CN117412517B - Composite welding quality control method and system based on visual recognition - Google Patents

Composite welding quality control method and system based on visual recognition Download PDF

Info

Publication number
CN117412517B
CN117412517B CN202311718802.1A CN202311718802A CN117412517B CN 117412517 B CN117412517 B CN 117412517B CN 202311718802 A CN202311718802 A CN 202311718802A CN 117412517 B CN117412517 B CN 117412517B
Authority
CN
China
Prior art keywords
laser
welding
tin
tin ring
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311718802.1A
Other languages
Chinese (zh)
Other versions
CN117412517A (en
Inventor
赵建涛
卢财源
杨进
何文强
颜飞亮
雷宝升
唐波
覃正玖
许必坚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Zichen Laser Equipment Co ltd
Original Assignee
Shenzhen Zichen Laser Equipment Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Zichen Laser Equipment Co ltd filed Critical Shenzhen Zichen Laser Equipment Co ltd
Priority to CN202311718802.1A priority Critical patent/CN117412517B/en
Publication of CN117412517A publication Critical patent/CN117412517A/en
Application granted granted Critical
Publication of CN117412517B publication Critical patent/CN117412517B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K3/00Apparatus or processes for manufacturing printed circuits
    • H05K3/30Assembling printed circuits with electric components, e.g. with resistor
    • H05K3/32Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits
    • H05K3/34Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits by soldering
    • H05K3/3447Lead-in-hole components
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K1/00Soldering, e.g. brazing, or unsoldering
    • B23K1/005Soldering by means of radiant energy
    • B23K1/0056Soldering by means of radiant energy soldering by means of beams, e.g. lasers, E.B.
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K3/00Tools, devices, or special appurtenances for soldering, e.g. brazing, or unsoldering, not specially adapted for particular methods
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K3/00Tools, devices, or special appurtenances for soldering, e.g. brazing, or unsoldering, not specially adapted for particular methods
    • B23K3/08Auxiliary devices therefor
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K3/00Apparatus or processes for manufacturing printed circuits
    • H05K3/30Assembling printed circuits with electric components, e.g. with resistor
    • H05K3/32Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits
    • H05K3/34Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits by soldering
    • H05K3/3457Solder materials or compositions; Methods of application thereof
    • H05K3/3485Applying solder paste, slurry or powder
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K3/00Apparatus or processes for manufacturing printed circuits
    • H05K3/30Assembling printed circuits with electric components, e.g. with resistor
    • H05K3/32Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits
    • H05K3/34Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits by soldering
    • H05K3/3489Composition of fluxes; Methods of application thereof; Other methods of activating the contact surfaces
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K3/00Apparatus or processes for manufacturing printed circuits
    • H05K3/30Assembling printed circuits with electric components, e.g. with resistor
    • H05K3/32Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits
    • H05K3/34Assembling printed circuits with electric components, e.g. with resistor electrically connecting electric components or wires to printed circuits by soldering
    • H05K3/3494Heating methods for reflowing of solder
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B23MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23KSOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
    • B23K2101/00Articles made by soldering, welding or cutting
    • B23K2101/36Electric or electronic devices
    • B23K2101/42Printed circuits
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K2203/00Indexing scheme relating to apparatus or processes for manufacturing printed circuits covered by H05K3/00
    • H05K2203/04Soldering or other types of metallurgic bonding
    • H05K2203/0465Shape of solder, e.g. differing from spherical shape, different shapes due to different solder pads
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K2203/00Indexing scheme relating to apparatus or processes for manufacturing printed circuits covered by H05K3/00
    • H05K2203/10Using electric, magnetic and electromagnetic fields; Using laser light
    • H05K2203/107Using laser light
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K2203/00Indexing scheme relating to apparatus or processes for manufacturing printed circuits covered by H05K3/00
    • H05K2203/16Inspection; Monitoring; Aligning
    • HELECTRICITY
    • H05ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
    • H05KPRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
    • H05K2203/00Indexing scheme relating to apparatus or processes for manufacturing printed circuits covered by H05K3/00
    • H05K2203/16Inspection; Monitoring; Aligning
    • H05K2203/166Alignment or registration; Control of registration

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Mechanical Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Optics & Photonics (AREA)
  • Laser Beam Processing (AREA)

Abstract

The invention discloses a vision recognition-based composite welding quality control method and a vision recognition-based composite welding quality control system, which relate to the technical field of semiconductor assembly quality control and comprise the following steps: setting a tin ring box, a tin paste dispensing assembly, a visual identification positioning device, a laser welding device and a PCB heating device; positioning and taking materials from the front surface of the tin ring; positioning and shifting the back of the tin ring; positioning a bonding pad pin; collar and verification; a spot solder paste, comprising: the solder paste is dispensed to the top end of the contact pin of the electronic device by the solder paste dispensing component, and the gap between the solder ring and the contact pin is shielded by the solder paste in a plan view; visual positioning; laser welding, comprising: the laser welding device is enabled to emit laser to heat the tin ring and the solder paste, and the PCB heating device is enabled to heat the bonding pad; and (5) detecting after welding. The present application has an effect of improving the soldering quality of the semiconductor device.

Description

Composite welding quality control method and system based on visual recognition
Technical Field
The application relates to the technical field of semiconductor assembly quality control, in particular to a vision recognition-based composite welding quality control method and system.
Background
The soldering of the discrete semiconductor device to the through hole of the PCB is actually soldering the through hole of the PCB and the pin of the semiconductor device and enabling to energize. The PCB surface has pads (typically copper plated, gold plated, and pads are typically circular, with the center points of the pads coinciding with the center points of the vias, so the pads are typically ring-shaped in nature), and the semiconductor pin is typically cylindrical.
The PCB itself has a thickness, typically between 0.1 and 6mm, i.e. the depth of the through holes in the PCB is equal to the thickness of the PCB. After the semiconductor pin is inserted into the PCB through hole, the semiconductor pin protrudes from the surface of the PCB by about 1-3 mm.
The upper surface pads of the PCB are filled with tin alloy and the through holes of the PCB are filled with tin alloy. The tin alloy can fill the PCB through holes, namely can reach the lower surface of the PCB, and the tin penetration rate is 100%; if the tin alloy penetrates only half the thickness of the PCB during the downward penetration from the upper surface of the PCB, this means a tin penetration rate of 50%. The higher tin permeability means better effect, namely, stronger welding and better electrifying effect. In the welding process, the temperature of the tin alloy is reduced, the tin alloy is gradually cooled and the fluidity is reduced as the tin alloy permeates downwards, so that the problem of approaching the tin permeability to 100% is solved.
The existing soldering iron welding, wave soldering and other modes, wherein a plating layer at the tip end part of the soldering iron head exists, the plating layer can be aged continuously along with the use, and the effect is gradually deteriorated due to the aging of the plating layer, so that the consistency of the welding effect is difficult to ensure when the soldering iron head is welded in mass production; the tin permeability of the latter is difficult to control, and the tin permeability of all through holes is difficult to be consistent, namely the welding quality is difficult to ensure, so the application provides a new technical scheme.
Disclosure of Invention
In order to improve the welding quality of a semiconductor device, the application provides a composite welding quality control method and system based on visual recognition.
In a first aspect, the present application provides a vision recognition-based composite welding quality control method, which adopts the following technical scheme:
a vision recognition-based composite welding quality control method comprises the following steps:
step one, hardware composition setting, which comprises: setting a tin ring box, a tin paste dispensing assembly, a visual identification positioning device, a laser welding device and a PCB heating device; wherein, a tin ring is placed in the tin ring box, the inner diameter of the tin ring is suitable for the contact pin of the electrical device, and the outer diameter of the tin ring is larger than the diameter of the through hole of the PCB;
step two, positioning and taking materials from the front surface of the tin ring, which comprises the following steps: the visual recognition positioning device is used for carrying out image acquisition on the tin ring box, positioning the position of the tin ring based on image recognition, and absorbing/grabbing the tin ring according to the position of the tin ring;
step three, positioning and shifting the back of the tin ring, which comprises the following steps: the visual recognition positioning device is enabled to collect images from the lower side of the captured tin ring, the inner diameter characteristics of the tin ring are extracted based on image recognition, and the tin ring is moved to the upper side of a contact pin of an electronic device;
fourth, pad contact pin location, it includes: the visual recognition positioning device is enabled to conduct image acquisition on the contact pin of the electronic device, and the positions of the pins are recognized and positioned based on the images;
step five, lantern ring and verification, it includes: sleeving the tin ring on a contact pin of the electronic device, enabling the visual identification positioning device to acquire an image of the bonding pad, and identifying the posture of the tin ring based on the image;
step six, dispensing solder paste, which comprises the following steps: the solder paste is dispensed to the top end of the contact pin of the electronic device by the solder paste dispensing component, and the gap between the solder ring and the contact pin is shielded by the solder paste in a plan view;
step seven, visual positioning, which comprises the following steps: identifying and positioning based on the image acquired by the visual identification and positioning device to obtain the position of the tin ring and the position of the contact pin which are sleeved;
step eight, laser welding, which comprises: the laser welding device is enabled to emit laser to heat the tin ring and the solder paste, and the PCB heating device is enabled to heat the bonding pad;
step nine, post-welding detection, which comprises: the visual recognition positioning device is enabled to collect images of welding pads which are welded, and based on image recognition analysis, the proportion of tin spreading area to the welding pads, whether short circuits exist between adjacent welding spots, whether defects generated by laser burn exist or not and welding leakage exist.
By adopting the technical scheme, the method comprises the following steps:
1) The laser welding is adopted, so that the process is contactless, stress-free and damage-free;
2) Ensuring that no laser is injected into a through hole of the PCB and no laser is injected into a semiconductor device on the lower surface of the PCB by a mode of tin ring and tin paste, thereby avoiding burn of the PCB and the semiconductor device; meanwhile, the tin alloy amount can be accurately controlled, so that the waste of the tin alloy is avoided, the tin permeability is ensured, and the short circuit caused by excessive tin alloy is avoided;
3) The automatic tin ring taking and sleeving of the tin ring on the contact pin can be realized by combining visual identification with an automatic mechanical arm, and the welding efficiency is relatively higher;
4) The lower surface of the PCB is provided with a heating design, so that the fluidity of tin alloy can be improved, and the tin permeability is further improved;
5) The post-welding detection is performed based on the visual recognition function, so that scrapping can be reduced, and the yield is improved;
i.e., the present method can be used to improve the soldering quality of semiconductor devices.
Optionally, the identifying and positioning process of the visual identifying and positioning device comprises preliminary positioning and accurate positioning.
By adopting the technical scheme, the method has more accurate positioning of the structures such as the tin ring, the contact pin and the like.
Optionally, the visual identifying and positioning device includes a first camera unit and an annular light source, the annular light source provides a plurality of light source light emitting units with the same spacing angle, and the first camera unit collects the target images under the sequential start and the simultaneous start of the plurality of light source light emitting units.
By adopting the technical scheme, the annular light source is established and divided into a plurality of independent light emitting modules with equal intervals, objects can be respectively irradiated from a plurality of angles, images with a plurality of angles are generated, and data obtained from the images with different angles can be used for improving positioning accuracy, so that identification in production has higher accuracy and reliability.
Optionally, the visual recognition positioning device is made to recognize and analyze dirt and oxidation information on the solder ring image and the solder pad image before the solder ring is sleeved when the light emitting units of the plurality of light sources are simultaneously started, and the visual recognition positioning device is made to recognize and analyze the posture of the solder ring and the area and position distribution of the solder paste respectively on the solder pad image after the solder ring is sleeved when the light emitting units of the plurality of light sources are simultaneously started and the solder pad image after the solder paste is coated.
By adopting the technical scheme, the defects of the tin ring, the welding disc and the tin paste can be found in time in the process, the adjustment and the stopping of relevant contents can be carried out in time, and the welding quality of the produced PCB is ensured.
Optionally, the laser welding device comprises a laser generating module, an infrared temperature measuring unit, a second camera shooting unit and a man-machine interaction screen, wherein the second camera shooting unit moves along with the laser generating module, and the man-machine interaction screen is connected with the second camera shooting unit and the infrared temperature measuring unit in a data mode;
the infrared temperature measurement unit is used for infrared temperature measurement of the welding position;
step eight, laser welding, it still includes: and adjusting the light-emitting power of the laser generating module according to the temperature information fed back by the infrared temperature measuring unit, so that the temperature of the welding position accords with a preset temperature threshold.
By adopting the technical scheme, on one hand, workers can observe the welding process in real time; on the other hand, the temperature of the welding area is monitored in real time, and the light emitting power of the laser is regulated in real time, so that the welding area can be always maintained at a set temperature, the welding area can be prevented from exceeding the melting temperature of tin alloy in the welding process, and the laser energy input capable of melting the tin alloy but not being too high can be always maintained, thereby better protecting the PCB and the semiconductor device from being damaged in the laser welding and heating processes.
Optionally, the PCB heating device is a non-contact heating structure and heats the solder pad with light or heat radiation.
By adopting the technical scheme, the heating design of the lower surface of the PCB can improve the fluidity of the tin alloy and further improve the tin permeability; meanwhile, because of non-contact, the probability of burning out the PCB in a heating structure out of control can be reduced, and the PCB can be more conveniently circulated on an automatic production line.
Optionally, further comprising line anomaly prediction, comprising:
invoking a pre-established neural network model to train the electrical control parameters, the electrical operation parameters and the images of all the steps in the production process;
and importing the electrical control parameters, the electrical operation parameters and the images of all the steps in the current production process into a trained neural network model, predicting a welding result, and generating a welding quality development trend table/graph.
By adopting the technical scheme, the production data can be analyzed by utilizing the mathematical model, the abnormality of the production process can be found in advance, the fault detection and the like can be performed in time, and the welding quality is ensured.
In a second aspect, the present application provides a quality control system applied to the vision recognition-based composite welding quality control method as described in any one of the above, and the following technical scheme is adopted:
the quality control system applied to the vision-recognition-based composite welding quality control method comprises a tin ring box, a spot solder paste assembly, a vision recognition positioning device, a laser welding device and a PCB heating device.
Optionally, one of the camera units one and one of the annular light sources are a group of visual units, the visual recognition positioning device comprises a plurality of groups of visual units, and at least one group of visual units is arranged on the spot solder paste assembly; the first image pick-up unit is positioned on the central axis of the annular light source.
By adopting the technical scheme, when the annular light sources are sequentially lightened, the image acquired by the first image pickup unit has symmetrical characteristics, and the image is relatively less in deflection and distortion caused by the shooting angle, and has fewer interference factors.
Optionally, the laser welding device comprises a laser generating module, an infrared temperature measuring unit, a second camera unit and a man-machine interaction screen, wherein the man-machine interaction screen is connected with the second camera unit and the infrared temperature measuring unit in data, and the infrared temperature measuring unit is used for measuring the temperature of the welding position in an infrared mode;
the laser generation module comprises an integrated shell, a central channel, a laser channel and a monitoring channel are arranged in the integrated shell, the central channel is vertical, an infrared temperature measurement unit is arranged at the upper end of the central channel, and a laser focusing lens is arranged at the other end of the central channel; the laser channel is L-shaped, the transverse end is communicated with the upper side wall of the central channel, and the other end is upward and is provided with a laser generating module; the monitoring channel is L-shaped, the transverse end is communicated with the side wall of the lower part of the central channel, and the other end is upward and is provided with a second camera unit;
the corner of the laser channel is provided with a laser total reflection lens, the upper part of the central channel is provided with a laser 45-degree reflection lens, the laser 45-degree reflection lens is parallel to the laser total reflection lens, the laser emitted by the laser generating module falls on the laser total reflection lens and is reflected to the laser 45-degree reflection lens, and the laser 45-degree reflection lens reflects the laser and enables the laser to be emitted from the lower port of the central channel;
the corner of the monitoring channel is provided with a monitoring total reflection lens, a monitoring 45-degree reflection lens is arranged below the laser 45-degree reflection lens, the monitoring 45-degree reflection lens is parallel to the monitoring total reflection lens, and an image below the central channel is reflected to the monitoring total reflection lens through the monitoring 45-degree reflection lens and is collected by a second image capturing unit.
By adopting the technical scheme, a worker can monitor the welding process through the monitoring picture displayed on the man-machine interaction screen, can monitor the temperature of a welding spot in the welding process, and can adjust the laser power in time, so that the probability of burning out the PCB is reduced; meanwhile, because the laser welding, the temperature measurement and the image acquisition are integrated, the laser welding, the temperature measurement and the image acquisition are coaxial, so that the monitoring effect is relatively better.
In summary, the present application includes at least one of the following beneficial technical effects:
1) The laser welding is adopted, so that the process is contactless, stress-free and damage-free;
2) Ensuring that no laser is injected into a through hole of the PCB and no laser is injected into a semiconductor device on the lower surface of the PCB by a mode of tin ring and tin paste, thereby avoiding burn of the PCB and the semiconductor device; meanwhile, the tin alloy amount can be accurately controlled, so that the waste of the tin alloy is avoided, the tin permeability is ensured, and the short circuit caused by excessive tin alloy is avoided;
3) The automatic tin ring taking and sleeving of the tin ring on the contact pin can be realized by combining visual identification with an automatic mechanical arm, and the welding efficiency is relatively higher;
4) The lower surface of the PCB is provided with a heating design, so that the fluidity of tin alloy can be improved, and the tin permeability is further improved;
5) The post-welding detection is performed based on the visual recognition function, so that scrapping can be reduced, and the yield is improved;
i.e., the present method can be used to improve the soldering quality of semiconductor devices.
Drawings
FIG. 1 is a schematic flow chart of the present method;
FIG. 2 is a schematic diagram of the line structure of the present method when it is carried out;
fig. 3 is a schematic structural diagram of an exploded PCB board and profiling jig according to an embodiment of the present method;
FIG. 4 is a schematic layout of the tin ring case of the method;
FIG. 5 is a schematic illustration of the structure of a spot solder paste assembly portion of the present method;
FIG. 6 is a schematic view of the structure of a laser welding apparatus of the present method;
fig. 7 is a schematic structural diagram of a laser generating module of the present method.
Detailed Description
The present application is described in further detail below in conjunction with fig. 1.
The embodiment of the application discloses a composite welding quality control method based on visual identification.
Referring to fig. 1, the vision recognition-based composite welding quality control method includes:
step one, hardware composition setting, which comprises: the device is provided with a tin ring box, a tin paste point assembly, a visual identification positioning device, a laser welding device and a PCB heating device.
Referring to fig. 2, it can be understood that when the method is implemented in a production line, the method can be used with a conveyor belt and a mechanical arm of the production line, and the corresponding profiling fixture matched with the PCB board is circulated on the production line, so that the upper surface of the PCB board is flat and stable; and it can be known that the PCB completes the operation of inserting the electronic device, the pins extend out of the upper/lower surfaces of the PCB, and then extend out from the upper or lower surfaces, which can be determined according to the arrangement of other components, and fig. 3 is a schematic diagram showing the structure of the PCB after explosion with the corresponding profiling fixture.
Referring to fig. 4, the tin ring case is installed near the transmission belt, the inner cavity of the tin ring case is opened upwards and is provided with a tin ring, and the inner diameter of the tin ring is adapted to the pins of the electrical device and the outer diameter of the tin ring is larger than the diameter of the through hole of the PCB. The tin ring box can be a plurality of tin ring boxes, and the tin ring boxes are arranged in an array, for example: the 6 tin ring boxes are distributed in two rows and one row of three, and tin rings with different specifications are placed in each tin ring box so as to meet various use requirements.
The inner diameter of the tin ring is as close as possible to the diameter of the contact pin, so that the tin ring can absorb more laser, and the situation that the PCB is damaged due to the fact that the laser is injected into the through hole is avoided; the outer diameter of the tin ring is larger than the diameter of the through hole so as to improve the probability that the melted tin alloy fills up the gap of the through hole of the PCB, and the tin permeability is as high as 100%.
Referring to fig. 5, in one embodiment of the method, the above-mentioned solder paste dispensing assembly is obtained based on an automated dispenser, and differs from the dispenser in that: the glue is changed into pasty tin alloy, and the tin alloy is extruded by a dispensing needle cylinder (head).
In one embodiment of the method, the visual recognition positioning device comprises a control module, a camera unit I and an annular light source. The control module can be a control host of a production line and a PLC controller, is electrically connected with the first camera unit and the annular light source, and can perform light control and analyze and process collected image data. The first image pickup unit comprises a CCD camera and a matched lens, the CCD camera and the matched lens are installed near the transmission belt through a bracket and the like, and the matched linear motor drives the CCD camera to shift when necessary.
The annular light source is 360 degrees, the light color of the annular light source can be white, and the annular light source is used for providing a plurality of light source light-emitting units with the same interval angle; taking 4 light source light-emitting units as an example, light-emitting at four angles of 90 degrees can be realized.
In the use process, the camera unit is enabled to collect target images under the condition that a plurality of light source light emitting units are sequentially started and simultaneously started; it can be seen that the object is irradiated from four angles of front, back, left and right respectively, so that four angles of images can be generated, under the irradiation of light sources at different positions, shadows of the object (such as pins) can form an obvious change on the bonding pad, the formed images are collected by the CCD camera and transmitted to the control host for processing, and the positioning accuracy is improved by data obtained through the images at different angles, so that the recognition has higher accuracy and reliability.
Referring to fig. 6, in an embodiment of the present application, the laser welding device may be a laser welding machine (mechanical arm) with matched power and freedom of movement of the laser welding head, and the laser welding device at least includes a laser generating module in addition to a structure for making the laser generating part work and move, so as to meet the requirement of laser welding operation.
Referring to fig. 6, in an embodiment of the present application, the PCB heating device is a non-contact heating structure and heats the solder pad with light or heat radiation. For example: the PCB heating device is made into a table shape and is arranged below the PCB in a welding state, and the welding Plate (PCB) is heated by infrared rays emitted by the infrared generating equipment.
The PCB heating device can keep the tin alloy in the PCB through hole in a liquid state and keep fluidity, so that the tin alloy gradually permeates to the lower surface of the PCB, the tin alloy permeation process is reduced, and the problems that the temperature of the tin alloy is reduced, the tin alloy gradually cools and the fluidity is reduced and the tin permeation rate cannot reach 100% along with the downward permeation are solved; the point works better as the thickness of the PCB is greater.
Step two, positioning and taking materials from the front surface of the tin ring, which comprises the following steps: the visual recognition positioning device is enabled to conduct image acquisition on the tin ring box, the position of the tin ring is positioned based on image recognition, and the tin ring is sucked/grabbed according to the position of the tin ring.
In one embodiment of the method, the identifying and locating process of the visual identifying and locating device comprises preliminary locating and accurate locating based on the preliminary locating result. This arrangement is to accurately obtain information such as the position and shape of the corresponding structure. The following description is made for several steps of identifying and locating in two steps.
The above-mentioned identification positioning process in this step, specifically:
1) Image preprocessing: the preprocessing flow is image noise reduction, image contrast enhancement, image quality improvement and processing efficiency improvement, and can be gray-scale image enhancement by using a spatial domain method, which are all the prior art, so that the description is omitted.
2) Edge feature extraction:
firstly, the initial positioning of the tin ring position is carried out, and after the gray processing is carried out on the image, the image is positioned. Such as: and (3) calculating the correlation coefficient between the gray values of a small area (namely the template) in the standard image and all areas with the same size in the image, and finding out the area which is most matched with the gray value of the template, so as to realize the initial positioning of the tin ring.
And extracting the characteristics of the preprocessed and initially positioned image. The feature extraction aims to obtain the coordinate position information of the tin ring. Such as: an edge detection algorithm is used to accurately find the edge of the outer ring of the tin ring near the edge of the outer ring of the tin ring after initial positioning, and image processing algorithms such as Sobel, prewitt, canny can be used. The edge points thus obtained can be used as a basis for the coordinate position of the tin ring.
3) Fine positioning:
and (3) according to the edge points of the outer ring of the tin ring obtained in the last step, using methods such as Hough transformation to find the center of the tin ring, and combining the obtained edges to obtain the position and shape information of the tin ring.
Through the steps, the coordinate position information of the tin ring is obtained, so that intelligent recognition and positioning of the tin ring are realized.
After the positioning is completed, in this embodiment, the mechanical arm can be used to cooperate with the claw to directly grasp or the suction nozzle driven by the mechanical arm to suck the tin ring.
Step three, positioning and shifting the back of the tin ring, which comprises the following steps: and (3) enabling the visual recognition positioning device to collect images from the lower part of the captured tin ring, extracting the inner diameter characteristics of the tin ring based on image recognition, and moving the tin ring to the upper part of the contact pin of the electronic device.
The shooting from the lower part or the back of the tin ring can verify whether the taking of the tin ring is finished or not, and on the other hand, because only one tin ring is arranged at the moment, the information such as the inner diameter size of the tin ring can be calculated more clearly and accurately.
Fourth, pad contact pin location, it includes: the visual recognition positioning device is enabled to conduct image acquisition on the contact pin of the electronic device, and the positions of the pins are recognized and positioned based on the images.
Identifying a positioning procedure, an exemplary description:
1) Image preprocessing:
firstly, adopting image enhancement technology such as contrast enhancement, histogram equalization and the like to improve the contrast of pins and bonding pads, and being easier to distinguish; meanwhile, a noise filter is applied to denoise the image, and noise in the image is removed for subsequent processing.
2) Whether the contact pin is identified or not:
and comparing the gray scale characteristics of each image with the gray scale characteristics of the template photo, and checking whether a pin exists in the photo.
3) And (5) definition identification:
and judging whether the pin image is focused or not by performing Fourier transformation on the image and analyzing the frequency spectrum of the image. If the pin is out of focus, indicating that the pin is too long or too short, it is necessary to stop identification and send a warning to the system.
4) And (3) extracting features:
since the image is taken from the vertical direction, the outline of the junction of the pin and the bonding pad is presented as a circle in the image, and we find the edge point of the circular outline by edge detection methods such as Canny edge detection, sobel edge detection and the like.
5) And (3) position identification:
and (3) carrying out data fusion on edge detection points obtained by shooting front, back, left and right illumination angles by adopting methods such as weighted average, maximum likelihood estimation and the like, and finding out the center of the round needle angle by adopting methods such as Hough transformation and the like, namely obtaining the accurate position of the stitch, wherein the center is the required coordinate.
Step five, lantern ring and verification, it includes: and sleeving the tin ring on a contact pin of the electronic device, and enabling the visual recognition positioning device to acquire images of the bonding pads, so as to recognize the posture of the tin ring based on the images.
In this step, the above-mentioned ring light source needs to emit light at the same time, shoot several MARK images with different exposure at the same position, in order to calculate the exposure parameter that the image contrast is best automatically according to these several images, then shoot the image again with the best exposure parameter, thus realize the automatic regulation exposure.
In this step, the image analysis: extracting gray features of the tin ring, identifying the outer edge of the tin ring, analyzing the roundness of the edge, if the tin ring is obviously elliptical, indicating that the tin ring is not tiled on the bonding pad, and is in an inclined state, and needing mechanical arm intervention, such as: and (3) stirring.
Step six, visual positioning, which comprises the following steps: and identifying and positioning based on the image acquired by the visual identification positioning device to obtain the position of the tin ring and the position of the contact pin which are sleeved.
This step is to allow positioning of the pins so that solder paste can be precisely extruded onto the tips of the pins.
Step seven, solder paste is dispensed, which comprises: the solder paste is dispensed to the top end of the pin of the electronic device by the solder paste dispensing component, and the gap between the solder ring and the pin is shielded by the solder paste in a top view.
The solder paste is dispensed on the top end of the pin, so that the gap between the solder ring and the pin is further shielded by the solder paste when seen from the upper side, and the probability that laser is injected into the PCB through hole through the gap between the solder ring and the pin is further reduced.
It will be appreciated that in this process, further analysis may be based on the image: according to the texture characteristics of the solder paste, an image segmentation algorithm is used for identifying the area and position distribution of the solder paste on the bonding pad, and the area and position distribution are compared with a standard template to judge whether the solder paste meets the requirements.
Step eight, laser welding, which comprises: the laser welding device is enabled to emit laser to heat the tin ring and the tin paste, and the PCB heating device is enabled to heat the bonding pad.
The position and the size of the tin ring can be obtained by the previous steps, so that the laser welding device can be controlled to adjust the position to emit laser to the target according to the tin ring. In this step, the heating pad (PCB board) can keep the tin alloy in the PCB through hole in a liquid state, keep fluidity, and gradually infiltrate into the lower surface of the PCB.
Step nine, post-welding detection, which comprises: the visual recognition positioning device is enabled to collect images of welding pads which are welded, and based on image recognition analysis, the proportion of tin spreading area to the welding pads, whether short circuits exist between adjacent welding spots, whether defects generated by laser burn exist or not and welding leakage exist.
Exemplary description: the annular light sources are respectively opened at the front, the back, the left and the right angles one by one, and the CCD camera sequentially lights the four angles to acquire images of the same welding spot, so that the surface detail information of the object is always different when the irradiated angles are different, and the information of the complete material welding spot can be acquired. Then, for image processing, using an image segmentation algorithm to identify the area and position distribution of the tin on the bonding pad, and comparing the area and position distribution with the standard template; laser burn features and missing weld features are also identified, and post-weld detection can be performed.
According to the above arrangement, the method comprises:
1) The laser welding is adopted, so that the process is contactless, stress-free and damage-free;
2) Ensuring that no laser is injected into a through hole of the PCB and no laser is injected into a semiconductor device on the lower surface of the PCB by a mode of tin ring and tin paste, thereby avoiding burn of the PCB and the semiconductor device; meanwhile, the tin alloy amount can be accurately controlled, so that the waste of the tin alloy is avoided, the tin permeability is ensured, and the short circuit caused by excessive tin alloy is avoided;
3) The automatic tin ring taking and sleeving of the tin ring on the contact pin can be realized by combining visual identification with an automatic mechanical arm, and the welding efficiency is relatively higher;
4) The lower surface of the PCB is provided with a heating design, so that the fluidity of tin alloy can be improved, and the tin permeability is further improved;
5) The post-welding detection is performed based on the visual recognition function, so that scrapping can be reduced, and the yield is improved;
i.e., the present method can be used to improve the soldering quality of semiconductor devices.
In one embodiment of the method, the visual recognition positioning device is enabled to recognize and analyze dirt and oxidation information on the tin ring image and the bonding pad image before the tin ring is sleeved when a plurality of light source light emitting units are simultaneously started.
Exemplary description:
1) And (3) image acquisition: all parts of the annular light source are lightened and photographed at the same time;
2) Preprocessing an image;
3) Image initial positioning: and (5) using a gray scale correlation method to initially position the tin ring.
4) Image analysis: and calculating whether the gray level distribution of the surfaces of the tin ring and the bonding pad is consistent with that of a normal product by using a gray level histogram, and if the difference is large, indicating that dirt or oxidation exists on the front surface of the tin ring.
According to the method, unqualified tin rings and PCB boards can be found in time in the process, so that welding quality is guaranteed.
In one embodiment of the method, the laser welding device comprises an infrared temperature measuring unit, a second camera unit and a man-machine interaction screen, wherein the second camera unit is a monitoring camera which can be arranged on a movable head of the laser welding machine to move along with the laser generating module and collect video images in the welding process; the man-machine interaction screen can be a touch display, the data of the man-machine interaction screen are connected to the second camera unit and the infrared temperature measuring unit, and the infrared temperature measuring unit can be infrared temperature measuring equipment, such as an infrared coaxial thermometer, which is used for infrared temperature measurement of welding positions.
At this time, step eight, laser welding, it still includes: and adjusting the light-emitting power of the laser generating module according to the temperature information fed back by the infrared temperature measuring unit, so that the temperature of the welding position accords with a preset temperature threshold.
According to the above-mentioned setting, on the one hand, the staff can look over real-time welding video image through human-computer interaction screen and observe welding process, and on the other hand uses infrared probe temperature measurement, but real-time supervision welded region (laser focus position, namely the central point of tin ring put) to adjust laser output in real time, thereby make welded region remain at the temperature of settlement all the time, like: the temperature can be set to be the melting temperature of the tin alloy, so that the welding area can not exceed the melting temperature of the tin alloy all the time in the welding process, and the laser energy (namely, heat) input which can melt the tin alloy but is not too high is always maintained, thereby better protecting the PCB and the semiconductor device from being damaged in the laser welding and heating processes.
The adjustment mode of the light-emitting power of the laser generation module at different temperature values is preset after the verification of staff.
In another embodiment of the method, the method further comprises: production line anomaly prediction, comprising:
invoking a pre-established neural network model to train the electrical control parameters, the electrical operation parameters and the images of all the steps in the production process;
and importing the electrical control parameters, the electrical operation parameters and the image analysis results of all the steps in the current production process into a trained neural network model, predicting the welding result, and generating a welding quality development trend table/graph.
Illustrative, neural network implementation:
feature extraction is performed by using a deep-learning Convolutional Neural Network (CNN), and batch processing is performed on data in the welding process, such as: the feature vectors such as the size/position of the tin ring, the movement control parameters in the tin ring sleeving process, the power of welding laser, the heating power of the PCB, the thickness of the PCB and the like form a feature map.
Classifying feature graphs using a fully-connected network (FCN) is a fully-automatic process that does not require manual parameter setting.
The gradients are updated by a back propagation algorithm, the weights of the network are optimized, and predictions are made on the validation set to adjust the model.
And repeating the training and verification processes until the model converges, finally testing on a test set, using the model which has passed the test evaluation as a result prediction based on the related data of each production, prompting workers in time to carry out production line maintenance and the like, and guaranteeing the welding quality of the product.
The embodiment of the application also discloses a quality control system of the composite welding quality control method based on visual identification.
The quality control system of the composite welding quality control method based on visual recognition comprises the following components: the device comprises a tin ring box, a solder paste dispensing assembly, a visual identification positioning device, a laser welding device and a PCB heating device.
The above structures have been described together in the embodiments of the method and are therefore not described in detail.
Referring to fig. 5, in one embodiment of the present system, one camera unit one and one annular light source are a set of vision units, the vision recognition positioning device includes a plurality of sets of vision units, and at least one set of vision units is mounted on the solder paste dispensing assembly, and another set of vision units is mounted on the mechanical arm for sucking/grabbing the solder ring, and is located beside the suction nozzle/claw. The first image capturing unit is located on the central axis of the annular light source, so that when the annular light source is sequentially lightened, the image acquired by the first image capturing unit has symmetrical characteristics, and the image is relatively less in skew and distortion caused by the shooting angle, and has fewer interference factors.
Referring to fig. 6 and 7, in one embodiment of the system, the laser generating module includes an integrated housing, a central channel, a laser channel and a monitoring channel are formed in the integrated housing, an infrared temperature measuring unit (such as a coaxial temperature measuring instrument is arranged at the upper end of the central channel vertically, an optical filter is arranged below the temperature measuring instrument to reduce interference), and a laser focusing lens is arranged at the other end of the central channel (a dustproof protection lens can be additionally arranged below the central channel).
The laser channel is L-shaped, the transverse end is communicated with the upper side wall of the central channel, the other end is upwards provided with a laser generating module (such as a laser optical fiber), and a laser collimating lens group is arranged below the laser generating module for adjustment.
The monitoring channel is L-shaped, the transverse end is communicated with the lower side wall of the central channel, the other end is upward and is provided with a second image pick-up unit (monitoring camera), a monitoring focusing lens is arranged below to assist focusing, and a monitoring adjusting grating is arranged below.
The corner of the laser channel is provided with a laser total reflection lens, the upper part of the central channel is provided with a laser 45-degree reflection lens, the laser 45-degree reflection lens is parallel to the laser total reflection lens, the laser emitted by the laser generating module falls on the laser total reflection lens and is reflected to the laser 45-degree reflection lens, and the laser 45-degree reflection lens reflects the laser and enables the laser to be emitted from the lower port of the central channel.
The monitoring total reflection lens is arranged on the crutch of the monitoring channel, the monitoring 45-degree reflection lens is arranged below the laser 45-degree reflection lens, the monitoring 45-degree reflection lens is parallel to the monitoring total reflection lens, and an image below the central channel is reflected to the monitoring total reflection lens through the monitoring 45-degree reflection lens and is collected by the second image pickup unit. A monitoring filter may also be installed at the lateral port of the monitoring channel to reduce interference.
Further, to improve the clarity of the monitoring graph, a coaxial light source may be further disposed around the lower port of the central channel.
According to the arrangement, a worker can monitor the welding process through the monitoring picture displayed on the man-machine interaction screen, can monitor the temperature of a welding spot in the welding process, and can adjust the laser power in time, so that the probability of burning out the PCB is reduced; meanwhile, because the laser welding, the temperature measurement and the image acquisition are integrated, the laser welding, the temperature measurement and the image acquisition are coaxial, the temperature measurement radiates from the periphery at the center of a laser welding spot, and the center temperature can be directly monitored; the monitoring pictures are coaxial, the observation effect is better, the monitoring effect is relatively better, and the space utilization rate of the laser welding head is improved.
The foregoing are all preferred embodiments of the present application, and are not intended to limit the scope of the present application in any way, therefore: all equivalent changes in structure, shape and principle of this application should be covered in the protection scope of this application.

Claims (5)

1. A vision recognition-based composite welding quality control method is characterized by comprising the following steps:
step one, hardware composition setting, which comprises: setting a tin ring box, a tin paste dispensing assembly, a visual identification positioning device, a laser welding device and a PCB heating device; wherein, a tin ring is placed in the tin ring box, the inner diameter of the tin ring is suitable for the contact pin of the electrical device, and the outer diameter of the tin ring is larger than the diameter of the through hole of the PCB;
step two, positioning and taking materials from the front surface of the tin ring, which comprises the following steps: the visual recognition positioning device is used for carrying out image acquisition on the tin ring box, positioning the position of the tin ring based on image recognition, and absorbing/grabbing the tin ring according to the position of the tin ring;
step three, positioning and shifting the back of the tin ring, which comprises the following steps: the visual recognition positioning device is enabled to collect images from the lower side of the captured tin ring, the inner diameter characteristics of the tin ring are extracted based on image recognition, and the tin ring is moved to the upper side of a contact pin of an electronic device;
fourth, pad contact pin location, it includes: the visual recognition positioning device is enabled to conduct image acquisition on the contact pin of the electronic device, and the positions of the pins are recognized and positioned based on the images;
step five, lantern ring and verification, it includes: sleeving the tin ring on a contact pin of the electronic device, enabling the visual identification positioning device to acquire an image of the bonding pad, and identifying the posture of the tin ring based on the image;
step six, dispensing solder paste, which comprises the following steps: the solder paste is dispensed to the top end of the contact pin of the electronic device by the solder paste dispensing component, and the gap between the solder ring and the contact pin is shielded by the solder paste in a plan view;
step seven, visual positioning, which comprises the following steps: identifying and positioning based on the image acquired by the visual identification and positioning device to obtain the position of the tin ring and the position of the contact pin which are sleeved;
step eight, laser welding, which comprises: the laser welding device is enabled to emit laser to heat the tin ring and the solder paste, and the PCB heating device is enabled to heat the bonding pad;
step nine, post-welding detection, which comprises: the visual recognition positioning device is used for carrying out image acquisition on the welding pads which are welded, and analyzing the proportion of the tin spreading area to the welding pads, whether short circuits exist between adjacent welding spots, whether defects generated by laser burn exist or not and whether welding leakage exists or not based on the image recognition;
the visual recognition positioning device comprises a preliminary positioning and a precise positioning in the recognition positioning process;
the visual recognition positioning device comprises a first camera unit and an annular light source, wherein the annular light source provides a plurality of light source light-emitting units with the same interval angle, and the first camera unit is used for collecting target images under the condition that the light source light-emitting units are sequentially started and simultaneously started;
the visual recognition positioning device is made to recognize and analyze dirt and oxidation information on the tin ring image and the solder pad image before the tin ring is sleeved when the light emitting units of the light sources are simultaneously started, and is made to recognize and analyze the gesture of the tin ring and the area and position distribution of the solder paste respectively on the solder pad image after the tin ring is sleeved and the solder pad image after the solder paste is coated when the light emitting units of the light sources are simultaneously started;
the laser welding device comprises a laser generation module, an infrared temperature measurement unit, a second camera unit and a man-machine interaction screen, wherein the second camera unit moves along with the laser generation module, and the man-machine interaction screen is connected with the second camera unit and the infrared temperature measurement unit in a data mode;
the infrared temperature measurement unit is used for infrared temperature measurement of the welding position;
step eight, laser welding, it still includes: adjusting the light-emitting power of the laser generating module according to the temperature information fed back by the infrared temperature measuring unit, so that the temperature of the welding position accords with a preset temperature threshold;
also included is production line anomaly prediction, comprising:
invoking a pre-established neural network model to train the electrical control parameters, the electrical operation parameters and the images of all the steps in the production process;
and importing the electrical control parameters, the electrical operation parameters and the images of all the steps in the current production process into a trained neural network model, predicting a welding result, and generating a welding quality development trend table/graph.
2. The vision-recognition-based composite welding quality control method of claim 1, wherein: the PCB heating device is of a non-contact heating structure and heats the welding disc by adopting light or heat radiation.
3. A quality control system for use in a vision-based composite welding quality control method as defined in any one of claims 1-2, characterized by: the device comprises a tin ring box, a solder paste dispensing assembly, a visual identification positioning device, a laser welding device and a PCB heating device.
4. A quality control system according to claim 3, characterized in that: the first camera unit and the annular light source are a group of visual units, the visual identification positioning device comprises a plurality of groups of visual units, and at least one group of visual units is arranged on the solder paste point assembly; the first image pick-up unit is positioned on the central axis of the annular light source.
5. A quality control system according to claim 3, characterized in that: the laser welding device comprises a laser generation module, an infrared temperature measurement unit, a second camera unit and a man-machine interaction screen, wherein the man-machine interaction screen is connected with the second camera unit and the infrared temperature measurement unit in data, and the infrared temperature measurement unit is used for infrared temperature measurement of a welding position;
the laser generation module comprises an integrated shell, a central channel, a laser channel and a monitoring channel are arranged in the integrated shell, the central channel is vertical, an infrared temperature measurement unit is arranged at the upper end of the central channel, and a laser focusing lens is arranged at the other end of the central channel; the laser channel is L-shaped, the transverse end is communicated with the upper side wall of the central channel, and the other end is upward and is provided with a laser generating module; the monitoring channel is L-shaped, the transverse end is communicated with the side wall of the lower part of the central channel, and the other end is upward and is provided with a second camera unit;
the corner of the laser channel is provided with a laser total reflection lens, the upper part of the central channel is provided with a laser 45-degree reflection lens, the laser 45-degree reflection lens is parallel to the laser total reflection lens, the laser emitted by the laser generating module falls on the laser total reflection lens and is reflected to the laser 45-degree reflection lens, and the laser 45-degree reflection lens reflects the laser and enables the laser to be emitted from the lower port of the central channel;
the corner of the monitoring channel is provided with a monitoring total reflection lens, a monitoring 45-degree reflection lens is arranged below the laser 45-degree reflection lens, the monitoring 45-degree reflection lens is parallel to the monitoring total reflection lens, and an image below the central channel is reflected to the monitoring total reflection lens through the monitoring 45-degree reflection lens and is collected by a second image capturing unit.
CN202311718802.1A 2023-12-14 2023-12-14 Composite welding quality control method and system based on visual recognition Active CN117412517B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311718802.1A CN117412517B (en) 2023-12-14 2023-12-14 Composite welding quality control method and system based on visual recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311718802.1A CN117412517B (en) 2023-12-14 2023-12-14 Composite welding quality control method and system based on visual recognition

Publications (2)

Publication Number Publication Date
CN117412517A CN117412517A (en) 2024-01-16
CN117412517B true CN117412517B (en) 2024-03-12

Family

ID=89498338

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311718802.1A Active CN117412517B (en) 2023-12-14 2023-12-14 Composite welding quality control method and system based on visual recognition

Country Status (1)

Country Link
CN (1) CN117412517B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN210334656U (en) * 2019-07-25 2020-04-17 深圳泰德激光科技有限公司 Laser soldering automation assembly line
CN212470114U (en) * 2020-07-06 2021-02-05 深圳市紫宸激光设备有限公司 Laser welding tin head with coaxial temperature measurement
CN212704875U (en) * 2020-07-14 2021-03-16 深圳市紫宸激光设备有限公司 Laser welding tin head with coaxial light source
CN114799588A (en) * 2022-06-07 2022-07-29 广东省索艺柏科技有限公司 Resistance spot welding pre-packaging method for pre-formed cover plate with gold-tin welding ring
CN115255631A (en) * 2022-07-29 2022-11-01 深圳市青虹激光科技有限公司 Laser welding equipment with temperature feedback and circuit board welding method
CN116329689A (en) * 2023-04-27 2023-06-27 西安微电子技术研究所 Laser welding process of high-density long needle plate connector
CN116673561A (en) * 2023-06-07 2023-09-01 北京计算机技术及应用研究所 Laser welding process
CN117177448A (en) * 2023-08-30 2023-12-05 武汉工程大学 Be applied to printed circuit board's accurate solder paste spray printing equipment

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN210334656U (en) * 2019-07-25 2020-04-17 深圳泰德激光科技有限公司 Laser soldering automation assembly line
CN212470114U (en) * 2020-07-06 2021-02-05 深圳市紫宸激光设备有限公司 Laser welding tin head with coaxial temperature measurement
CN212704875U (en) * 2020-07-14 2021-03-16 深圳市紫宸激光设备有限公司 Laser welding tin head with coaxial light source
CN114799588A (en) * 2022-06-07 2022-07-29 广东省索艺柏科技有限公司 Resistance spot welding pre-packaging method for pre-formed cover plate with gold-tin welding ring
CN115255631A (en) * 2022-07-29 2022-11-01 深圳市青虹激光科技有限公司 Laser welding equipment with temperature feedback and circuit board welding method
CN116329689A (en) * 2023-04-27 2023-06-27 西安微电子技术研究所 Laser welding process of high-density long needle plate connector
CN116673561A (en) * 2023-06-07 2023-09-01 北京计算机技术及应用研究所 Laser welding process
CN117177448A (en) * 2023-08-30 2023-12-05 武汉工程大学 Be applied to printed circuit board's accurate solder paste spray printing equipment

Also Published As

Publication number Publication date
CN117412517A (en) 2024-01-16

Similar Documents

Publication Publication Date Title
CN108802046B (en) Optical detection device and detection method for defects of hybrid integrated circuit assembly
US5058178A (en) Method and apparatus for inspection of specular, three-dimensional features
JP5421763B2 (en) Inspection apparatus and inspection method
CN108489986A (en) A kind of increasing material manufacturing on-line checking and restorative procedure
US20110222655A1 (en) X-ray inspection apparatus, x-ray inspection method, x-ray inspection program, and x-ray inspection system
CN1960602A (en) Automatic detection, repair and apparatus and method for automatic marking printed circuit board
CN114505553B (en) Laser soldering method, device and system
CN113504239B (en) Quality control data analysis method
WO1999000661A1 (en) Method and apparatus for inspecting a workpiece
CN111735850A (en) Scanning type circuit board solder joint insufficient solder joint automatic detection system and detection method
CN106226270B (en) The method of the dirty defect of detection image sensor surface
JP6785407B2 (en) Bump electronic component mounting device and bump electronic component mounting method
CN114749745B (en) Automatic laser soldering method, device and system
JP2024520368A (en) High-speed automatic film foreign body detection device
CN117412517B (en) Composite welding quality control method and system based on visual recognition
Lertrusdachakul et al. Vision-based control of wire extension in GMA welding
CN112588607A (en) Multi-view soldering tin defect detection device based on deep learning
CN116012285A (en) BGA welding spot defect detection system and detection method based on laser ultrasound
JP6387620B2 (en) Quality control system
CN106910694B (en) A kind of automatic detection at circuit devcie angularity method and device
CN206677406U (en) Integrated welding, which is agreed to play, takes the photograph structure and battery core automatic welding detection means
JP3355978B2 (en) Bond coating inspection apparatus and bond coating inspection method
EP0430095A2 (en) Inner lead bonding inspecting method and inspection apparatus therefor
CN111230299A (en) Laser welding online detection device and method
CN112730453B (en) Bottom visual inspection device after multi-pin element pin insertion

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant