Could you clarify if there are specific aspects of the article you want to focus on or any particular style you prefer?
1. It is estimated that the consumption of flip-chip chips will rise from 500 million in 1996 to 2.5 billion by the end of this century, while the consumption of TAB/TCP has either stagnated or shown negative growth. As anticipated, it was around 700 million in 1995.
2. PCB Mounting Method
3. The mounting requirements vary, and so does the mounting method (principle). These requirements include component pick-and-place capability, placement strength, accuracy, speed, and flux fluidity. Placement speed often hinges on placement accuracy.
4. Pick and Place
5. Fewer placement heads on the equipment generally mean higher placement accuracy. The accuracy of the x, y, and θ positioning axes impacts overall placement accuracy. The placement head is mounted on the xy plane of the support frame, with the rotation axis being crucial. However, the movement accuracy of the z-axis should not be overlooked. In high-performance placement systems, z-axis movement is controlled by a microprocessor, with vertical movement and placement force managed by sensors.
6. A key advantage of placement is that the precision head can move freely in the x and y planes, including retrieving materials from the waffle disk and conducting multiple measurements on the device with a fixed upward-looking camera.
7. The most advanced systems can achieve 4 sigma and 20μm accuracy on the x and y axes. However, the main disadvantage is a lower placement speed, typically under 2000 cph, excluding additional actions like flip chip coating flux.
8. Simple systems with a single placement head are being phased out in favor of more flexible systems. These systems feature a high-precision placement head and a revolver head, capable of handling large BGA and QFP packages. The rotating (or shooter) head can manage irregular devices, fine-pitch flip-chip chips, and μBGA/CSP chips with pin pitches as small as 0.5mm. This method is known as “collect, pick and place.”
9. High-performance SMD placement equipment with flip-chip spin heads is now available. It can mount flip-chips and μBGA and CSP chips with a ball grid diameter of 125μm and pin pitch of about 200μm at high speed. Equipment with collecting, picking, and placing capabilities can achieve a speed of around 5000 cph.
10. Traditional Wafer Sniffer
11. This system features a rotating head that moves horizontally while picking up components from a moving feeder and placing them onto a moving PCB circuit board. Theoretically, the placement speed can reach 40,000 cph, but it has limitations:
12. Wafer picking cannot exceed the grid plate where the device is placed;
13. The spring-driven vacuum nozzle limits work-hour optimization during z-axis movement or fails to reliably pick up the die from the conveyor belt;
14. For most surface array packages, placement accuracy may not meet requirements, with typical values exceeding 10μm at 4 sigma;
15. The application of soldering flux for micro flip chips is not feasible.
16. Collection and Placement
17. In the “collect and place” sniffer system, both rotating heads are mounted on the x-y support frame. Each head has 6 or 12 suction nozzles that can reach any position on the grid plate. For standard SMD chips, this system achieves a placement accuracy of 80μm and a placement speed of 20,000 pch at 4 sigma (including theta deviation). By adjusting the system’s dynamic positioning and ball grid search algorithms, the system can achieve placement accuracy of 60μm to 80μm and speeds over 10,000 pch for surface array packages at 4 sigma.
18. Placement Accuracy
19. To understand different placement equipment, it is crucial to know the main factors affecting placement accuracy for area array packages. Ball grid placement accuracy P//ACC// depends on ball grid alloy type, number of ball grids, and package weight.
20. These factors are interrelated. Surface array packages generally have lower mounting accuracy requirements compared to ICs in QFP and SOP packages with the same pitch. Note: Insert equation
21. For round pads without solder mask, the maximum allowable mounting deviation equals the radius of the PCB pad. If the mounting error exceeds this radius, there will still be mechanical contact between the ball grid and the PCB pad. Assuming the diameter of the PCB pad is roughly equal to the ball grid diameter, the placement accuracy for μBGA and CSP packages with a 0.3mm ball grid diameter and 0.5mm pitch is 0.15mm; if the ball grid diameter is 100μm and the pitch is 175μm, the accuracy requirement is 50μm.
22. For tape ball grid array packages (TBGA) and heavy ceramic ball grid array packages (CBGA), self-alignment is limited if it occurs. Thus, the placement precision requirements are high.
23. Application of Flux
24. Standard large-scale reflow soldering of flip-chip ball grids requires flux. Modern general-purpose SMD placement equipment often includes built-in flux application devices, with coating and dip soldering being the two common methods.
25. The coating unit is positioned near the placement head, applying flux to the placement area before flip-chip placement. The amount applied depends on the flip-chip size and solder wettability. Ensuring adequate flux coverage is crucial to avoid missed pads due to errors.
26. For effective filling in a no-clean process, the flux must be no-clean (residue-free). Liquid flux, containing minimal solid matter, is ideal for no-clean processes.
27. However, due to the liquid flux’s fluidity, flip-chip placement may cause inertial displacement. Solutions include:
28. Setting a waiting period before PCB transfer to allow flux evaporation, which improves adhesion but may reduce yield.
29. Adjusting conveyor belt acceleration and deceleration to match flux adhesion. Smooth belt movement minimizes wafer displacement.
30. A main disadvantage of flux coating is its longer cycle time, with mounting time increasing by about 1.5s per device.
31. PCB Dip Soldering Method
32. In this method, a rotating bucket is used for the flux carrier, and a blade scrapes it into a flux film (around 50μm). This method suits high-viscosity fluxes and reduces solder consumption by only dipping the solder at the bottom of the ball grid.
1. It is estimated that the consumption of flip-chip chips will rise from 500 million in 1996 to 2.5 billion by the end of this century, while the consumption of TAB/TCP has either stagnated or shown negative growth. As anticipated, it was around 700 million in 1995.
2. PCB Mounting Method
3. The mounting requirements vary, and so does the mounting method (principle). These requirements include component pick-and-place capability, placement strength, accuracy, speed, and flux fluidity. Placement speed often hinges on placement accuracy.
4. Pick and Place
5. Fewer placement heads on the equipment generally mean higher placement accuracy. The accuracy of the x, y, and θ positioning axes impacts overall placement accuracy. The placement head is mounted on the xy plane of the support frame, with the rotation axis being crucial. However, the movement accuracy of the z-axis should not be overlooked. In high-performance placement systems, z-axis movement is controlled by a microprocessor, with vertical movement and placement force managed by sensors.
6. A key advantage of placement is that the precision head can move freely in the x and y planes, including retrieving materials from the waffle disk and conducting multiple measurements on the device with a fixed upward-looking camera.
7. The most advanced systems can achieve 4 sigma and 20μm accuracy on the x and y axes. However, the main disadvantage is a lower placement speed, typically under 2000 cph, excluding additional actions like flip chip coating flux.
8. Simple systems with a single placement head are being phased out in favor of more flexible systems. These systems feature a high-precision placement head and a revolver head, capable of handling large BGA and QFP packages. The rotating (or shooter) head can manage irregular devices, fine-pitch flip-chip chips, and μBGA/CSP chips with pin pitches as small as 0.5mm. This method is known as “collect, pick and place.”
9. High-performance SMD placement equipment with flip-chip spin heads is now available. It can mount flip-chips and μBGA and CSP chips with a ball grid diameter of 125μm and pin pitch of about 200μm at high speed. Equipment with collecting, picking, and placing capabilities can achieve a speed of around 5000 cph.
10. Traditional Wafer Sniffer
11. This system features a rotating head that moves horizontally while picking up components from a moving feeder and placing them onto a moving PCB circuit board. Theoretically, the placement speed can reach 40,000 cph, but it has limitations:
12. Wafer picking cannot exceed the grid plate where the device is placed;
13. The spring-driven vacuum nozzle limits work-hour optimization during z-axis movement or fails to reliably pick up the die from the conveyor belt;
14. For most surface array packages, placement accuracy may not meet requirements, with typical values exceeding 10μm at 4 sigma;
15. The application of soldering flux for micro flip chips is not feasible.
16. Collection and Placement
17. In the “collect and place” sniffer system, both rotating heads are mounted on the x-y support frame. Each head has 6 or 12 suction nozzles that can reach any position on the grid plate. For standard SMD chips, this system achieves a placement accuracy of 80μm and a placement speed of 20,000 pch at 4 sigma (including theta deviation). By adjusting the system’s dynamic positioning and ball grid search algorithms, the system can achieve placement accuracy of 60μm to 80μm and speeds over 10,000 pch for surface array packages at 4 sigma.
18. Placement Accuracy
19. To understand different placement equipment, it is crucial to know the main factors affecting placement accuracy for area array packages. Ball grid placement accuracy P//ACC// depends on ball grid alloy type, number of ball grids, and package weight.
20. These factors are interrelated. Surface array packages generally have lower mounting accuracy requirements compared to ICs in QFP and SOP packages with the same pitch. Note: Insert equation
21. For round pads without solder mask, the maximum allowable mounting deviation equals the radius of the PCB pad. If the mounting error exceeds this radius, there will still be mechanical contact between the ball grid and the PCB pad. Assuming the diameter of the PCB pad is roughly equal to the ball grid diameter, the placement accuracy for μBGA and CSP packages with a 0.3mm ball grid diameter and 0.5mm pitch is 0.15mm; if the ball grid diameter is 100μm and the pitch is 175μm, the accuracy requirement is 50μm.
22. For tape ball grid array packages (TBGA) and heavy ceramic ball grid array packages (CBGA), self-alignment is limited if it occurs. Thus, the placement precision requirements are high.
23. Application of Flux
24. Standard large-scale reflow soldering of flip-chip ball grids requires flux. Modern general-purpose SMD placement equipment often includes built-in flux application devices, with coating and dip soldering being the two common methods.
25. The coating unit is positioned near the placement head, applying flux to the placement area before flip-chip placement. The amount applied depends on the flip-chip size and solder wettability. Ensuring adequate flux coverage is crucial to avoid missed pads due to errors.
26. For effective filling in a no-clean process, the flux must be no-clean (residue-free). Liquid flux, containing minimal solid matter, is ideal for no-clean processes.
27. However, due to the liquid flux’s fluidity, flip-chip placement may cause inertial displacement. Solutions include:
28. Setting a waiting period before PCB transfer to allow flux evaporation, which improves adhesion but may reduce yield.
29. Adjusting conveyor belt acceleration and deceleration to match flux adhesion. Smooth belt movement minimizes wafer displacement.
30. A main disadvantage of flux coating is its longer cycle time, with mounting time increasing by about 1.5s per device.
31. PCB Dip Soldering Method
32. In this method, a rotating bucket is used for the flux carrier, and a blade scrapes it into a flux film (around 50μm). This method suits high-viscosity fluxes and reduces solder consumption by only dipping the solder at the bottom of the ball grid.
