-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueBox Build
One trend is to add box build and final assembly to your product offering. In this issue, we explore the opportunities and risks of adding system assembly to your service portfolio.
IPC APEX EXPO 2024 Pre-show
This month’s issue devotes its pages to a comprehensive preview of the IPC APEX EXPO 2024 event. Whether your role is technical or business, if you're new-to-the-industry or seasoned veteran, you'll find value throughout this program.
Boost Your Sales
Every part of your business can be evaluated as a process, including your sales funnel. Optimizing your selling process requires a coordinated effort between marketing and sales. In this issue, industry experts in marketing and sales offer their best advice on how to boost your sales efforts.
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
Web Exclusive: Part II: The French Rework Connection
December 31, 1969 |Estimated reading time: 11 minutes
The continuation of the story of two companies that developed a universal method of verifying all types of rework equipment. The tool moves from initial design and test to use in real-world conditions.By Emmanuel Smague, Philippe Raout, Julien Ratie and Mike Hayward
When the French division of an international electronics manufacturing services (EMS) company* determined that a machine quality management (MQM) tool was needed to validate rework equipment at their facility, they formed a team to develop specifications and outline the capabilities needed in a tool of that type. Joining with a U.S. vendor company**, a new tool was born.
After several iterations in hardware design and software, a tool evolved that met all criteria. The MQM tool differed from those used for reflow or wave in that it was designed to detect and measure top and bottom temperatures — both air and mass — and both the physical design and its accompanying software package were geared to detailed monitoring of a static, rather than dynamic, production process.
The following parameters were selected because their results could be plotted and the interaction between them studied. By identifying the interactions, fault diagnosis could be conducted and failure modes assigned:
- T1 Top — topside, ambient air temperature in soak zone
- T1 Bot — bottom-side, ambient air temperature in soak zone
- t1 — Soak time commencing when TPCB (program defined) is reached
- T2 Top — topside, ambient air temperature in peak zone. Commencing at the end of t1
- T2 Bot — bottom-side, ambient air temperature in peak zone. Commencing at the end of t1
- t2 — peak time commencing at the end of t1
- Mass Top — topside component temperature throughout the process
- Mas Bottom — bottom-side component temperature throughout the process
- Delta T Top — the difference in temperature (Tb - Ta) at predetermined times in the process.
MQM construction consisted of a 200 x 254 mm pallet of composite delmat material (CDM) with a main sensor and zero mass thermocouple combo attached to the top and bottom side of the pallet. A section of the pallet held a portable data acquisition (thermal profiling) device to record input from the sensors.
Figure 1. The original MQM tool was fitted with an integral shield that held a portable data acquisition device (profiler).
In operation, the pallet is placed on the rework station table and a nozzle comes down to enclose the top sensors (Figure 1). The second sensor assembly, mounted under the pallet, detects heat energy from the bottom because most rework machines have a heating element underneath to minimize temperature differentials on the board that potentially could absorb the energy intended for the device under rework.
Tool Fine-tuning
After the final tests were completed and the tool went onto the production floor, some unexpected issues arose. Fine-tuning was done to both the tool's physical aspects and software.
Hardware. Once the new tool passed all tests, the EMS engineers used it frequently. Soon, some issues arose that changed the design and, unintentionally, added capabilities not considered at the start of the project.
The first change was in the basic design. Figure 1 shows that the original MQM pallet incorporated a holder for a data acquisition tool, which was a thermal profiler used in reflow oven profiling. In tests, this worked fine. Under real-world conditions and repeated use, the profiler was getting "toasted." It soon became apparent that the localized heat generated from stationary rework nozzles was far more concentrated than the profiler would see in a moving oven conveyor, even using a protective barrier.
Figure 2. Design changes included removing the integral data logger, centering the sensors and changing the type of plugs used on the unit.
Therefore, the data acquisition module was removed from the pallet (Figure 2). While this changed the tool from a completely self-contained unit, it resulted in several surprising benefits.
Figure 3. The new configuration allows a direct plug-in to the computer for real-time data acquisition and manipulation in an SPC program.
First, it reduced the weight of the pallet and allowed better centralization of the sensor mechanisms. Next, the original pallet had what were termed "micro plugs" for plugging into the data acquisition module. Removing the module allowed replacement of these with miniature K-type plugs that offer various options. The sensors now interface with any standard data acquisition profiler, with a high-end profiler/scan tool for real-time data transfer to a computer or directly into the computer control of rework machines (Figure 3).
The next physical change was the main topside sensor. There was a slight problem between the stainless steel sensor and the laser rework machine. The EMS engineers had done a "work-around" by covering the sensor with black tape to keep from reflecting the laser beam. This led the vendor company to develop a blackening process of the sensor housing to allow it to work with all rework machines, laser included, without in-house modifications.
Finally, there was a change to increase durability. The air sensor/thermocouple next to the main sensor originally was housed in a small rigid pipe. This housing was changed to a flexible spring assembly to make it more robust. If the unit gets knocked around on the shop floor, the sensors will not pop off or break.
Software. The initial software package presented everything the EMS team requested. They soon found out that "everything" was overkill. Because the software package was developed from cutting back features of a dynamic reflow oven MQM, it did the job; however, it was cumbersome. It was tempting to be able to track and log every aspect of the process, but by presenting too much information it was difficult to extract the key essential data needed most. The engineers had to hash out a specific list of the most useful parameters.
Once that was trimmed back, it left the essentials that give a true picture of what is going on at the rework station level. In terms of functionality, the statistical process control (SPC) works the same way as the other MQM (oven and wave) tool software. This makes the new tool's software friendly for those already used to working with the reflow and wavesolder MQM tools.
The final version of the rework MQM tool passed all tests for quantifying rework stations and has been in use for the last half of the year at the EMS company. It can measure top- and bottom-side temperatures for both mass and air temperatures with a high degree of integrity and accuracy (±-2.5°C) on hot air, IR and laser rework equipment equally. Tests show that it is capable of detecting 0.25 mm variation in nozzle heights, which is more than originally outlined.
For basic maintenance, all "perfect" profiles for each rework machine are stored in a database and checked regularly for machine drift, as is done with the ovens and wave machines. This is fairly standard and the new tool handles the job well.
A significant aspect this tool brings is the confidence to pinpoint trouble when it occurs. Even with standard maintenance, some rework machines can be temperamental. If one acts up, the engineer can check directly with the rework MQM tool and find the exact point where the trouble started. That gives maintenance a jump on finding and correcting the cause.
The tool also runs full diagnostics. The frequency depends on the type of drift seen on an individual machine. For example, the ovens are rather stable so eventually the diagnostic routine can be reduced from weekly to perhaps monthly for only preventive maintenance. For rework, because of the different types of machines, it is never known exactly from one day to another when big drifts may happen. Initially, it was expected that a rework machine would be easier to control because it is a much less dynamic process than a reflow oven. However, the opposite proved to be true. As a direct result of this project, the engineers found that rework machines are far more susceptible to change than reflow ovens. In ovens, the conveyor is a dynamic system but the temperature is a stable parameter during the process cycle. In rework machines, there is no conveyor system but the temperature is a dynamic parameter, a function of the time. That is why the temperature regulation is more critical on rework machines than on the ovens.
MQM in Use at the EMS Facility
SPC and process optimization are a reality now that the software creates a database of information on all the rework machines at the EMS facility. This enables the comparison of operating flow to tie in with the other equipment in the same manner as is done with data from the ovens. There now are profiles of all the rework machines. SPC charts are compiled for point comparisons key to the rework process. The EMS engineers now can apply tolerances in the software so that after each test it indicates immediately if the process is in or out of its limits. Initially, 20 to 30 cycles are run with the rework MQM to establish the normal variation within any system. Tolerances then are applied to these. Each set of results from later runs should be within those limits. If the software indicates a potential problem, the cycle is repeated to confirm and maintenance is called. The rework MQM tool helps maintenance personnel by highlighting the element that is at fault (top, bottom, convection power, etc.). Maintenance only has to look at the different SPC charts to see where the default is and then correct it.
Rework machines are not in constant use. They may be used for short periods and then stand idle for many hours. The result is that after a period of idleness the machines need to "warm through" before they can be used with confidence. The new rework MQM tool proves that a cold machine does not deliver the same signature as a warm one, and the first cycle always is a bit colder than subsequent cycles. This has been figured into procedures and the machines are warmed before running either a board or the MQM. This influences SPC charting and gives a truer picture of what is going on. One or two heat cycles are run before recording data with the rework MQM or when the machine is hot.
Case History Examples Figure 4. Delamination problem: On the first profile, notice the peak on the top air temperature.
Case Number 1. Recently, the EMS engineers had a problem on one card reworked on the SRT machine. There was a delamination problem on the PCB around the component. A recording cycle was done using the rework MQM. The resulting profile is shown in Figure 4. This profile shows peak on the top air temperature. On a second cycle, it recorded the same program's profile, increasing the time of the second zone, shown in Figure 5. From this profile it is apparent that there is an overshoot on the top air temperature regulation control. Analyzing the physical aspects of the machine closely found that the problem was stemming from a broken thermocouple in the nozzle. After changing it, a control run was made and the SRT returned to its normal configuration, as shown in Figure 6.
Figure 5. On a second cycle, the same program's profile was recorded, increasing the time of the second zone.
Case Number 2. To test the capacity of the new MQM, one engineer built some faults into one of the laser rework machines and then conducted a capability study on the laser machine using the new tool. Faults were introduced to see if they could be followed accurately as they were happening.
Figure 6. After changing the broken thermocouple in the nozzle, a control run showed that the SRT had returned to normal.
The charts in Figure 7 show laser drift and illustrate a simulation of the laser power decreasing (5°C by 5°C) on five runs. It shows that the response recorded by the rework MQM corresponded with the temperatures programmed on the laser. This not only proved that the tool works precisely in profiling the temperatures of the laser rework system, but also that the laser had no problems with the new blackened sensor.
Figure 7. These graphs illustrate a simulation of the laser power decreasing (5°C by 5°C) on five runs.
The changes in the tool's design that were borne of necessity resulted in benefits, particularly the aspect of running the data in real-time from the rework MQM on the rework machine platform as opposed to downloading it from the previous design using an on-board profiler. Now the engineers see the machine profile data unfolding as it happens with a direct connection to their computer.
Although the initial goal was to design a repeatability measurement tool, it now is believed that the software can be used to set up a product for rework. It is an intriguing prospect because of the tool's real-time nature and because of the parameters that are being measured. The peak, time above, time to peak and the average temperature are useful data for setting up an actual process. Therefore, although it is not what was intended, the engineers feel that it is a potential spin-off. Testing this theory is not part of the present agenda because it negates the simplicity of the tool's primary purpose. However, it is an interesting option.
Currently, both the EMS engineers and the vendor company engineers are satisfied that the two teams have developed a tool to streamline rework machine validation, and add necessary SPC. Because of its reliable precision, this tool has cut labor time in diagnostics at the EMS facility by at least 50 percent. This also reduces maintenance time by eliminating the need for troubleshooting. Now, when a machine does need repair, maintenance knows exactly what needs to be done. The tool also saves time and production costs within the rest of the process because it avoids the probability of damaging boards in rework.SMT
*Solectron Corp., Bordeaux, France** Electronic Controls Design Inc. (ECD)
Julien Ratie, process technician, may be contacted at JulienRatie@FR.SLR.COM; +33.05.57.12.78.81. Emmanuel Smague, process engineer, may be contacted at EmmanuelSmague@FR.SLR.com; +33.05.57.12.87.76. Philippe Raout, process engineer, may be contacted at PhilippeRaout@FR.SLR.com; +33.05.57.12.78.81. Mike Hayward, managing director, may be contacted at mike.hayward@ecd.com; +44 1633 768714.