4 Sensors for Process Monitoring272
33
Heuer, W., Dissertation; Universität Han-
nover, 1992.
34
Tönshoff, H.K., Falkenberg, Y., Mohl-
feld, A., IDR Ind. Diamanten Rundsch. 1
(1995) 43–48.
35
Heinzel, C., Dissertation; Universität Bre-
men, 1999.
36
Karpuschewski, B., Brunner, G., Falk-
enberg, Y., Jahrbuch Schleifen, Honen,
Läppen und Polieren; Essen: Vulkan-Verlag,
Ausgabe 58, 1997, pp. 146–158.
37
Brinksmeier, E., Heinzel, C., Witt-
mann, M., Ann. CIRP 48 (1999) 581–598.
38
Chang, Y. P., Dornfeld, D. A., Ann.
CIRP 45 (1996) 331–334.
39
Sabotka, I., Dissertation; TU Berlin, 1991.
40
Funck, A., Dissertation; TU Berlin, 1994.
41
Bönsch, G., Dissertation; RWTH Aachen,
1992.
42
Choi, G. S., Choi, G. H., Int. J. Machine

Hinkenhuis, H., Regent, C., in: VDS-
Fachtagung ‘Schleiftechnik im Wettbewerb’,
9–10 October 1997, Aachen, pp. 8/1–13.
52
Tönshoff, H. K., Karpuschewski, B.,
Paul, T., VDI-Z. 133 (7) (1991) 58–63.
53
Paul, T., Dissertation; Universität Hanno-
ver, 1994.
54
Rowe, W. B., Yan, L., Inasaki, I., Malkin,
S., Ann. CIRP 43 (1994) 1–11.
55
Inasaki, I., VDI-Ber. 1179 (1995) 31–45.
4.5
Laser Processing
V. Kral, O. Hillers, Laser Zentrum Hannover, Hannover, Germany
4.5.1
Introduction
The field of laser manufacturing has been expanding rapidly in the last 20 years.
New materials and new laser sources have considerably increased the potential of
laser applications. This expansion has led to a necessity for higher quality and re-
producibility when using lasers. To satisfy this demand, a considerable amount of
research has been expended into the use of sensing for laser applications.
The goals of implementing sensing technology in laser manufacturing can be
categorized into two groups.
1. Sensors that monitor the process parameters. These measure the external variables
that affect the laser process. This may mean monitoring of the beam character-
istics, the workpiece-head distance, the geometrical accuracy of the workpiece,
the process gas, filler material feed, the material quality of the workpiece, and

Imaging systems, such as the SCOUT [1], are used as seam tracking devices. A
camera is mounted on the robot arm and allows the system to adjust the robot
motion relative to the seam position. Camera-based systems often have problems
when dealing with highly reflective surfaces, since scratches and reflections can
be misinterpreted. Another approach is the use of a mechanical guide to follow
the seam [2], as shown in Figure 4.5-1. The use of this method is limited by the
feed rate, since the mechanical guide is usually spring loaded. The advantage is
that a certain robustness is guaranteed.
4.5.2.2
Sensors for Identifying the Workpiece Quality
Both the material quality and the surface cleanliness can considerably affect laser
welding or cutting processes. Errors due to unacceptable material quality are diffi-
cult to detect on-line. One approach is to use the process radiation from a high-
power laser pulse to classify the type and quality of the steel. The process radia-
tion is analyzed using a spectral analyzer, and the individual material-specific
4.5 Laser Processing 273
emission peaks are correlated with that of a reference steel. Using this method,
undesirable changes in alloy concentration can be detected.
4.5.2.3
Sensors for Beam Characterization
In laser processing, the quality of the beam plays a considerable role in achieving
high process quality. For this purpose, sensors are being developed for monitoring
the beam power distribution. One common approach is to use a shaft with a pin-
hole that scans across the beam [3]. Coupled to a mirror, the radiation is reflected
to a sensor. With such a system, the intensity distribution across the beam is mea-
sured. Another system requires the placing of a thin wire grid in the beam path
[4]. From the temperature change of the individual wires, a complete beam power
distribution can be reconstructed.
4.5.2.4
Focal Position and Gas Pressure

plexity of the failure identification algorithms, and has so far not been industrially
implemented.
The optical sensor approach is the most promising approach to identifying
faulty laser processes. Unfortunately, even though variations in the process radia-
tion can be correlated with faults, some faults are not always detectable by observ-
ing the process radiation signals.
4.5.3.2
Acoustic Sensors
Another approach is to use microphones or ultrasonic sensors to measure process
faults. This is particularly useful in the piercing phase of a cutting process [9].
Using these sensors, the piercing time can be optimized. Other approaches in la-
ser welding have shown that it is possible to correlate the acoustic signals with
weld-pool vibrations when a pulsed laser process is being used. In micro-structur-
ing using excimer lasers, ultrasonic measurements on the workpiece have shown
that the removal rate for each pulse can be monitored [10].
4.5.3.3
Visual-based Sensing
This approach deals with the complete visualization of the effects of the process.
In welding, for example, key-hole diameter and weld-pool dynamics [11] can be
monitored using charge-coupled device (CCD) cameras. The only difficulty lies in
choosing the proper filter so that the bright process radiation does not overexpose
the image. This is particularly important when dealing with processes that pro-
4.5 Laser Processing 275
duce considerable plasma radiation. Another approach is to use thermal imaging
after the process. From the thermal distribution, irregularities in welding pro-
cesses can be identified (Figure 4.5-2). To increase the signal processing speed, in-
frared line-arrays can be implemented [12]. Another purely scientific approach is
to use X-ray imaging to identify the vertical melt-pool dynamics [13]. Such a re-
search tool has been used to identify the formation of pores in welds.
Unfortunately, visual systems are complex and expensive. Fast feature extraction