Estimated reading time: 10 minutes

SMT Solver: Developing a Reflow Profile

• Provide an overview of various types of thermal profiles, purpose, key requirements, and challenges in developing thermal profiles.
• Explain the importance of different soldering zones—such as preheat, soak, reflow, and cooling—and their impact on solder quality.
• Discuss the details of how and where to attach thermocouples to achieve the desired soldering temperatures in various soldering zones.
• Emphasize the importance and difference between time above liquidus (TAL) and true TAL and their impact on the quality of solder joints, especially on head-on-pillow.
• Provide specific recommendations for thermal profiles for most commonly used Pb-free (SAC and low-temp), Sn-Pb, and mixed alloys.
• Discuss conflicting requirements of different types of packages and alloys on the same board and show some examples of profile related defects

What Is a Thermal Profile? 
A thermal profile is a unique temperature vs. time plot for each fully populated printed wiring board assembly (PWBA) using thermocouples attached to the solder joints with high-temperature solder, copper, or aluminum tapes to selected representative components on the board as the board travels at a given belt speed through various temperature zones of an oven or soldering system. There is a lot in that one sentence, but it is worth paying attention to some of the things I mentioned. 
 
For example, you need to use a fully populated board and definitely not a bare board or another board that possibly looks like the board but a fully populated board that comes  as close as possible to the thermal mass of the product you are profiling. You are using thermocouples that are attached with high-temperature solder or copper or aluminum tapes and not Kapton tapes used by many people. The location of attachment is generally the solder joint itself (unless you’re specifically monitoring other spots, such as temperature-sensitive body components or the bare board itself, when using low-Tg PCBs.) And even the length (about a meter) of the thermocouple wires and their gauges (about 36 gauge) is very important so that you don’t get the incorrect temperature.

And you need a unique profile for each product. One profile will not work for all products, even if they tend to look alike. The reason is simple. Each PWBA has a unique thermal mass. Different boards have varying thermal mass because they have a various number of layers, and the number and location of ground and power planes may be different, and they certainly have different types of components. Even the same board will need two profiles if it has components on both sides. For example, a single-sided board will need different temperature and belt speed settings than the same board loaded with components on both sides. Think of a single-sided board as a chicken and the double-sided heavier board as a turkey. Chicken and turkey require very different thermal profiles to prevent overcooking or undercooking. PWAs are no different. 
 
What Is a Misleading Profile? 
Profiles not correctly developed are misleading profiles. For example, profiles using bare boards will give you a misleading profile. You must use a fully loaded board. Profiles using rejected bare PCBs and dummy components or pennies that simulate the thermal mass of actual components is a good place to start if real boards and components are not available or affordable. 
 
Profiles using Kapton tapes to attach thermocouples will give you wrong temperatures, especially since they tend to come off during the profiling process. You must attach TCs using high-temperature solder or copper or aluminum tapes. Thermocouples attached to the PCB surface and not attached to solder joints will give you a misleading profile. It is important to keep in mind that developing the thermal profile is a destructive operation but needs to be performed only once during the entire life of that product. 
 
Developing a profile is like using correct DFM. You have to do it only once (DFM and profile development). But it is amazing to hear how the process engineers complain about DFM but don’t take the time, barely an hour, to develop a unique profile for each product.
 
Purpose and Types of Thermal Profiles
What is the reason for developing a unique profile for each product? There is only one purpose for using the correct profile: to reduce defects and to produce acceptable electronic assemblies. Yes, there are many causes of defects, and the right profile is not the panacea for solving all defects. Nothing is because there are many causes of defects in an electronic assembly, but the thermal profile plays a leading role. It is the simplest thing to take care of, but many people don’t take the time. There are many ways to develop a thermal profile, but there is only one way to correctly develop the thermal profile, which I mentioned earlier. 
 
There are two types of profiles: ramp to peak (RP) and ramp to soak to peak (RSP). The key difference between an RSP and an RP profile is the absence of the soak zone in the RP profile. An RSP profile allows a more uniform temperature across the PWBA, and it is very useful in achieving uniform temperature in a PWBA with a large variation in thermal masses of different components. RSP profiles also make it easier to achieve lower voids in solder joints, especially in BGAs. 
 
RP profiles may increase the incidence of voids in solder joints, but they minimize the incidence of head-on-pillow in BGAs, which is a serious defect. Here is what you need to keep in mind. The presence of voids is not a serious concern for product reliability, but head-on-pillow is an open causing a board to be nonfunctional; so it very simple. If there are BGAs on your board, worry about head-on-pillow and not voids. There are many causes of head-on-pillow—a topic that would take more than one column to discuss—but at least using the right profile is the easiest thing to take care of. 
 
Zones in a Profile
There are only two things we control in developing a profile: belt speed and temperature settings in each zone. Further, there are four zones in any profile: preheat, soak, reflow, and cooling. Here is a brief summary of each zone. 
 
1. Preheat Zone
The temperature in the preheat zone can be from 30–175°C. Component suppliers recommend 2–4°C per second ramp rate to avoid thermal shock to sensitive components. Such guidelines are considered conservative since some capacitors are wave soldered and go from a preheat temperature of nearly 120°C to a wave pot temperature of 260°C. A high ramp rate increases the potential for solder balls, so keep it as low as feasible with consideration for the recommended ramp rate for the temperature-sensitive components on the board.
 
2. Soak Zone
The soak zone is intended to raise the temperature of the entire PWBA to a uniform temperature. The recommended ramp rate in soak zone: 100–180°C for Sn-Pb and 140–220°C for SAC. The soak zone also acts as the flux activation zone for solder paste. The consequences of having too high a temperature in the soak zone can include solder balls, solder splatter due to excessive oxidation of paste, and spent flux activation capability (burnt-out flux that fails to scrub the oxidized surface, its main function). Also, the purpose of a long soak zone is to minimize voids, especially in BGAs and BTCs. 
 
As mentioned previously, it is also common practice to not use a soak zone but to steadily ramp the temperature from the preheat zone to peak reflow. The likelihood of voids may increase when ramping steadily (RP profile) to peak reflow temperature.
 
3. Reflow Zone
The peak temperature in the reflow zone should be high enough to obtain good wetting and to create a strong metallurgical bond. However, the temperature should not be so high as to cause the component or PWBA damage or discoloration or, in the worst case, delamination or charring of the PWBA. On the other hand, a temperature that is too low may result in cold and grainy solder joints, non-melted solder, or poor intermetallic bonding. As a general rule of thumb, a higher peak is preferable to a lower peak temperature to prevent opens/non-wetting. 
 
The recommended peak temperature in reflow zone is between 210–220°C (absolute minimum of 205°C) for SnPb and 235–245°C (absolute minimum of 230°C) for Pb-free solder alloy. The TAL should be 60–90 seconds but closer to 60 seconds. Extended duration above the solder melting point, or TAL, will damage temperature-sensitive components. It also will result in excessive intermetallic growth, which makes the solder joint brittle and reduces solder joint fatigue resistance.
 
4. Cooling Zone
During the cooling zone, various materials will cool at different rates. The BGA package typically will cool faster than the BGA solder joints and much faster than the bare board. This differential cooling can create a mechanical strain on the weakest spot in the interconnection, which is the laminate below the BGA pad, potentially resulting in pad  ?cratering. 
 
A faster cooling rate decreases the grain size, improves the joint strength, but increases the warpage and the potential for pad cratering. Pad cratering defects have become more common due to the increased stiffness of SAC solders and Pb-free laminates. Pad cratering depends not only on cooling rate but many other factors, such as stiffer lead-free solder and stiffer Pb-free laminate. As a practical matter, in most ovens, turning the cooling fan on and off are the only options for controlling the cooling rate unless the cooling zone has an option for blowing cold air.
 
Challenges in Developing Thermal Profiles
There are many challenges in developing a profile. For example, all solder joints must reach the minimum soldering temperature (15–20°C above liquidus) to allow wetting of the solder surfaces and the formation of intermetallic but also not exceed a maximum peak to prevent damage. However, there are components of different thermal mass (socket, BGA, chip components, etc.) that require different thermal input. A minimum soldering temperature is essentially determined by the largest component, such as a BGA, but the maximum is determined by smaller and temperature-sensitive components.
 
Even though different products, based on their thermal mass, require different amounts of thermal input, all products must achieve the minimum temperature (temperature above liquidus) without exceeding the maximum temperature (without damage to any components) within a defined time period (thermal profile). This is the key reason for developing a unique profile for each product. Developing a good profile is a balancing act to ensure intermetallic formation in heavier components without causing dewetting in smaller components due to overheating. 
 
Future Columns
In future columns, I will take on some of the other key points in reflow profile development that I mentioned in the beginning.