Most shakers for general use are set up with current amplifiers. When using some of the more common shaker excitation techniques used today for modal testing, this does not provide for good frequency response measurements. This is especially true for burst random excitation which is very widely used in modal testing with single or multiple shakers. When using burst random excitation, the response of the system needs to decay to zero before the end of the sample interval of the FFT analyzer time capture. With current amplifiers, the armature of the shaker coil is allowed to freely float after the excitation is terminated. For very lightly damped systems, the excitation and response may linger on well beyond the sample interval.
However, when the amplifier is set up as a voltage amplifier, the back EMF effect (the electromotive force caused by the structure motion driving the shaker armature through the coil) provides resistance to the armature and helps to cause the system response to decay more quickly. This may seem to be inappropriate because the shaker system is then supplying damping to the measurement, but is not an issue as long as the force is measured for the entire measurement. Then the correct input-output relationship is measured. (It is also important to note here that the force needs to be measured and not the electrical parameters of the amplifier in order to make the correct measurement.)