|
The beam focus of
high-power fiber lasers can be characterized via imaging of Rayleigh
scattering in air-an indirect detection method that avoids damage to
detectors due to ultrahigh-power beams.
07/08/2014 KEVIN KIRKHAM
John William Strutt was born on Nov. 12, 1842, in Essex, England. His father
was the second Baron of Rayleigh, and upon his death, Strutt became the third
Baron of Rayleigh. In 1871, Lord Rayleigh explained that the sky was blue
because shorter wavelengths of light are more easily scattered by the gas
molecules that make up our atmosphere (see Fig. 1). Longer wavelengths are
similarly scattered but at a statistically lower amount (inversely
proportional to the fourth power of the wavelength).
|
Lord Rayleigh
could not have predicted the uses of the scattering phenomenon he described;
Rayleigh scattering has shown to be an excellent way to monitor the
performance of high-power lasers and the optical delivery systems used to
bring the beams to the workpiece.
Since the early 1960s, the fiber laser has held the promise of an inexpensive
coherent light source that can easily be brought to the workpiece. These
workpieces can now include the human eye (via laser surgery), a sheet of
steel (via industrial laser cutting and welding) or an enemy rocket (via
military lasers now in development), among countless others.
With the advent of cladding pumping, ytterbium (Yb)-doped fibers, and
high-power/high-brightness diodes, modern high-power fiber lasers have become
an impressive reality. Fiber-laser architecture is modular, permitting a
near-limitless amount of total power to be realized by combining the output
of multiple modules. In one example, IPG Photonics (Oxford, MA) recently
installed a record 100 kW materials-processing fiber laser for Japanese
joining technology supplier NADEX (Nagoya City, Japan).
Measuring
high-power lasers
All laser processes benefit from the use of diagnostic equipment, if only to
assure the operator that the laser and laser beam delivery components are
still operating as expected. Of course, the ability to effectively monitor
any process is a necessary precursor to mastering that process.
More and more applications require fiber-laser parameters to be optimized and
consistent. Predictable and desired outcomes can be better assured by
providing accurate, repeatable laser performance to the workpiece. Parameters
important for efficient laser processing of the product or process include:
beam intensity, waist size, and waist location of the X, Y, Z process volume.
Most important is the ability to provide stable processing profiles of these
measurements over time.
Compounded laser measurements, such as beam quality (m2) and beam parameter
product (BPP), are other ways of defining or predicting the primary
laser-processing parameter, which is spot brightness. Commercial
beam-monitoring systems use beamsplitters, waveguides, or scanning slits to
sample and attenuate high-power lasers.
While these systems are practical in that they use these sampling techniques
to effectively match the sampled laser-beam intensity to the usable range of
the profiling sensor, the sampling techniques add some distortion and
uncertainty to the measurement data. Traditional diagnostic techniques
measure one or two parameters at once, providing a myopic view of laser
performance and argue the need for a number of diagnostic tools to be used
sequentially or in tandem.
Laser-beam diagnostics have always included a way to measure the intensity of
the beam?either the average power or the energy of each laser pulse. This is
accomplished using thermopile and quantum-type sensors to measure the average
power or with pyroelectric sensors to measure the energy per pulse.
More recently, ways to monitor the laser mode or beam profile have become
available to provide greater knowledge of the laser system's performance to
the operator. These systems consist of scanning-slit or CCD/CMOS-camera
sensors that measure the beam's 1D or 2D intensity distribution. These
systems are also capable of measuring the location of the focused spot or
minimized beam in the processing zone. Optimally, beam-profiling systems
should indicate how the power density changes across the area of the focused
spot.
More complex systems that measure M2, BPP, divergence, and mode-field
diameter are now available. These systems are based on beam profilers that
can translate the spatial-measurement setup along the beam-propagation axis.
This results in a more complete understanding of laser performance. This
toolbox approach to monitoring multiple laser characteristics is possible but
not always practical, and certainly not effective in a manufacturing
environment.
Most current techniques require the beam to be optically sampled; anything
placed in the beam will cause some amount of distortion to the sampled beam
as well to the working beam. Focused multikilowatt fiber lasers achieve power
densities of many megawatts per square centimeter. Thermal lensing of the
sampling optics and the potential for damage of transmissive optics push
traditional beam-sampling techniques to their limit and expose the diagnostic
system to ultimate failure.
Noncontact
beam measurement
We have developed a Rayleigh-scatter-based profiler that provides instantaneous
measurements of spot size, focal-spot location, focal-plane shift in the
propagation and lateral axes, and m2 and BPP measurements. This beam
profiler, BeamWatch, can assess lasers that have power densities of 2 MW/cm2
or greater without requiring any optics or waveguides to be placed in the
beam path. Because no optic interacts with the beam, there is no practical
limit to observable power densities.
The profiler uses a CCD camera to image the laser path over a distance of
more than 25 mm as it propagates through focus. The camera frame rate permits
the focal plane, and thus any focal-plane shift, to be continuously monitored
in ≤40 ms intervals.
The VECSEL chip itself has 10 indium gallium arsenide (InGaAs) quantum wells
(QWs) 8 nm
|
The operator
screen displays the beam caustic in false color, along with the initial and
current focal plane location. Calculated results include m2, waist width,
waist location, focal-plane shift, divergence, BPP, power, and centroid
location (see Fig. 2). These measurements can be analyzed to test if they
fall within a pass/fail window. Results that fall outside of the predefined
minimum-maximum windows are immediately enunciated and the operator alerted.
|