BRUEL & KJAER System V Instruction Manual page 54

sponse of the Head itself is eliminated as a source of error. The next consideration is the
response of the Standard Accelerometer. Below 5 kHz this response is independent of
frequency. Above 5 kHz there is a slight upward swing to a maximum of + 0,5 dB (6%) at
10kHz. If a ca libration is to be run at these high frequencies, it is a simple matter to
adjust the sensitivity setting on the preamplifier to compensate for the increased output,
by reference to the accelerometer calibration chart. The influence of test levelon sensi­
tivity is so small as to be not measurable over the dynamic range of the 4815.
The final consideration is the cross-motion of the Head. If, for example, this is taken as
2%, the contribution to the measured acceleration level is dependent on the difference in
transverse sensitivity between the Standard Accelerometer and the test accelerometer.
For example, if the Exciter is running at 100 g at 1 kHz, there is a transverse acceleration
of about 2 g. If the Standard Accelerometer has a transverse sensitivity of 1,7% it will be
reading 2 x 0,017 g high, i. e. 1 00,034g. If the test accelerometer has a transverse sen­
sitivity of 3%, it will be reading 2 x 0,03 high, or 100,060 g. Since the ca libration of the
test accelerometer is based, not on what the absolute acceleration level is, but on what
the Standard Accelerometer measures, there will be an error of + 0,026 g introduced. If
both accelerometers had the same maximum transverse sensitivity and were aligned in
the direction of greatest sensitivity, there would be no error introduced by cross-motion.
53