Koji and i updated the shaker today. We replaced the short multilayer piezoelectric actuator (10mm) by a longer one (20mm) from the NEC TOKIN’s we have. (datasheet)
The pzt is glued to a brass disk, about 2" diameter and clamped between the side of the table and the steel frame around it using a aluminum base on the other side. (will add photo later).
We use a modified PMC servo card as a HV piezo driver. The modified schematic can be found below.
We added a 1kOhm resistor in the output which forms a ~100Hz low-pass with the 1.5uF capacitance of the PZT.
We get a good SNR ratio for TF measurements even when using white noise as the source. doing some low-frequency TF measurement over night.
WiIl also try a swept-sine measurement if required, but takes too long at low frequencies.
We also tried to build a simple loop using two stanford preamps to suppress the horizontal seismic motion of the table but couldn't see any improvement. Will wait for the measured TF to design the right loop.
HV amplifier schematic (modified PMC servo):