Output signal from the quadrant detector. 

Contexts in source publication

Context 1

... A , B , C , and D are signals from individual elements of the quadrant photodiode, Fig. 3. A differential amplifier is then used to condition the output signal as shown in Fig. 3. In Fig. 3 ͑ a ͒ , the elliptical irradiance pattern produces the strongest signal in quadrants A and B , hence ͑ the workpiece is above the focal point ͒ a positive voltage is generated ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭϩ v e . In Fig. 3 ͑ c ͒ when the elliptical irradiance pattern is strik- ing mainly quadrants C and D ͑ the workpiece surface is below the focus ͒ , a negative voltage is generated. ͓͑ A ϩ B ͒ Ϫ ͑ C ϩ D ͔͒ ϭϪ v e . When the beam waist coincides with the workpiece surface, Fig. 3 ͑ b ͒ , Z ϭ 0, the system is in focus. In this case, the light collected by the detector produces a zero output voltage from the differential amplifier ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭ 0. A closed-loop control system, using a dc servomotor supported by a commercial fuzzy logic controller ͑ NLX230, RS catalogue ͒ , is used to measure and control the stand-off distance. The fuzzy logic controller is utilized to compensate for different material types, which greatly affects the surface and reflectivity function and achieve normalization of the output. The bandwidth of the signal is controlled by the detector response time, which is 15 ns. The effectiveness of laser cutting processes depends on many parameters such as: laser power, beam feed rate, gas type and pressure, and the focal height offset stand-off distance. Stand-off distance is the distance between the cutting nozzle and workpiece surface. This controls the position of the beam focus relative to the surface. If the stand-off distance can be accurately measured, the processing beam can be controlled to accurately follow a workpiece profile. In the laser cutting process, the cut quality is mainly assessed by the kerf side wall surface roughness, which is characterized by the surface striation frequency, dross attachment, and the extent of the heat-affected zone. A coaxial type nozzle head with gas jet was used for the cutting process. Figure 4 illustrates the system setup. The laser beam is focused onto the workpiece through a convex- planar lens; the focal point of the lens is positioned 1.0 mm beneath the nozzle tip. The cutting nozzle is mounted on an IBM 7540 SCARA Robot. The robot is moving in the work- space via three axes X , Y , and Z . The optical sensor is fixed directly to the nozzle and connected to the servo control system to drive the robot end effector in the Z axis direction. A program written using the AML/E language is used to drive the robot and follow the cut trajectory ͑ X, Y ͒ . Focus control occurs just ahead of the laser cut or weld. The robot control program ensures that the measurement zone is kept in the optimum position depending on the type of process. An optical fiber is used to deliver the laser beam from the Nd 3 ϩ -YAG laser to the cutting nozzle. The laser is a Lumon- ics model MS 830 and generates a beam power of 400 W. The laser energy is available in the form of a pulsed beam of monochromatic and highly directional light at a wavelength of 1064 nm. Two main series of experiments were conducted using mild steel ͑ Fig. 5 ͒ and stainless ͑ Fig. 6 ͒ . Experiments were for stand-off distance ͑ 0.0 mm ͒ , ͑ 0.5, 1.0, and 1.5 mm ͒ above the workpiece surface and ͑ 0.5, 1.0 mm ͒ beneath the surface. Figures 5 and 6 illustrate the variation of the cut quality for different stand-off distances for the stainless steel and mild steel, respectively. From the figures it can be concluded that the kerf surface roughness increases when the nozzle offset ͑ stand-off ͒ distance is increased or decreased above or below the optimum zero value. This is because the spot size on the workpiece increases, above or below the focus, which results in a lower irradiance. The surface roughness for the different kerfs were measured and plotted with respect to the variation in the stand-off distance, Fig. 7. It can be seen that the optimum cut quality was achieved when the displacement from the focus was zero. The Taylor–Hobson, Talysurf 10 which give us a meter indication of the roughness average on a dual scale ͑ metric/British units ͒ meter is used as surface roughness measuring instrument. The system was calibrated experimentally by measuring surface light variations with the stand-off controller and comparing this output with that from Taylor–Hobson, Talysurf 10 surface roughness measurement device. There is no trade-off between resolution and response time. However, at 1 ␮ m, the measuring range is Ϯ 5 mm. An on-axis noncontact sensor capable of accurately sensing and measuring the workpiece surface position in real time has been developed. The autofocus system is located ahead of the processing beam and not coaxial, the measurement position is controlled to be in the optimum position via the robot controller. The output of the sensor is a voltage signal, which represents the axial displacement of the stand- off distance between the robot end effector and a workpiece surface. Changing the size of the pinhole, type and quality of the collimated beam splitter, lens size, and the angle of the workpiece surface has an effect on the characteristics of the sensor output. As the pinhole diameter is decreased, the sensitivity of the sensor increases; this, however, leads to a noisy signal due to speckle. The optimum size for a particular design is arrived at by compromising between sensitivity, output signal truncation, and noise effects. Using a smaller main lens and bigger image lens gives a more sensitive output. Changing the angle of incidence on the workpiece by a few degrees has little effect on the sensor output. Curve No. 1 on Fig. 8 was plotted for an angle of incidence on the workpiece of zero degrees; this provided a reference signal. Curve No. 2 shows the output signal after the workpiece was rotated clockwise through 5° from the horizontal feedrate vector, while curve No. 3 was produced by a counter clockwise rotation of 15° from the horizontal feedrate vector. The shifting position of the output curves occurred because the rotational axis ͑ offset ͒ of the workpiece was not coincident with the focused spot on the workpiece surface. From these results it can be seen that sensor output is largely indepen- dent of the angle of the incident beam. An on-axis high-bandwidth noncontact optical sensor capable of accurately sensing and measuring the workpiece surface position in real time has been presented. The sensor utilizes a low power diode laser beam and integrated closed- loop control system for materials processing. The device maintains the robot end effector at a specified stand-off distance from a workpiece. In material processing, most workpiece surfaces are matt and hence generate a diffuse reflection, which destroys the Gaussian intensity distribution of the laser beam upon reflection; this explains why the system still works well when the workpiece is set at an angle. The system gives an extremely useful linear response region which has an accuracy resolution of 1 ␮ m; it achieved a lateral resolution for 20.5 ␮ m. The control of stand-off distance is important to achieve good quality machining with laser ...

Context 2

... A , B , C , and D are signals from individual elements of the quadrant photodiode, Fig. 3. A differential amplifier is then used to condition the output signal as shown in Fig. 3. In Fig. 3 ͑ a ͒ , the elliptical irradiance pattern produces the strongest signal in quadrants A and B , hence ͑ the workpiece is above the focal point ͒ a positive voltage is generated ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭϩ v e . In Fig. 3 ͑ c ͒ when the elliptical irradiance pattern is strik- ing mainly quadrants C and D ͑ the workpiece surface is below the focus ͒ , a negative voltage is generated. ͓͑ A ϩ B ͒ Ϫ ͑ C ϩ D ͔͒ ϭϪ v e . When the beam waist coincides with the workpiece surface, Fig. 3 ͑ b ͒ , Z ϭ 0, the system is in focus. In this case, the light collected by the detector produces a zero output voltage from the differential amplifier ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭ 0. A closed-loop control system, using a dc servomotor supported by a commercial fuzzy logic controller ͑ NLX230, RS catalogue ͒ , is used to measure and control the stand-off distance. The fuzzy logic controller is utilized to compensate for different material types, which greatly affects the surface and reflectivity function and achieve normalization of the output. The bandwidth of the signal is controlled by the detector response time, which is 15 ns. The effectiveness of laser cutting processes depends on many parameters such as: laser power, beam feed rate, gas type and pressure, and the focal height offset stand-off distance. Stand-off distance is the distance between the cutting nozzle and workpiece surface. This controls the position of the beam focus relative to the surface. If the stand-off distance can be accurately measured, the processing beam can be controlled to accurately follow a workpiece profile. In the laser cutting process, the cut quality is mainly assessed by the kerf side wall surface roughness, which is characterized by the surface striation frequency, dross attachment, and the extent of the heat-affected zone. A coaxial type nozzle head with gas jet was used for the cutting process. Figure 4 illustrates the system setup. The laser beam is focused onto the workpiece through a convex- planar lens; the focal point of the lens is positioned 1.0 mm beneath the nozzle tip. The cutting nozzle is mounted on an IBM 7540 SCARA Robot. The robot is moving in the work- space via three axes X , Y , and Z . The optical sensor is fixed directly to the nozzle and connected to the servo control system to drive the robot end effector in the Z axis direction. A program written using the AML/E language is used to drive the robot and follow the cut trajectory ͑ X, Y ͒ . Focus control occurs just ahead of the laser cut or weld. The robot control program ensures that the measurement zone is kept in the optimum position depending on the type of process. An optical fiber is used to deliver the laser beam from the Nd 3 ϩ -YAG laser to the cutting nozzle. The laser is a Lumon- ics model MS 830 and generates a beam power of 400 W. The laser energy is available in the form of a pulsed beam of monochromatic and highly directional light at a wavelength of 1064 nm. Two main series of experiments were conducted using mild steel ͑ Fig. 5 ͒ and stainless ͑ Fig. 6 ͒ . Experiments were for stand-off distance ͑ 0.0 mm ͒ , ͑ 0.5, 1.0, and 1.5 mm ͒ above the workpiece surface and ͑ 0.5, 1.0 mm ͒ beneath the surface. Figures 5 and 6 illustrate the variation of the cut quality for different stand-off distances for the stainless steel and mild steel, respectively. From the figures it can be concluded that the kerf surface roughness increases when the nozzle offset ͑ stand-off ͒ distance is increased or decreased above or below the optimum zero value. This is because the spot size on the workpiece increases, above or below the focus, which results in a lower irradiance. The surface roughness for the different kerfs were measured and plotted with respect to the variation in the stand-off distance, Fig. 7. It can be seen that the optimum cut quality was achieved when the displacement from the focus was zero. The Taylor–Hobson, Talysurf 10 which give us a meter indication of the roughness average on a dual scale ͑ metric/British units ͒ meter is used as surface roughness measuring instrument. The system was calibrated experimentally by measuring surface light variations with the stand-off controller and comparing this output with that from Taylor–Hobson, Talysurf 10 surface roughness measurement device. There is no trade-off between resolution and response time. However, at 1 ␮ m, the measuring range is Ϯ 5 mm. An on-axis noncontact sensor capable of accurately sensing and measuring the workpiece surface position in real time has been developed. The autofocus system is located ahead of the processing beam and not coaxial, the measurement position is controlled to be in the optimum position via the robot controller. The output of the sensor is a voltage signal, which represents the axial displacement of the stand- off distance between the robot end effector and a workpiece surface. Changing the size of the pinhole, type and quality of the collimated beam splitter, lens size, and the angle of the workpiece surface has an effect on the characteristics of the sensor output. As the pinhole diameter is decreased, the sensitivity of the sensor increases; this, however, leads to a noisy signal due to speckle. The optimum size for a particular design is arrived at by compromising between sensitivity, output signal truncation, and noise effects. Using a smaller main lens and bigger image lens gives a more sensitive output. Changing the angle of incidence on the workpiece by a few degrees has little effect on the sensor output. Curve No. 1 on Fig. 8 was plotted for an angle of incidence on the workpiece of zero degrees; this provided a reference signal. Curve No. 2 shows the output signal after the workpiece was rotated clockwise through 5° from the horizontal feedrate vector, while curve No. 3 was produced by a counter clockwise rotation of 15° from the horizontal feedrate vector. The shifting position of the output curves occurred because the rotational axis ͑ offset ͒ of the workpiece was not coincident with the focused spot on the workpiece surface. From these results it can be seen that sensor output is largely indepen- dent of the angle of the incident beam. An on-axis high-bandwidth noncontact optical sensor capable of accurately sensing and measuring the workpiece surface position in real time has been presented. The sensor utilizes a low power diode laser beam and integrated closed- loop control system for materials processing. The device maintains the robot end effector at a specified stand-off distance from a workpiece. In material processing, most workpiece surfaces are matt and hence generate a diffuse reflection, which destroys the Gaussian intensity distribution of the laser beam upon reflection; this explains why the system still works well when the workpiece is set at an angle. The system gives an extremely useful linear response region which has an accuracy resolution of 1 ␮ m; it achieved a lateral resolution for 20.5 ␮ m. The control of stand-off distance is important to achieve good quality machining with laser ...

Context 3

... A , B , C , and D are signals from individual elements of the quadrant photodiode, Fig. 3. A differential amplifier is then used to condition the output signal as shown in Fig. 3. In Fig. 3 ͑ a ͒ , the elliptical irradiance pattern produces the strongest signal in quadrants A and B , hence ͑ the workpiece is above the focal point ͒ a positive voltage is generated ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭϩ v e . In Fig. 3 ͑ c ͒ when the elliptical irradiance pattern is strik- ing mainly quadrants C and D ͑ the workpiece surface is below the focus ͒ , a negative voltage is generated. ͓͑ A ϩ B ͒ Ϫ ͑ C ϩ D ͔͒ ϭϪ v e . When the beam waist coincides with the workpiece surface, Fig. 3 ͑ b ͒ , Z ϭ 0, the system is in focus. In this case, the light collected by the detector produces a zero output voltage from the differential amplifier ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭ 0. A closed-loop control system, using a dc servomotor supported by a commercial fuzzy logic controller ͑ NLX230, RS catalogue ͒ , is used to measure and control the stand-off distance. The fuzzy logic controller is utilized to compensate for different material types, which greatly affects the surface and reflectivity function and achieve normalization of the output. The bandwidth of the signal is controlled by the detector response time, which is 15 ns. The effectiveness of laser cutting processes depends on many parameters such as: laser power, beam feed rate, gas type and pressure, and the focal height offset stand-off distance. Stand-off distance is the distance between the cutting nozzle and workpiece surface. This controls the position of the beam focus relative to the surface. If the stand-off distance can be accurately measured, the processing beam can be controlled to accurately follow a workpiece profile. In the laser cutting process, the cut quality is mainly assessed by the kerf side wall surface roughness, which is characterized by the surface striation frequency, dross attachment, and the extent of the heat-affected zone. A coaxial type nozzle head with gas jet was used for the cutting process. Figure 4 illustrates the system setup. The laser beam is focused onto the workpiece through a convex- planar lens; the focal point of the lens is positioned 1.0 mm beneath the nozzle tip. The cutting nozzle is mounted on an IBM 7540 SCARA Robot. The robot is moving in the work- space via three axes X , Y , and Z . The optical sensor is fixed directly to the nozzle and connected to the servo control system to drive the robot end effector in the Z axis direction. A program written using the AML/E language is used to drive the robot and follow the cut trajectory ͑ X, Y ͒ . Focus control occurs just ahead of the laser cut or weld. The robot control program ensures that the measurement zone is kept in the optimum position depending on the type of process. An optical fiber is used to deliver the laser beam from the Nd 3 ϩ -YAG laser to the cutting nozzle. The laser is a Lumon- ics model MS 830 and generates a beam power of 400 W. The laser energy is available in the form of a pulsed beam of monochromatic and highly directional light at a wavelength of 1064 nm. Two main series of experiments were conducted using mild steel ͑ Fig. 5 ͒ and stainless ͑ Fig. 6 ͒ . Experiments were for stand-off distance ͑ 0.0 mm ͒ , ͑ 0.5, 1.0, and 1.5 mm ͒ above the workpiece surface and ͑ 0.5, 1.0 mm ͒ beneath the surface. Figures 5 and 6 illustrate the variation of the cut quality for different stand-off distances for the stainless steel and mild steel, respectively. From the figures it can be concluded that the kerf surface roughness increases when the nozzle offset ͑ stand-off ͒ distance is increased or decreased above or below the optimum zero value. This is because the spot size on the workpiece increases, above or below the focus, which results in a lower irradiance. The surface roughness for the different kerfs were measured and plotted with respect to the variation in the stand-off distance, Fig. 7. It can be seen that the optimum cut quality was achieved when the displacement from the focus was zero. The Taylor–Hobson, Talysurf 10 which give us a meter indication of the roughness average on a dual scale ͑ metric/British units ͒ meter is used as surface roughness measuring instrument. The system was calibrated experimentally by measuring surface light variations with the stand-off controller and comparing this output with that from Taylor–Hobson, Talysurf 10 surface roughness measurement device. There is no trade-off between resolution and response time. However, at 1 ␮ m, the measuring range is Ϯ 5 mm. An on-axis noncontact sensor capable of accurately sensing and measuring the workpiece surface position in real time has been developed. The autofocus system is located ahead of the processing beam and not coaxial, the measurement position is controlled to be in the optimum position via the robot controller. The output of the sensor is a voltage signal, which represents the axial displacement of the stand- off distance between the robot end effector and a workpiece surface. Changing the size of the pinhole, type and quality of the collimated beam splitter, lens size, and the angle of the workpiece surface has an effect on the characteristics of the sensor output. As the pinhole diameter is decreased, the sensitivity of the sensor increases; this, however, leads to a noisy signal due to speckle. The optimum size for a particular design is arrived at by compromising between sensitivity, output signal truncation, and noise effects. Using a smaller main lens and bigger image lens gives a more sensitive output. Changing the angle of incidence on the workpiece by a few degrees has little effect on the sensor output. Curve No. 1 on Fig. 8 was plotted for an angle of incidence on the workpiece of zero degrees; this provided a reference signal. Curve No. 2 shows the output signal after the workpiece was rotated clockwise through 5° from the horizontal feedrate vector, while curve No. 3 was produced by a counter clockwise rotation of 15° from the horizontal feedrate vector. The shifting position of the output curves occurred because the rotational axis ͑ offset ͒ of the workpiece was not coincident with the focused spot on the workpiece surface. From these results it can be seen that sensor output is largely indepen- dent of the angle of the incident beam. An on-axis high-bandwidth noncontact optical sensor capable of accurately sensing and measuring the workpiece surface position in real time has been presented. The sensor utilizes a low power diode laser beam and integrated closed- loop control system for materials processing. The device maintains the robot end effector at a specified stand-off distance from a workpiece. In material processing, most workpiece surfaces are matt and hence generate a diffuse reflection, which destroys the Gaussian intensity distribution of the laser beam upon reflection; this explains why the system still works well when the workpiece is set at an angle. The system gives an extremely useful linear response region which has an accuracy resolution of 1 ␮ m; it achieved a lateral resolution for 20.5 ␮ m. The control of stand-off distance is important to achieve good quality machining with laser ...

Context 4

... A , B , C , and D are signals from individual elements of the quadrant photodiode, Fig. 3. A differential amplifier is then used to condition the output signal as shown in Fig. 3. In Fig. 3 ͑ a ͒ , the elliptical irradiance pattern produces the strongest signal in quadrants A and B , hence ͑ the workpiece is above the focal point ͒ a positive voltage is generated ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭϩ v e . In Fig. 3 ͑ c ͒ when the elliptical irradiance pattern is strik- ing mainly quadrants C and D ͑ the workpiece surface is below the focus ͒ , a negative voltage is generated. ͓͑ A ϩ B ͒ Ϫ ͑ C ϩ D ͔͒ ϭϪ v e . When the beam waist coincides with the workpiece surface, Fig. 3 ͑ b ͒ , Z ϭ 0, the system is in focus. In this case, the light collected by the detector produces a zero output voltage from the differential amplifier ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭ 0. A closed-loop control system, using a dc servomotor supported by a commercial fuzzy logic controller ͑ NLX230, RS catalogue ͒ , is used to measure and control the stand-off distance. The fuzzy logic controller is utilized to compensate for different material types, which greatly affects the surface and reflectivity function and achieve normalization of the output. The bandwidth of the signal is controlled by the detector response time, which is 15 ns. The effectiveness of laser cutting processes depends on many parameters such as: laser power, beam feed rate, gas type and pressure, and the focal height offset stand-off distance. Stand-off distance is the distance between the cutting nozzle and workpiece surface. This controls the position of the beam focus relative to the surface. If the stand-off distance can be accurately measured, the processing beam can be controlled to accurately follow a workpiece profile. In the laser cutting process, the cut quality is mainly assessed by the kerf side wall surface roughness, which is characterized by the surface striation frequency, dross attachment, and the extent of the heat-affected zone. A coaxial type nozzle head with gas jet was used for the cutting process. Figure 4 illustrates the system setup. The laser beam is focused onto the workpiece through a convex- planar lens; the focal point of the lens is positioned 1.0 mm beneath the nozzle tip. The cutting nozzle is mounted on an IBM 7540 SCARA Robot. The robot is moving in the work- space via three axes X , Y , and Z . The optical sensor is fixed directly to the nozzle and connected to the servo control system to drive the robot end effector in the Z axis direction. A program written using the AML/E language is used to drive the robot and follow the cut trajectory ͑ X, Y ͒ . Focus control occurs just ahead of the laser cut or weld. The robot control program ensures that the measurement zone is kept in the optimum position depending on the type of process. An optical fiber is used to deliver the laser beam from the Nd 3 ϩ -YAG laser to the cutting nozzle. The laser is a Lumon- ics model MS 830 and generates a beam power of 400 W. The laser energy is available in the form of a pulsed beam of monochromatic and highly directional light at a wavelength of 1064 nm. Two main series of experiments were conducted using mild steel ͑ Fig. 5 ͒ and stainless ͑ Fig. 6 ͒ . Experiments were for stand-off distance ͑ 0.0 mm ͒ , ͑ 0.5, 1.0, and 1.5 mm ͒ above the workpiece surface and ͑ 0.5, 1.0 mm ͒ beneath the surface. Figures 5 and 6 illustrate the variation of the cut quality for different stand-off distances for the stainless steel and mild steel, respectively. From the figures it can be concluded that the kerf surface roughness increases when the nozzle offset ͑ stand-off ͒ distance is increased or decreased above or below the optimum zero value. This is because the spot size on the workpiece increases, above or below the focus, which results in a lower irradiance. The surface roughness for the different kerfs were measured and plotted with respect to the variation in the stand-off distance, Fig. 7. It can be seen that the optimum cut quality was achieved when the displacement from the focus was zero. The Taylor–Hobson, Talysurf 10 which give us a meter indication of the roughness average on a dual scale ͑ metric/British units ͒ meter is used as surface roughness measuring instrument. The system was calibrated experimentally by measuring surface light variations with the stand-off controller and comparing this output with that from Taylor–Hobson, Talysurf 10 surface roughness measurement device. There is no trade-off between resolution and response time. However, at 1 ␮ m, the measuring range is Ϯ 5 mm. An on-axis noncontact sensor capable of accurately sensing and measuring the workpiece surface position in real time has been developed. The autofocus system is located ahead of the processing beam and not coaxial, the measurement position is controlled to be in the optimum position via the robot controller. The output of the sensor is a voltage signal, which represents the axial displacement of the stand- off distance between the robot end effector and a workpiece surface. Changing the size of the pinhole, type and quality of the collimated beam splitter, lens size, and the angle of the workpiece surface has an effect on the characteristics of the sensor output. As the pinhole diameter is decreased, the sensitivity of the sensor increases; this, however, leads to a noisy signal due to speckle. The optimum size for a particular design is arrived at by compromising between sensitivity, output signal truncation, and noise effects. Using a smaller main lens and bigger image lens gives a more sensitive output. Changing the angle of incidence on the workpiece by a few degrees has little effect on the sensor output. Curve No. 1 on Fig. 8 was plotted for an angle of incidence on the workpiece of zero degrees; this provided a reference signal. Curve No. 2 shows the output signal after the workpiece was rotated clockwise through 5° from the horizontal feedrate vector, while curve No. 3 was produced by a counter clockwise rotation of 15° from the horizontal feedrate vector. The shifting position of the output curves occurred because the rotational axis ͑ offset ͒ of the workpiece was not coincident with the focused spot on the workpiece surface. From these results it can be seen that sensor output is largely indepen- dent of the angle of the incident beam. An on-axis high-bandwidth noncontact optical sensor capable of accurately sensing and measuring the workpiece surface position in real time has been presented. The sensor utilizes a low power diode laser beam and integrated closed- loop control system for materials processing. The device maintains the robot end effector at a specified stand-off distance from a workpiece. In material processing, most workpiece surfaces are matt and hence generate a diffuse reflection, which destroys the Gaussian intensity distribution of the laser beam upon reflection; this explains why the system still works well when the workpiece is set at an angle. The system gives an extremely useful linear response region which has an accuracy resolution of 1 ␮ m; it achieved a lateral resolution for 20.5 ␮ m. The control of stand-off distance is important to achieve good quality machining with laser ...

Context 5

... A , B , C , and D are signals from individual elements of the quadrant photodiode, Fig. 3. A differential amplifier is then used to condition the output signal as shown in Fig. 3. In Fig. 3 ͑ a ͒ , the elliptical irradiance pattern produces the strongest signal in quadrants A and B , hence ͑ the workpiece is above the focal point ͒ a positive voltage is generated ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭϩ v e . In Fig. 3 ͑ c ͒ when the elliptical irradiance pattern is strik- ing mainly quadrants C and D ͑ the workpiece surface is below the focus ͒ , a negative voltage is generated. ͓͑ A ϩ B ͒ Ϫ ͑ C ϩ D ͔͒ ϭϪ v e . When the beam waist coincides with the workpiece surface, Fig. 3 ͑ b ͒ , Z ϭ 0, the system is in focus. In this case, the light collected by the detector produces a zero output voltage from the differential amplifier ͓ ( A ϩ B ) Ϫ ( C ϩ D ) ͔ ϭ 0. A closed-loop control system, using a dc servomotor supported by a commercial fuzzy logic controller ͑ NLX230, RS catalogue ͒ , is used to measure and control the stand-off distance. The fuzzy logic controller is utilized to compensate for different material types, which greatly affects the surface and reflectivity function and achieve normalization of the output. The bandwidth of the signal is controlled by the detector response time, which is 15 ns. The effectiveness of laser cutting processes depends on many parameters such as: laser power, beam feed rate, gas type and pressure, and the focal height offset stand-off distance. Stand-off distance is the distance between the cutting nozzle and workpiece surface. This controls the position of the beam focus relative to the surface. If the stand-off distance can be accurately measured, the processing beam can be controlled to accurately follow a workpiece profile. In the laser cutting process, the cut quality is mainly assessed by the kerf side wall surface roughness, which is characterized by the surface striation frequency, dross attachment, and the extent of the heat-affected zone. A coaxial type nozzle head with gas jet was used for the cutting process. Figure 4 illustrates the system setup. The laser beam is focused onto the workpiece through a convex- planar lens; the focal point of the lens is positioned 1.0 mm beneath the nozzle tip. The cutting nozzle is mounted on an IBM 7540 SCARA Robot. The robot is moving in the work- space via three axes X , Y , and Z . The optical sensor is fixed directly to the nozzle and connected to the servo control system to drive the robot end effector in the Z axis direction. A program written using the AML/E language is used to drive the robot and follow the cut trajectory ͑ X, Y ͒ . Focus control occurs just ahead of the laser cut or weld. The robot control program ensures that the measurement zone is kept in the optimum position depending on the type of process. An optical fiber is used to deliver the laser beam from the Nd 3 ϩ -YAG laser to the cutting nozzle. The laser is a Lumon- ics model MS 830 and generates a beam power of 400 W. The laser energy is available in the form of a pulsed beam of monochromatic and highly directional light at a wavelength of 1064 nm. Two main series of experiments were conducted using mild steel ͑ Fig. 5 ͒ and stainless ͑ Fig. 6 ͒ . Experiments were for stand-off distance ͑ 0.0 mm ͒ , ͑ 0.5, 1.0, and 1.5 mm ͒ above the workpiece surface and ͑ 0.5, 1.0 mm ͒ beneath the surface. Figures 5 and 6 illustrate the variation of the cut quality for different stand-off distances for the stainless steel and mild steel, respectively. From the figures it can be concluded that the kerf surface roughness increases when the nozzle offset ͑ stand-off ͒ distance is increased or decreased above or below the optimum zero value. This is because the spot size on the workpiece increases, above or below the focus, which results in a lower irradiance. The surface roughness for the different kerfs were measured and plotted with respect to the variation in the stand-off distance, Fig. 7. It can be seen that the optimum cut quality was achieved when the displacement from the focus was zero. The Taylor–Hobson, Talysurf 10 which give us a meter indication of the roughness average on a dual scale ͑ metric/British units ͒ meter is used as surface roughness measuring instrument. The system was calibrated experimentally by measuring surface light variations with the stand-off controller and comparing this output with that from Taylor–Hobson, Talysurf 10 surface roughness measurement device. There is no trade-off between resolution and response time. However, at 1 ␮ m, the measuring range is Ϯ 5 mm. An on-axis noncontact sensor capable of accurately sensing and measuring the workpiece surface position in real time has been developed. The autofocus system is located ahead of the processing beam and not coaxial, the measurement position is controlled to be in the optimum position via the robot controller. The output of the sensor is a voltage signal, which represents the axial displacement of the stand- off distance between the robot end effector and a workpiece surface. Changing the size of the pinhole, type and quality of the collimated beam splitter, lens size, and the angle of the workpiece surface has an effect on the characteristics of the sensor output. As the pinhole diameter is decreased, the sensitivity of the sensor increases; this, however, leads to a noisy signal due to speckle. The optimum size for a particular design is arrived at by compromising between sensitivity, output signal truncation, and noise effects. Using a smaller main lens and bigger image lens gives a more sensitive output. Changing the angle of incidence on the workpiece by a few degrees has little effect on the sensor output. Curve No. 1 on Fig. 8 was plotted for an angle of incidence on the workpiece of zero degrees; this provided a reference signal. Curve No. 2 shows the output signal after the workpiece was rotated clockwise through 5° from the horizontal feedrate vector, while curve No. 3 was produced by a counter clockwise rotation of 15° from the horizontal feedrate vector. The shifting position of the output curves occurred because the rotational axis ͑ offset ͒ of the workpiece was not coincident with the focused spot on the workpiece surface. From these results it can be seen that sensor output is largely indepen- dent of the angle of the incident beam. An on-axis high-bandwidth noncontact optical sensor capable of accurately sensing and measuring the workpiece surface position in real time has been presented. The sensor utilizes a low power diode laser beam and integrated closed- loop control system for materials processing. The device maintains the robot end effector at a specified stand-off distance from a workpiece. In material processing, most workpiece surfaces are matt and hence generate a diffuse reflection, which destroys the Gaussian intensity distribution of the laser beam upon reflection; this explains why the system still works well when the workpiece is set at an angle. The system gives an extremely useful linear response region which has an accuracy resolution of 1 ␮ m; it achieved a lateral resolution for 20.5 ␮ m. The control of stand-off distance is important to achieve good quality machining with laser ...

Context 6

... manufacturing applications continue to make in- creasing use of robots to carry the laser beam for cutting, welding, and surface engineering processes. The quality of such processes is greatly influenced by the precision with which the laser can be translated across a workpiece surface; 1 this article reports on a sensor that has been suc- cessfully integrated into a robot control system so as to give the precise stand-off control required. The drive towards au- tomation has the following advantages: improved product quality, reduced labor and cost of production, increased pro- ductivity and accuracy, better working conditions, and greater production reliability. Unless specifically designed for laser processing applications, many robots produce a jerking incremental trajectory and usually suffer from position dependent displacement of the end effector. 2–7 With the addition of a focal height servomechanism the high-bandwidth optical range sensor described herein allows the use of such robots for fiber manipu- lation and hence facilitates high quality laser cutting and welding operations using equipment that may already exist. A focus control system for Nd:YAG laser welding system based on an optical sensor incorporated into the fiber delivery system to detect light generated by the process was de- tailed in Refs. 8 and 9. Radiation from a low power laser diode is focused onto the workpiece surface; light reflected from the surface is collected through a main lens and directed into an imaging lens ͑ astigmatic lens ͒ , 10 which focuses the signal onto a quadrant photodiode. The irradiance of the laser beam is detected by a photodiode, a differential amplifier is used to generate an output signal called focus error signal ͑ FES ͒ ; this determines the magnitude and direction of the required robot end- effector displacement. The system measures the stand-off distance in real time. The system reported herein has far greater accuracy than capacitive or inductive based systems and there is no restric- tion on the type of materials on which it can be used. Fur- thermore, its performance is not degraded near workpiece edges or holes. The system is more accurate and compact than laser line projection or triangulation systems. Mechanical ride on system can damage workpiece surfaces and have a slow response. 11–15 The sensor researched in this work is a noncontact ex- ternal sensor. It is a high-bandwidth optical range sensor for local stand-off control of robotically manipulated optical fiber laser beam delivery systems. The system facilitates a measuring range of Ϯ 5 mm of focus error; its effectiveness in controlling the stand-off distance for laser materials processing applications is assessed herein. The work involved the design of a noncontact device utilizing a laser beam and integrated closed-loop control system, which maintains the robot end-effector at a specified stand-off distance from a workpiece surface. The working principle of the sensor is shown in Fig. 1. A collimated laser beam, which passes through a polarizing beam splitter, is focused onto the workpiece surface by the main lens to form a new beam waist. 16–18 The light is reflected from the workpiece surface, collected through the main lens, and redirected into an astigmatic image lens through the polarizing beam splitter, to focus on quadrant photodiode through a pinhole. The pinhole is incorporated to reduce noise from ambient light and laser light resulting from back reflect and from optical surface and speckle noise. The position of the pinhole was determined experimentally and greatly improved the performance of the system. 19 The effect of a small longitu- dinal displacement of the workpiece in the direction Z , par- allel to the beam axis, will substantially alter the magnitude and shape of the irradiance arriving at the plane of the photodiode, where Z is offset distance of the workpiece from the sensor, Fig. 1. As a result, the power detected by the four individual cells of the photodiode will vary. A differential amplifier is then used to condition the output signal and hence determine the magnitude and direction of displacement, Fig. 1. The output signal is called the FES. A special filter is utilized in front of the detectors to allow the probe beam to pass through. By this means, the probe beam light can be separated from the separated light emitted by the laser manufacturing process. The system is similar to a CD player focus control system but the optical design facilities are for greater control range ͑ i.e., Ϯ 5 mm ͒ . Focus error signal is related to the axial distance from the active medium to the focal point. In this work, the astigmatic technique is used to generate the FES. A typical FES is shown in Fig. 2. Point A represents the workpiece located at the system focus. Region B is the range over which the FES is linear; the magnitude of its slope controls the system’s sensitivity. Region C is termed the lock-on range and repre- sents the symmetric region of space surrounding the desired workpiece position ͑ best focus location A ͒ . It can be seen in Fig. 2 that the relationship between the distance and the output voltage is nonlinear, except for a portion between the positive and negative peaks, region B. This linear range is determined by the size and position of lenses, pinhole, and photodiode. The important parameters that quantify the performance of the sensor and characterize the FES are as follows: slope through the best focus, symmetric lock-on range about the best focus, linearity within the measuring range and focus offset. These parameters will affect the control system of the sensor, which determines the best position of the workpiece. As a result of previous experiments on the sensor prototype, it was found to be advan- tageous to have high gain, a large symmetric lock-on range, a high degree of linearity, and no focus offset. Performance parameters are determined by the optical system, design, detector geometry, and working environment. The FES is entirely controlled by the characteristics of the returning beam. If the workpiece is in the required position the returned collimated laser beam will form a uniform circle on the quadrant photodiode, see Fig. 3. If, however, the workpiece surface is too far above or below the required position, the reflected beam will be more convergent. There- fore, if the point of focus lies above the workpiece surface, the beam on the detector is either a vertical elliptical beam or a horizontal elliptical beam, see Fig. 3. The focus error signal is simply given by the normalized output of the detector, which ...

Context 7

... manufacturing applications continue to make in- creasing use of robots to carry the laser beam for cutting, welding, and surface engineering processes. The quality of such processes is greatly influenced by the precision with which the laser can be translated across a workpiece surface; 1 this article reports on a sensor that has been suc- cessfully integrated into a robot control system so as to give the precise stand-off control required. The drive towards au- tomation has the following advantages: improved product quality, reduced labor and cost of production, increased pro- ductivity and accuracy, better working conditions, and greater production reliability. Unless specifically designed for laser processing applications, many robots produce a jerking incremental trajectory and usually suffer from position dependent displacement of the end effector. 2–7 With the addition of a focal height servomechanism the high-bandwidth optical range sensor described herein allows the use of such robots for fiber manipu- lation and hence facilitates high quality laser cutting and welding operations using equipment that may already exist. A focus control system for Nd:YAG laser welding system based on an optical sensor incorporated into the fiber delivery system to detect light generated by the process was de- tailed in Refs. 8 and 9. Radiation from a low power laser diode is focused onto the workpiece surface; light reflected from the surface is collected through a main lens and directed into an imaging lens ͑ astigmatic lens ͒ , 10 which focuses the signal onto a quadrant photodiode. The irradiance of the laser beam is detected by a photodiode, a differential amplifier is used to generate an output signal called focus error signal ͑ FES ͒ ; this determines the magnitude and direction of the required robot end- effector displacement. The system measures the stand-off distance in real time. The system reported herein has far greater accuracy than capacitive or inductive based systems and there is no restric- tion on the type of materials on which it can be used. Fur- thermore, its performance is not degraded near workpiece edges or holes. The system is more accurate and compact than laser line projection or triangulation systems. Mechanical ride on system can damage workpiece surfaces and have a slow response. 11–15 The sensor researched in this work is a noncontact ex- ternal sensor. It is a high-bandwidth optical range sensor for local stand-off control of robotically manipulated optical fiber laser beam delivery systems. The system facilitates a measuring range of Ϯ 5 mm of focus error; its effectiveness in controlling the stand-off distance for laser materials processing applications is assessed herein. The work involved the design of a noncontact device utilizing a laser beam and integrated closed-loop control system, which maintains the robot end-effector at a specified stand-off distance from a workpiece surface. The working principle of the sensor is shown in Fig. 1. A collimated laser beam, which passes through a polarizing beam splitter, is focused onto the workpiece surface by the main lens to form a new beam waist. 16–18 The light is reflected from the workpiece surface, collected through the main lens, and redirected into an astigmatic image lens through the polarizing beam splitter, to focus on quadrant photodiode through a pinhole. The pinhole is incorporated to reduce noise from ambient light and laser light resulting from back reflect and from optical surface and speckle noise. The position of the pinhole was determined experimentally and greatly improved the performance of the system. 19 The effect of a small longitu- dinal displacement of the workpiece in the direction Z , par- allel to the beam axis, will substantially alter the magnitude and shape of the irradiance arriving at the plane of the photodiode, where Z is offset distance of the workpiece from the sensor, Fig. 1. As a result, the power detected by the four individual cells of the photodiode will vary. A differential amplifier is then used to condition the output signal and hence determine the magnitude and direction of displacement, Fig. 1. The output signal is called the FES. A special filter is utilized in front of the detectors to allow the probe beam to pass through. By this means, the probe beam light can be separated from the separated light emitted by the laser manufacturing process. The system is similar to a CD player focus control system but the optical design facilities are for greater control range ͑ i.e., Ϯ 5 mm ͒ . Focus error signal is related to the axial distance from the active medium to the focal point. In this work, the astigmatic technique is used to generate the FES. A typical FES is shown in Fig. 2. Point A represents the workpiece located at the system focus. Region B is the range over which the FES is linear; the magnitude of its slope controls the system’s sensitivity. Region C is termed the lock-on range and repre- sents the symmetric region of space surrounding the desired workpiece position ͑ best focus location A ͒ . It can be seen in Fig. 2 that the relationship between the distance and the output voltage is nonlinear, except for a portion between the positive and negative peaks, region B. This linear range is determined by the size and position of lenses, pinhole, and photodiode. The important parameters that quantify the performance of the sensor and characterize the FES are as follows: slope through the best focus, symmetric lock-on range about the best focus, linearity within the measuring range and focus offset. These parameters will affect the control system of the sensor, which determines the best position of the workpiece. As a result of previous experiments on the sensor prototype, it was found to be advan- tageous to have high gain, a large symmetric lock-on range, a high degree of linearity, and no focus offset. Performance parameters are determined by the optical system, design, detector geometry, and working environment. The FES is entirely controlled by the characteristics of the returning beam. If the workpiece is in the required position the returned collimated laser beam will form a uniform circle on the quadrant photodiode, see Fig. 3. If, however, the workpiece surface is too far above or below the required position, the reflected beam will be more convergent. There- fore, if the point of focus lies above the workpiece surface, the beam on the detector is either a vertical elliptical beam or a horizontal elliptical beam, see Fig. 3. The focus error signal is simply given by the normalized output of the detector, which ...