3-Component-L2F devices are well suited for turbo machinery measurements but there are applica-tions where either the solid angle for optical access is too large or where not enough space is available to place and operate the optical head. In these cases the 3-Component -Doppler-L2F can be used which operates with the same confocal optical set-up as a usual 2-component L2F system, thus enabling 3-component measurements even under difficult conditions, e.g. in centrifugal compressors. This technique combines the principle of the L2F method with the principle of the Doppler Global velocimetry. From the time-of-flight data, the standard two-component measure-ments yield the velocity vector component in the plane perpendicular to the optical axis. An addi-tional frequency analysis of the scattered light determines the Doppler frequency shift, which is a measure of the velocity component along the optical axis.
Since a velocity component in the direction of the optical axis causes a frequency shift of the scattered light due to the Doppler Effect, it can be determined by analyzing the scattered light frequency. The Doppler shift of the backscattered light is a function of the laser frequency, the speed of light and the on-axis particle velocity component. The frequency shift is analyzed by measuring the transmission of the detected scattered light through an iodine cell which serves as a frequency-to-intensity converter. The transmission is the ratio of the light intensity emerging from the iodine cell (signal) related to the light intensity entering the cell (reference).
The transmission function of an iodine vapour absorption line has a very strong frequency depend-ency. In order to take advantage of this high frequency sensitivity, the laser frequency should be tuned to a certain position of the transmission slope and stabilized there. If the scattered light undergoes a frequency shift, the transmission derived from the photo detector signals changes. With the known transmission function, the Doppler shift and the on-axis velocity component can be determined.
The set-up of the 3-Component-Doppler-L2F device shown here uses as the light source a fre-quency stabilized argon ion laser with intra-cavity etalon. While operating in multi-colour mode, the Ar+-laser was frequency stabilized using its green line (=514 nm). The laser light is fibre linked to the probe head. The scattering light pulses from the 488nm and 496nm laser beams in the probe volume are guided to a L2F processing unit, where the start and stop signals for the time-of-flight analysis are generated. These signals deliver the two velocity components in the plane perpendicu-lar to the optical axis.
The velocity component along the optical axis is deduced from the Doppler frequency shift analysis using the scattering light pulses from the 514nm laser beam in the probe volume. The 514nm scattering light is collimated and split by a non-polarizing beam splitter into two beams of equal intensity. One beam is guided through an iodine cell providing the signal light pulses while the other beam is guided through a multimode fibre that serves as an optical delay line. Along this path the reference light pulses are transmitted. Both pulses are detected by a single photomultiplier sepa-rated by a constant time delay determined by the fibre length. The amplitude ratio of both pulses is proportional to the transmission of the iodine cell, from which the scattered light frequency can be derived.
In order to account for the light intensity losses within the iodine cell and the delay fibre a special calibration test is necessary. For this reason the laser frequency is established far outside the range of the absorption line where the transmission value is known to be one and constant over a wide frequency range. Care is taken to set identical integration ranges for both signal and the reference pulse. These ranges can easily be matched to each other with the aid of the known time delay of the two pulses. In the presented system the signal processing of the time-of-flight measurement and the Doppler shift analysis is performed independently, however, correlated three component velocity measurements should be feasible in principle.