Laser weld quality monitoring method and system

• US 20020144984A1
• Filed: 02/25/2002
• Published: 10/10/2002
• Est. Priority Date: 02/23/2001
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Motor Co., Ltd.
• Inventors: Mori, Kiyokazu, Tarui, Taishi, Takemura, Shinsuke
• Granted Patent: US 6,710,283 B2
• Time in Patent Office: Days
• US Class Current: 219/121.64
• CPC Class Codes: B23K 2101/006 Vehicles
• CPC Class Codes: B23K 2101/35 Surface treated articles
• CPC Class Codes: B23K 2103/04 Steel or steel alloys
• CPC Class Codes: B23K 2103/08 Non-ferrous metals or alloys
• CPC Class Codes: B23K 2103/50 Inorganic material, e.g. metals, not provided for in B23K2103/02 – B23K2103/26
• CPC Class Codes: B23K 26/03 Observing, e.g. monitoring, the workpiece
• CPC Class Codes: B23K 26/244 Overlap seam welding
• CPC Class Codes: B23K 26/32 taking account of the properties of the material involved
• CPC Class Codes: B23K 31/125 Weld quality monitoring

Abstract

Infrared reflection of a laser beam (LB3) irradiated on a welding part (WP) of work (5) is detected by a sensor (6a) high of elevation angles where it is converted into an electrical signal, which is processed by a measuring circuit (MC) to be input to a quality monitor (QM), where it is stored as a data in a memory (7g), which data is processed by way of a spectral analysis, which calculates a spectral distribution of electrical signal, and a signal power sum in a particular frequency band, to be compared with a threshold value for decision on occurrence of a significant porous state.

Claims

1. A laser weld quality monitoring method comprising:

welding a part of work with a laser beam irradiated thereon from a YAG laser;

detecting a varying intensity of light from the welding part to provide a detection signal;

determining a value of signal power of a frequency spectrum in a specified frequency band of the detection signal; and

making a decision for a porous state of the welding part

to be significant as the value of signal power exceeds a threshold of weld quality, and

to be insignificant as the value of signal power does not exceed the threshold of weld quality.

2. A laser weld quality monitoring method according to claim 1, wherein the detection signal comprises a varying electrical signal representing the varying intensity of the light from the welding part, and the determining the value of signal power comprises calculating a set of frequency spectra of the varying electrical signal.

3. A laser weld quality monitoring method according to claim 1, wherein the specified frequency band is varied depending on one of a thickness of the work, a welding speed, and an aspect ratio of a keyhole at the welding part.

4. A laser weld quality monitoring method according to claim 1, wherein the determining the value of signal power comprises one of passing the electrical signal to a band-pass filter and applying a Fourrier transform to data of the electrical signal.

5. A laser weld quality monitoring method comprising:

irradiating a laser beam from a YAG laser to a welding part of work;

detecting light reflected from the welding part;

calculating a frequency distribution from a set of data of the detected light within a interval of time;

calculating, from the frequency distribution, a first signal power sum in one of a first frequency band for detecting an under-filled state and a second frequency band for detecting a porous state, and a second signal power sum in a third frequency band for detecting a non-welded state;

mapping a combination of calculated values of the first and second signal power sums, in a region defined by a combination of a first axis representing the first signal power sum and a second axis representing the second signal power sum, including a sub-region representing a non-conforming state as one of the under-filled state, the porous state, and the non-welded stare; and

making a decision for the welding part to have the non-conforming state, as the combination of calculated values is mapped in the sub-region.

6. A laser weld quality monitoring method according to claim 5, wherein the calculating the frequency distribution comprises converting the detected light into an electrical signal, storing data on time-dependant variations of the electrical signal, and calculating the frequency distribution from the stored data.

7. A laser weld quality monitoring method according to claim 5, wherein the region includes sub-regions representing the under-filled state, the porous state, and the non-welded state, respectively,

8. A laser weld quality monitoring method according to claim 5, wherein the region includes a sub-region representing a conforming state of the work.

9. A laser weld quality monitoring method according to claim 5, wherein the region includes a sub-region representative of at least tow of the under-filled state, the porous state, and the non-welded state.

10. A laser weld quality monitoring method according to claim 5, further comprising:

calculating, from a subset of the set of data, a subsidiary frequency distribution of the detected light within a sub-interval of the interval of time;

calculating, from the subsidiary frequency distribution, a first subsidiary signal power sum in one of a first subsidiary frequency band for detecting an under-filled state in a sub-section of the welding part corresponding to the sub-interval and a second subsidiary frequency band for detecting a porous state in the sub-section, and a second subsidiary signal power sum in a third subsidiary frequency band for detecting a non-welded state in the sub-section;

mapping in the region a combination of calculated subsidiary values of the first and second subsidiary signal power sums;

making a decision for the sub-section of the welding part to have the non-conforming state, as the combination of calculated subsidiary values is mapped in the sub-region; and

concluding a weld quality of the welding part based on the decision for the subsection.

11. A laser weld quality monitoring method according to claim 10, wherein the concluding the weld quality depends on a conforming proportion of the sub-section to the welding part.

12. A laser weld quality monitoring method according to claim 10, wherein one of the first, second, and third subsidiary frequency bands is varied depending on one of a thickness of the work, a welding speed, and an aspect ratio of a keyhole at the sub-section of the welding part.

13. A laser weld quality monitoring system comprising;

a welder configured to weld a part of work with a laser beam irradiated thereon from a YAG laser;

a detector configured to detect a varying intensity of light reflected from the welding part to provide a detection signal;

a value determiner configured to determine a value of signal power of a frequency spectrum in a specified frequency band of the detection signal; and

a decision-maker configured to make a decision for a porous state of the welding part

to be significant as the value of signal power exceeds a threshold of weld quality, and

to be insignificant as the value of signal power does nor exceed the threshold of weld quality.

14. A laser weld quality monitoring system comprising:

welding means for welding a part of work with a laser beam irradiated thereon from a YAG laser;

detecting means for detecting a varying intensity of light reflected from the welding part to provide a detection signal;

value determining means for determining a value of signal power of a frequency spectrum in a specified frequency band of the detection signal; and

decision-making means for making a decision for a porous state of the welding part

to be significant as the value of signal power exceeds a threshold of weld quality, and

to be insignificant us the value of signal power does not exceed the threshold of weld quality.

15. A laser weld quality monitoring system comprising:

a laser welder configured to irradiate a laser beam from a YAG laser to a welding part of work;

a detector configured to detect light reflected from the welding part;

a calculator configured to calculate a frequency distribution from a set of data of the detected light within a interval of time;

a calculator configured to calculate, from the frequency distribution, a first signal power sum in one of a first frequency band for detecting an under-filled state and a second frequency band for detecting a porous stale, and a second signal power sum in a third frequency band for detecting a non-welded state;

an operator configured to map a combination of calculated values of the first and second signal power sums, in a region defined by a combination of a first axis representing the first signal power sum and a second axis representing the second signal power sum, including a sub-region representing a non-conforming state as one of the under-filled state, the porous state, and the non-welded state; and

a decision-maker configured to make a decision for the welding part to have the non-conforming state, as the combination of calculated values is mapped in the sub-region.

16. A laser weld quality monitoring system comprising;

laser welding means for irradiating a laser beam from a YAG laser to a welding part of work;

detecting means for detecting light reflected from the welding part;

calculating means for calculating a frequency distribution from a set of data of the detected light within a interval of time;

calculating means for calculating, from the frequency distribution, a first signal power sum in one of a first frequency band for detecting an under-filled state and a second frequency band for detecting a porous state, and a second signal power sum in a third frequency band for detecting a non-welded state;

operator means for mapping a combination of calculated values of the first and second signal power sums, in a region defined by a combination of a first axis representing the first signal power sum and a second axis representing the second signal power sum, including a sub-region representing a non-conforming state as one of the under-filled state, the porous state, and the non-welded state; and

decision-making means for making a decision for the welding part to have the non-conforming state, as the combination of calculated values is mapped in the sub-region.