PCB Blog - Flip-Chip Assembly Technology for PCB board

With the emergence of miniaturized high-density packaging, the requirements for high-speed PCB board and high-precision assembly have become more critical, and related assembly equipment and processes have become more advanced and flexible. Since flip-chip has a smaller form factor, smaller ball diameter and ball pitch than BGA or CSP, it offers unprecedented possibilities for ball placement process, substrate technology, material compatibility, manufacturing process, and inspection equipment and methods. challenge. Nowadays, there are more and more miniaturized and high-density packaging forms of electronic devices, such as multi-module packaging (MCM), system-in-package (SiP), flip-chip (FC, Flip-Chip) and other applications. The advent of these technologies has further blurred the line between packaging and secondary assembly. Undoubtedly, with the emergence of miniaturized high-density packaging, the requirements for high-speed and high-precision assembly have become more critical, and related assembly equipment and processes have become more advanced and flexible. Since flip-chip has a smaller form factor, smaller ball diameter and ball pitch than BGA or CSP, it offers unprecedented possibilities for ball placement process, substrate technology, material compatibility, manufacturing process, and inspection equipment and methods. challenge. These requirements are analyzed in detail below:
1. For the requirements of mounting pressure control, considering that the flip-chip substrate for PCB is relatively brittle silicon, if a large pressure is applied during the process of material reclaiming and flux dipping, it is easy to be fractured, and at the same time Small solder bumps are also easily deformed during this process, so try to use a relatively low mounting pressure. The general requirement is about 150g. For ultra-thin chips, such as 0.3mm, sometimes the mounting pressure is even required to be controlled at 35g.

2. For the requirements of placement accuracy and stability, for devices with ball pitch as small as 0.1mm, what placement accuracy is required to achieve a high yield? The warpage and deformation of the substrate, the size and position deviation of the solder mask window, and the accuracy of the machine will all affect the final placement accuracy. We will not discuss the influence of substrate design and manufacturing on placement, but here we only discuss the placement accuracy of the machine.

3. The requirements of the chip assembly process for the placement equipment, in order to answer the above questions, let's build a simple hypothetical model:
1) Assume that the solder bumps of the flip-chip PCB board are spherical, and the corresponding pads on the substrate are circular and have the same diameter;
2) It is assumed that there is no influence of substrate warpage and manufacturing defects;
3) Does not consider the effects of Theta and shock;
4) During the reflow soldering process, the device is self-neutral, and 50% of the contact between the solder ball and the wetted surface can be "pulled up" during the soldering process. Then, based on the above assumptions, if the diameter of the solder ball with a diameter of 25μm is 50μm, the left and right position deviation (X axis) or the front and rear position deviation (Y axis) is 50% of the pad size. The balls are always on the pads. For flip-chip PCB boards with a solder ball diameter of 25μm, if the process capability Cpk is to reach 1.33, the precision of the machine must reach 12μm.

4. According to the requirements of cameras and image processing technology, a digital camera with megapixels is required to process the images of flip-chip PCB boards with fine solder ball pitches. Digital cameras with higher pixels have higher magnifications, however, the higher the pixels, the smaller the field of view (FOV), which means that larger devices may need to be "photographed" multiple times. The light source of the camera is generally light-emitting diode, which is divided into side light source, front light source and axial light source, and can be controlled independently. The imaging light source of flip chip for PCB board adopts side light, front light, or a combination of both. So how do you choose a camera for a given device? This mainly depends on the algorithm of the image. For example, it takes N pixels to distinguish one solder ball, and 2N pixels are required to distinguish the ball pitch. Taking the Magellan digital camera on Universal Instruments' placement machine as an example, it takes 4 pixels to distinguish one solder ball. Choose a camera, assuming that all images are 75% of the actual object size. The image processing of flip-chip fiducials for PCB boards is similar to that of ordinary fiducials. Flip-chip mounting for PCB boards often uses local fiducials in addition to GLOBAL fiducials. At this time, the fiducials will be small (0.15-1.0mm), and the selection of cameras refers to method above. The selection of the light source needs to be considered. Generally, the light source of the camera on the SMD head is red light, and the effect is very poor when dealing with the reference point on the flexible circuit board, and the reference point cannot even be found. The reason is that the surface of the reference point (copper) The color is very close to the substrate color, and the color difference is not obvious. If the blue light source technology of Universal Instruments is used, this problem can be solved very well.

5. Selection of nozzles
Since the flip-chip substrate for PCB is silicon, the upper surface is very flat and smooth, and the head is a rigid plastic material with a porous ESD nozzle. If you choose a nozzle with a rubber head, as the rubber ages, the device may stick to the device during the placement process, causing the placement to shift or take away the device.

6. Requirements for the flux application unit. The flux application unit is an important part of controlling the flux dipping process. The basic principle of its work is to obtain a stable flux film with a set thickness, so that each solder ball of the device can be easily dipped. Take the same amount of flux. To stably control the thickness of the flux film while meeting the requirements of high-speed dipping, the flux application unit must meet the following requirements:
1) It can meet the requirements of dipping multiple devices with flux at the same time (such as dipping 4 or 7 pieces at the same time) to increase the output;
2) The unit for flux should be simple, easy to operate, easy to control and easy to clean;
3) It can handle a wide range of fluxes or solder pastes. The viscosity range of the fluxes suitable for the dipping process is wide, and it can handle both thinner and more viscous fluxes, and the obtained film thickness should be uniform;
4) The dipping process can be controlled, and the dipping process parameters will be different due to different materials, so the process parameters of the dipping process must be individually controlled, such as downward acceleration, pressure, dwell time, upward acceleration, etc.

7. For the requirements of the feeder, to meet the production of high-speed and high-yield batches, the feeding technology is also very critical. Flip-chip packaging methods for PCB boards mainly include: 2*2 or 4*4 inch JEDEC reels, 200mm or 300mm wafer reels (Wafer), and reel reels (Reel). The corresponding feeders are: Stationary tray feeder, Automated stackable feeder, Wafer feeder, and tape feeder. All of these feeding technologies must be capable of high-speed feeding, and the wafer feeder is also required to be able to handle a variety of device packaging methods, such as: Device packaging can be JEDEC trays, or bare wafers, or even complete chips in the machine. Flip action. Let's take an example to illustrate the characteristics of Unovis's bare die feeder (DDF Direct Die Feeder):
1) Can be used in hybrid circuits or sensors, multi-chip modules, system-in-package, RFID and 3D assembly;
2) The disc can be fed vertically to save space, and one machine can install multiple DDFs;
3) The chip can be flipped in DDF;
4) Can be installed on a variety of patch platforms.

8. Requirements for board support and positioning system, some PCB flip chips are used on flexible circuit boards or thin circuit boards. At this time, the flat support of the substrate is very important. The solution often uses a carrier plate and a vacuum suction system to form a flat support and positioning system that meets the following requirements:
1) Support control in the Z direction of the substrate, and programming adjustment of the support height;
2) Provide customized board support interface;
3) Complete vacuum generator;
4) Non-standard and standard carrier boards can be applied.

9. Inspection after reflow soldering and filler curing, there are non-destructive inspection and destructive inspection for the inspection of the product after the underfill is completed. The non-destructive inspection includes:
1) Use an optical microscope to conduct visual inspection, such as checking whether the filler climbs on the side of the device, whether a good edge fillet is formed, and whether the surface of the device is dirty, etc.;
2) Use an X-ray inspector to check whether the solder joints are short-circuited, open-circuited, offset, wetted, voids in the solder joints, etc.;
3) Electrical test (Continuity test), which can test whether there is any problem with the electrical connection. For some test boards with a daisy-chain design, the location of the solder joint failure can also be determined through the continuity test;
4) Use ultrasonic scanning microscope (C-SAM) to check whether there are voids, stratification and complete flow after underfilling. Destructive inspections can section solder joints or underfill, combined with optical microscopy, metallographic microscopy or scanning electron microscopy and energy dispersive analyzers (SEM/EDX) to examine the microstructure of solder joints, e.g., microcracks/micropores, tin crystallization, intermetallic compounds, soldering and wetting conditions, whether the underfill has voids, cracks, delamination, and whether the flow is complete, etc. Common defects of products after reflow soldering and underfill process are: solder joint bridging/open circuit, poor solder joint wetting, solder joint void/bubble, solder joint cracking/brittleness, underfill and chip delamination and chip cracking, etc. . For the completeness of the underfill, whether there are voids, cracks and delamination in the filler, it needs to be observed by an ultrasonic scanning microscope (C-SAM) or a flat section parallel to the bottom surface of the chip. Defects add to the difficulty. Delamination between the underfill material and the chip tends to occur at the four corners of stressed devices or at the interface of the filler and solder joints.

Flip-chip for PCB has shown advantages in terms of product cost, performance and high-density packaging, and its application has gradually become mainstream. Due to the small size of the flip-chip used for PCB boards, it is necessary to ensure high precision, high yield and high repeatability, which brings challenges to our traditional equipment and processes, which are reflected in the following aspects:
1) The design of the substrate (hard board or soft board);
2) Assembly and inspection of equipment;
3) Manufacturing process, chip mounting process, PCB manufacturing process, SMT process;
4) Material compatibility.
A comprehensive understanding of the above issues is the basis for a successful flip-chip assembly process for PCB board.