|
Laser soldering systems have been available for many years, but with the development of newer, high power diode lasers, selective soldering with laser is now gaining wider acceptance in the industry for the production of miniaturized interconnects. Similarly, lead-free soldering has been introduced as a response to the proposed legislative restrictions on the useof lead in electronics in the Far East and Europe. Diode lasers are becoming increasingly attractive as an option for selective soldering in the microelectronics industry due in part to their increased reliability, better electrical conversion efficiency and cost-effectiveness. The other optical soldering technology, Xenon short-arc heating lamp (as known as Soft beam) has been in the market for years now. Both optical soldering technologies have inherent advantages because of their non-contactnature. The majority of laser soldering systems currently being used produce joints sequentially as opposed to conventional soldering processes, such as wave soldering. Sequential soldering of single joints normally requires less than 10 watts of power and is therefore within the powerrange of different types of diode laser devices. Soft Beam is a trademark of the light beam heater with an optical fiber developed by Panasonic Factory Automation Company. The light from a xenon-arc amp is passed through an optical fiber of diameter 5mm, which is connected, to a converging lens, which focuses the lightbeam to a spot diameter of 1mm. Advantages of diode laser soldering
With interest in diode laser soldering being rejuvenated, it was deemed necessary to compare it with the original Xeon Arc technology which many customers havepreviously used. The spectral content of the xenon lamp is spread over a range of wavelengths between 350 to 1100nm, whereas the laser diode is highly monochromatic between 810 to 980nm. Thus the laser diode has a higher power density/ convergence compared to the xenon lamp. A diode laser output could be focused to a much smaller spot diameter(0.4mm) than the Soft Beam light (1mm). Basically, this means that the diode laser has the ability to focus the energy beam precisely onto the work piece with minimal heating of the surrounding parts. It has a small heat affected zone, hence creating less heating of the board. Thus it is suitable for use when heat sensitive components are present in the vicinity of the part to be soldered. It avoids the need to preheat the components when high thermal mass substances have to be soldered. Furthermore, rapid heating and cooling of the solder results in finegrained solder microstructure, with improved mechanicalproperties. Evaluating performance on both tin-lead and lead-free solders
Lead-free soldering is currently being introduced on a wide scale as a response to the current/proposed legislative restrictions on the use of lead in electronics, especially in the Far East and Europe. Various alternatives have been proposed, but tin-silver (Ag-Sn) group of alloys has been shown to be functionally equivalent to lead-tin (Pb-Sn) alloys in certain situations, although there is still some concern over replacements for higher temperature (~300ºC) lead based solders. However a combination of factors such as availability, mechanical properties, and melting point led to choosing tin-silver (96.5% Sn 3.5% Ag) as the materialcomposition to study. In order to evaluate the solder joint quality, predict cycle time and select process parameters, it is necessary to model the thermal dynamic response of the solder joint. In an earlier research work conducted by Spectra Technologies employees, a mathematical model was developed for a‘through hole” laser soldering process in order to determinethe thermal signatures of the solder. The lumped-parametermodel can be used to study the temperature response ofthe solder as well as the component under different processparameters like laser power and pulse duration. In addition, a direct diode laser and Soft Beam lamp were integrated into similar Adept robots to develop automated selective soldering machines of equal motion control and movement. The following attempts to compare the two heating technologies for both tin-lead as well as lead-free solders, and examine the quality of the finishedsolder joints. Results for experiments with tin-lead solders
Soldering experiments were conducted using a diodelaser and Soft Beam for 60-40 lead-tin solder. • Diode laser soldering
A 25W diode laser was used to conduct the soldering experiments. The heart of the system is a diode laser system (QUATTRO FAP system). A Fiber Array Package (FAP) module, built into the diode laser system, generates the diode laser beam. This is a standard commercially available system. In this case, the FAP module uses a diode bar emitting at a wavelength of 980nm. The FAP module is built around a single 19 emitter diode laser bar, with each of these emitters coupled into individual small diameter fibers. These 19 fibers are then brought together into a flexible fiber bundle with a diameter of 800nm. The delivery fiber is coupled to a standard Optical Imaging Accessory (OIA). This OIA consists simply of two lenses: a collimating and a focusing lens, which produce a 1:1 image of the fiber tip at the working distance of 33mm from the focusing lens. The laser was connected to a z-axis of a Cartesian robotic system which was integrated with a micro camera to make a vision-guided laser soldering work station. At a focal distance of 33mm, the laser is focused to a spot diameter of 0.8mm on the top of the FR4 printed circuit board. Tests were conducted at different power levels and pulse durations using 60-40 solder preforms as shown in figure 1 with the following dimensions: OD 1.90mm x ID 1.20mm x thickness 1.50 mm. The dimensions of the PCB are 89mm x 51mm x 1.5mm. The copper track has thickness of 0.25mm. The component to be soldered is a standard resistor network. • Soft Beam
Similar soldering tests were conducted using Soft Beam as a heating source with the results shown in figure 2. Further experiments were carried out using 60-40 solderwire for different joints. Thus we see that for soldering a particular joint using a 60-40 solder preform, the process parameters are: • Soft Beam : 60W / 1s• Diode laser: 25W / 0.8s These results indicate that lesser power and shorter heating time is required when using diode laser. Experiments with lead-free solder
A single, standard solder joint configuration was used to demonstrate a generic experimental approach to minimizing heat input. A small FR4 circuit board was designed specifically for investigations into diode laser soldering. In addition, for this work, tin plated copper through holes on a pitch of 0.1” were provided on the board such that a standard 44 pin chip carrier could be soldered into the board. The through holes were standard tin plated copper. To achieve reproducible solder joints, 1.7mm OD (0.07”) solder preforms were used. The preform was selected over solder paste or solder wire feeding, as the use of the preforms provides more reproducible results especially in a laboratory environment. In this case, 96.5%/3.5% tin-silver solder preforms were used, melting point 221ºC, supplied by Alpha Fry Technologies. These are coated with an RMAtype low solids flux, making it a ‘no clean’ solder. 0.61mm thick solder preforms were used, 1.65mm OD, 0.56mm ID. These were pre-placed by hand onto the pins. To ensure a valid optimization trial, soldering time was fixed at 0.8s. Average power and solder composition were therefore the primary variables studied. Average power was changed in increments of 2W over the range at which solder joints were produced. At each setting, a total of 13 pins were soldered. Results over the optimum range of 6 to12W average power are reported here. Inference
It can be seen that in this particular joint configuration, for a soldering time of 0.8s using the particular lead-free solder composition, an average power of 10W produces the best quality solder joint. When compared with a standard 60/40 PbSn solder composition, an increase of 2W in average power is required to produce an optimized joint. This power increase helps achieves the increased temperature required for the soldering of higher melting point lead-free solders. Conclusions
The diode laser system successfully soldered both leadfree and lead-containing solder joints with higher quality in less time. Very thin intermetallic layers and small grain size were observed on the diode laser soldered joints. This suggests that the mechanical properties of the joints would be excellent. Diode laser has shown to be very successful in producing high quality, low heat input joints for the type of through-hole electronic components used in this experiment. The diode laser clearly out-performed Soft Beam in soldering speed. The diode laser (at 810nm) was able to penetrate the solder with much more energy efficiency, reducing the time required to create a good solder joint. Diode laser soldering systems may be extremely useful in high mix production areas that require both tin lead and lead free soldering capabilities within the same line. Finally, based upon previous experience integrating diode laser systems, the 810nm rated diode offers more power at the joint but is more reflective while the 980nm rated diode is more penetrative and offers less reflectivity. Thus, diode laser nm rating should be selected based upon application(s) and production requirements. Gary Goldberg Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Table 1 Table 2 Table 3 References: [1] Panasonic Factory Automation Company, Operations Manual, Panasonic Soft Beam system
[2] Syed Naveed, Robert Woods, “Diode Laser Soldering – A Lumped-parameter Mathematical Model and Comparison of Different Soldering Technologies”, Proceedings of the conference on High power diode laser technologies and applications, Photonics West 2003, San Jose, CA, Jan 2003
[3] Coherent Semiconductor Division Data sheet, FAP System.
[4] Promation Robotic Soldering Division and Trotter Controls (Gary Goldberg & Victor Trotter) | | 
