raw data from the FARO arm, i.e., the capability of the Viewing Wand to compute the position of the tip of the probe was not used.