02.11.2005
The resonances and disturbing motions of acceleration exciters have been investigated with the help of a 3D-Laservibrometer and a Scanning-Laservibrometer. As a consequence the influence of these disturbances on the results of primary calibration of accelerometers has been reduced largely.
The high precision calibration of accelerometer transfer standards at the PTB in a frequency range up to 20 kHz with an expanded uncertainty of 0,1 % to 1 % demands comprehensive and detailed knowledge about the performance of the facilitated acceleration exciters at high frequencies. By application of 3D-Laservibrometry and scanning-vibrometry it was possible to measure special motion disturbances (transverse motion, tilting, bending and armature resonances) of acceleration exciters as well as the influence of the mass loading on the exciter up to very high frequencies (50 kHz).
With its 3 simultaneously measuring Laser beams the 3D-Laservibrometer is able to measure the vibration velocity vector at distinctive points on the surface of the exciter's armature. By this means it is possible to measure transverse motion components together with the intended axial acceleration. The scanning-vibrometry in turn is able to measure the dynamic tilt, or deformation of the mounting surface of this armature as well as the local acceleration distribution in axial direction (fig. 1).
Figure 1: Pseudocolor display of the acceleration distribution on the mounting surface of an acceleration exciter measured at 20 kHz with a Polytec Scanning-vibrometer
The knowledge of this differences in acceleration over the surface is extremely valuable for calibrations where the measuring beam has to be aimed at a point lateral from the transducer under test. For example a membrane like deformation leads to a decrease of the axial acceleration with increasing distance from the centre of the armature's surface. Even the high quality exciters developed and optimized at PTB exhibit different instantaneous accelerations or acceleration amplitudes over the surface of the airborne armature, which might have influence on the calibration result. However, these disturbances have been identified with the described methods and are further taken into account by adapted calibration methods (fig. 2), which are capable of eliminating the major part of these influences.
Figure 2: Interferometric two channel measurement for the elimination of the influence of tilting motions during accelerometer calibration
Since the acceleration at diametric opposing measuring points may differ in amplitude and phase, the PTB applies a synchronous two channel measurement of amplitude and phase. The result is then derived from the sum of the measurement at 0° position and 180° position. Subsequently the two interferometer measuring beams are adjusted to 90° position and 270° position, respectively, and the calibration procedure is repeated. The final result is given as the average of those two measurements (0°, 180°) and (90°, 270°).
In order to minimize the influence of transverse motion and the sensor cable the interferometric measurement procedure described so far is applied for four different mounting positions of the transducer. These mounting positions are realized by turning the transducer 90° around its sensitivity axis.
With the application of the adapted calibration methods described above it was possible to extend the frequency range for the calibration of transfer standard accelerometers in the PTB to 20 kHz (formerly 10 kHz) while the expanded measurement uncertainty was simultaneously reduced from =2% to =1% at 20 kHz. Correspondingly adapted CMC entries have been submitted to the BIPM and were approved on 18th of October 2005.
Contact person:
Th. Bruns, FB 1.3, AG 1.31, E-mail: thomas.bruns@ptb.de