by Ray Prasad, SMT Columnist
April 2004
I hope you had a chance to visit the premier show in our industry, APEX/IPC Printed Circuits Expo in Anaheim, Calif. at the end of February. As usual, many kinds of assembly equipment such as paste printers, pick-and-place, reflow ovens, as well as inspection and rework systems were on display, along with some laser soldering systems for both through-hole and SMT. These subjects will be discussed in both this and my next columns.
Laser soldering has been of interest since the mid-’80s, but more as a curiosity and for niche applications than as a mainstream technology. However, this is about to change. First, the laser technology itself has gone through a tremendous change, from traditional YAG and CO2 lasers to semiconductor technology of diode lasers. Second, many applications exist where laser soldering will be more cost-effective than traditional hand, wave, reflow or hot air soldering methods.
Hand soldering is where it all started. It was then replaced by wave soldering, especially for high-volume applications. Both hand and wave soldering primarily are used only for through-hole components, although a few SMT components can be soldered by these methods. More than 90 percent of components in all electronic products today are SMT, and this number gets closer to 100 percent in some applications.
Hot air and convection reflow soldering methods are used primarily for SMT components, the former for low-volume applications (primarily rework) and the latter for mass production. However, soldering options are complicated by mixed-assembly boards with both SMT and through-hole components.
Through-hole components will remain for many years in electronic products. Reasons include component availability, solder joint reliability and the need for socketability. In a mixed-assembly board, through-hole components can be selectively soldered by three methods:
Laser soldering has been of interest since the mid-’80s, but more as a curiosity and for niche applications than as a mainstream technology. However, this is about to change. First, the laser technology itself has gone through a tremendous change, from traditional YAG and CO2 lasers to semiconductor technology of diode lasers. Second, many applications exist where laser soldering will be more cost-effective than traditional hand, wave, reflow or hot air soldering methods.
Hand soldering is where it all started. It was then replaced by wave soldering, especially for high-volume applications. Both hand and wave soldering primarily are used only for through-hole components, although a few SMT components can be soldered by these methods. More than 90 percent of components in all electronic products today are SMT, and this number gets closer to 100 percent in some applications.
Hot air and convection reflow soldering methods are used primarily for SMT components, the former for low-volume applications (primarily rework) and the latter for mass production. However, soldering options are complicated by mixed-assembly boards with both SMT and through-hole components.
Through-hole components will remain for many years in electronic products. Reasons include component availability, solder joint reliability and the need for socketability. In a mixed-assembly board, through-hole components can be selectively soldered by three methods:
- Wave soldering, with or without specially designed fixtures;
Convection reflow using a paste-in-hole process with specially designed stencil apertures; andHand soldering.
When selective wave soldering and the paste-in-hole process are not technically feasible or too expensive, hand soldering is used as the “last resort.” An example where wave soldering can be used for through-hole, but fixtures that add considerable variable cost are necessary is as follows. Whenever the board’s secondary side contains devices that cannot go through the wave solder process, the through-hole components must be selectively soldered without impact to the adjacent components already soldered in the reflow oven. Examples of components on the bottom side that cannot go through the wave soldering process are any components that have leads on all four sides (PLCC, QFP) or on the package bottom (BGA and CSP).
When the component is temperature-sensitive, has too many rows of pins to allow sufficient solder paste deposition, or the board is not designed for the paste-in-hole process to begin with (e.g., other components may be too close), the paste-in-hole process using convection reflow cannot be used.
To summarize, selective soldering by wave using test fixtures can be expensive, depending on the application, and the paste-in-hole process using convection reflow is not always feasible.
The last resort, hand soldering, is not only slow and expensive, but also operator-dependent, resulting in inconsistent solder joint quality. Additionally, hand-soldered joints are susceptible to external and internal damage.
This is where laser soldering comes in for cost-effective selective soldering of through-hole components when hand, wave or reflow is either not technically feasible or desirable or too expensive. Laser soldering does not need special fixtures that add to variable cost and is less than half as expensive as dedicated selective wave soldering machines with various configurations of solder fountains.
The high cost of dedicated selective wave soldering machines can be justified for low-mix, high-volume applications. However, such high-volume products increasingly are being outsourced to low-wage countries, where even hand soldering can be cost-effective. Based on cost-effectiveness alone, selective laser soldering systems are worth a serious look for many applications, especially relatively lower-volume, high-mix applications where the cost of fixtures can be prohibitive.
The use of expensive, temperature-sensitive components used in the telecommunication industry should also add to the widespread use of laser soldering systems, as will the impending use of lead-free solders. With the deadline for the use of lead-free solder approaching, capital equipment decisions must be made about soldering high melting point lead-free solders. Those companies that want to get away from the hot, messy wave solder pots of traditional or selective soldering systems must take a serious look at laser soldering. Laser soldering systems are versatile, do not need fixtures, and can be cost-effective.
Laser soldering parameters are easy to understand; think small. For example, the variables that control solder joint quality are laser power (8 to 30 W), wave length (800 to 900 nanometers) and time in milliseconds (generally less than a second for a typical solder joint), with or without the use of built-in preheat. In addition to using traditional pre-heaters, the laser itself also is programmed for pre- and post-heat operations (just a few milliseconds) to achieve needed solder joint quality. Because laser soldering time is measured in milliseconds (as opposed to seconds), the intermetallic thickness is under 1 µm (or less than 40 µm-inch), greatly improving solder joint reliability in many applications.
Through-hole components are not the only drivers for selective laser soldering. Surface mount components need to be selectively soldered as well; the subject of my next column.
When the component is temperature-sensitive, has too many rows of pins to allow sufficient solder paste deposition, or the board is not designed for the paste-in-hole process to begin with (e.g., other components may be too close), the paste-in-hole process using convection reflow cannot be used.
To summarize, selective soldering by wave using test fixtures can be expensive, depending on the application, and the paste-in-hole process using convection reflow is not always feasible.
The last resort, hand soldering, is not only slow and expensive, but also operator-dependent, resulting in inconsistent solder joint quality. Additionally, hand-soldered joints are susceptible to external and internal damage.
This is where laser soldering comes in for cost-effective selective soldering of through-hole components when hand, wave or reflow is either not technically feasible or desirable or too expensive. Laser soldering does not need special fixtures that add to variable cost and is less than half as expensive as dedicated selective wave soldering machines with various configurations of solder fountains.
The high cost of dedicated selective wave soldering machines can be justified for low-mix, high-volume applications. However, such high-volume products increasingly are being outsourced to low-wage countries, where even hand soldering can be cost-effective. Based on cost-effectiveness alone, selective laser soldering systems are worth a serious look for many applications, especially relatively lower-volume, high-mix applications where the cost of fixtures can be prohibitive.
The use of expensive, temperature-sensitive components used in the telecommunication industry should also add to the widespread use of laser soldering systems, as will the impending use of lead-free solders. With the deadline for the use of lead-free solder approaching, capital equipment decisions must be made about soldering high melting point lead-free solders. Those companies that want to get away from the hot, messy wave solder pots of traditional or selective soldering systems must take a serious look at laser soldering. Laser soldering systems are versatile, do not need fixtures, and can be cost-effective.
Laser soldering parameters are easy to understand; think small. For example, the variables that control solder joint quality are laser power (8 to 30 W), wave length (800 to 900 nanometers) and time in milliseconds (generally less than a second for a typical solder joint), with or without the use of built-in preheat. In addition to using traditional pre-heaters, the laser itself also is programmed for pre- and post-heat operations (just a few milliseconds) to achieve needed solder joint quality. Because laser soldering time is measured in milliseconds (as opposed to seconds), the intermetallic thickness is under 1 µm (or less than 40 µm-inch), greatly improving solder joint reliability in many applications.
Through-hole components are not the only drivers for selective laser soldering. Surface mount components need to be selectively soldered as well; the subject of my next column.
SMT
Ray P. Prasad is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. and founder of the Ray Prasad Consultancy Group. Contact him at 15375 SW Beaverton Creek Court, Beaverton, OR 97006; (503) 646-3224; Fax: (503) 646-1654;
e-mail: smtsolver@aol.com; Web site: http://www.rayprasad.com/.
PULL QUOTE: “Laser soldering systems are versatile, do not need fixtures and can be cost-effective.”
By Tony Hoult, Circuits Assembly
(February 2004)
(February 2004)
Non-contact selective soldering with high power diode lasers
The trends toward miniaturization in the electronics industry and toward automation in the telecom equipment industry have led to the demand for new, highly controllable selective laser soldering technology in the electronics industry modern high-density electronic and electro-optic subassemblies frequently include delicate heat- and/or debris-sensitive components
By Ray Prasad
(July 2002)
In today’s mixed-assembly board, surface mount components represent the majority of all components. However, it is rare to see complete absence of through-hole components in most assemblies.
In some cases, selective soldering can be avoided by using the paste-in-hole process for both surface mount and through-hole components. However, the paste-in-hole process is not always feasible. This happens when the component is temperature-sensitive, has too many rows of pins to allow sufficient solder paste deposition or the board is not designed for the paste-in-hole process (e.g., other components may be too close).
Whenever the board’s secondary side contains devices that cannot go through wave solder, the through-hole components must be selectively soldered without impacting the adjacent components already soldered in the reflow oven. Examples of components on the bottom side that cannot go through the wave soldering process are any components that have leads on all four sides (PLCC, QFP) or on the package bottom (BGA and CSP).
This adds up to a strong case for co-existence of through-hole and surface mount packages on the same board (mixed assembly). Continued use of through-hole components like sockets and connectors along with fine-pitch and BGAs means that we have no option but to use selective soldering for the foreseeable future. So what are the options for selectively soldering through-hole components in a mixed-assembly board? Following are some commonly used selective soldering options, along with some that are not used currently but may become more popular in the near future.
In some cases, selective soldering can be avoided by using the paste-in-hole process for both surface mount and through-hole components. However, the paste-in-hole process is not always feasible. This happens when the component is temperature-sensitive, has too many rows of pins to allow sufficient solder paste deposition or the board is not designed for the paste-in-hole process (e.g., other components may be too close).
Whenever the board’s secondary side contains devices that cannot go through wave solder, the through-hole components must be selectively soldered without impacting the adjacent components already soldered in the reflow oven. Examples of components on the bottom side that cannot go through the wave soldering process are any components that have leads on all four sides (PLCC, QFP) or on the package bottom (BGA and CSP).
This adds up to a strong case for co-existence of through-hole and surface mount packages on the same board (mixed assembly). Continued use of through-hole components like sockets and connectors along with fine-pitch and BGAs means that we have no option but to use selective soldering for the foreseeable future. So what are the options for selectively soldering through-hole components in a mixed-assembly board? Following are some commonly used selective soldering options, along with some that are not used currently but may become more popular in the near future.
- Today, the use of non-metallic fixtures is the most common method to selectively solder through-hole components in a mixed-assembly board. However, this method works only when the board is designed correctly. Otherwise, the fixture requires quite a few iterations to come up with the final fixture that can only expose the through-hole components and completely hide the surface mount components on the bottom side. This option can be very expensive. Also, it is easy to run out of storage space for fixtures because about half a dozen fixtures are needed for each product.
- This method, generally referred to as solder fountains, uses a metallic fixture that covers the solder pot. The solder comes out like a fountain at the designated locations under the through-hole leads. These fixtures also can be very expensive and take more than a month to design and fabricate. The solder defect levels can be high because the solder fountains change the wave’s flow dynamics.
- Generally known as site-specific soldering in which a robotic carrier moves the board to the solder fountain, this method is becoming common. A variation of this method is a “dancing” wave, in which the board remains stationary but the solder fountain travels to protruding leads. The solder fountain solders each lead or row of leads at a time. These site-specific soldering machines have built-in fluxer, preheater and solder fountain, and tend to simulate the standard wave soldering process. Such machines are flexible and do not use fixtures but are expensive.
-
The next method uses intense focused light to heat the leads and provides solder by wire feeders. The key disadvantage of this process is that allowance must be made at the board design stage to keep solder mask away from the pad. Otherwise, the solder mask close to the through-hole pads tends to burn or discolor. Additionally, such machines are flexible and do not use fixtures but are expensive.
-
An up-and-coming method for selective soldering is a laser that also uses wire feeders. The built-in IR sensors in some of the commercially available systems use high powered diode lasers and can turn off soldering when the joint achieves the desired solder temperature and volume. Some of these systems also can use the paste-in-hole process, eliminating the need for wire feeders. These machines use fiducials for alignment and CAD data for locating the leads and, therefore, do not need fixtures.
The laser can be focused from the top or bottom. Eliminating the need to flip the board before soldering is one advantage of using the laser and wire feed from the bottom. Another advantage is that lasers are environmentally more desirable because they do not require hot metal pots, as is the case with selective soldering machines that use solder pots for soldering — with or without fixtures. Additionally, laser systems can be less expensive than most other selective soldering options. What are the disadvantages? They may be slower. Also, diode laser soldering is a new technology and not everyone feels comfortable being on the leading edge.
What is the best method for selective soldering of though-hole components in a mixed-assembly? It depends on the application and capital budget. To select the right solution for your application, examine your product volume, mix and complexity. And then weave through both technical and business issues to come up with the right answer.
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. in Portland, OR and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax: (503) 297-0330
Web site: http://www.rayprasad.com/.
Pull Quote “Continued use of through-hole components like sockets and connectors along with fine-pitch and BGAs means that we have no option but to use selective soldering for the foreseeable future.”
By Ray Prasad
April 2002
Any money spent on inspection, test and repair is wasted money. It adds only to cost and, hence, adversely impacts the bottom line. It does not add any value. So why do we need to perform these non-value added functions? The reasons are simple. As long as humans are involved, and to err is only human, we are stuck with inspection, test and rework. My modest goal in this column is to focus on ways to reduce the total cost of inspection (namely automated optical inspection [AOI]), test and repair. Of these three non-value added functions, AOI is the only function that can be performed at any point in the line. Test and repair functions generally are performed at the end of the line. Also, AOI and test are somewhat interchangeable. Both these functions accomplish the same goal – finding faults, although their capabilities may vary and they may be performed expensively or cheaply.
April 2002
Any money spent on inspection, test and repair is wasted money. It adds only to cost and, hence, adversely impacts the bottom line. It does not add any value. So why do we need to perform these non-value added functions? The reasons are simple. As long as humans are involved, and to err is only human, we are stuck with inspection, test and rework. My modest goal in this column is to focus on ways to reduce the total cost of inspection (namely automated optical inspection [AOI]), test and repair. Of these three non-value added functions, AOI is the only function that can be performed at any point in the line. Test and repair functions generally are performed at the end of the line. Also, AOI and test are somewhat interchangeable. Both these functions accomplish the same goal – finding faults, although their capabilities may vary and they may be performed expensively or cheaply.
In-circuit vs. Functional Test
Let us take test first. Taking a high-level view, there are two types of tests: in-circuit (also known as bed of nails or automated test equipment [ATE]) test (ICT) and functional test. ICT requires design stage planning so that sufficient test points can be inserted to allow bed-of-nails testing. The beauty of ICT is that it can diagnose the problem up to the component level, eliminating time spent in debugging the failed boards. ICT’s drawback is that the equipment is expensive and the test fixtures for each board can cost anywhere from $10,000 to $40,000. This does not include the cost of developing the test programs. Functional test, on the other hand, does not require any test fixture or elaborate test setup. It simply is a go/no go test performed by plugging the board in the final system. Functional test’s disadvantage comes in not knowing the nature of the problem when the board does fail. Depending on the technician’s skill and experience, it could take anywhere from a few minutes to hours to find the source of the problem. One way to reduce the total cost of AOI, test and repair is to not use ICT and depend entirely on functional test to ensure your customer does not receive faulty boards. However, to enable complete reliance on functional test, your defects must be very low. Now we are back to the interrelationships between inspection, test and repair mentioned earlier. We need to ask the question: “Where do the defects come from and how effective are the AOI systems available on the market?”
Major Defect Sources
Again taking a high-level view, there are three major defect sources, the first being design for manufacture (DFM). There is no AOI or test method that addresses DFM problems. More importantly, there are few companies that even feature in-house DFM.
The blame for defects typically is placed on the assemblers because that is where the defects are discovered, even though the assembly processes have nothing to do with those defects.
The second major source of defects is the quality of incoming materials such as the bare board, components, adhesive, solder paste, flux, etc. Again, like the DFM violations, no AOI or test method can help catch incoming material-related problems.
The third source of defects is the assembly process and equipment used in manufacturing. Let us look at the various manufacturing processes where the defects could come from. Before reflow, they could come from solder paste printing or placement machines. After reflow, they could come from the reflow oven itself. There are AOI systems on the market to perform inspection after any of these processes. If you have the budget, you can put an AOI system after each manufacturing process & after printing, placement and reflow. However, this is not the way to achieve the lowest total cost.
The blame for defects typically is placed on the assemblers because that is where the defects are discovered, even though the assembly processes have nothing to do with those defects.
The second major source of defects is the quality of incoming materials such as the bare board, components, adhesive, solder paste, flux, etc. Again, like the DFM violations, no AOI or test method can help catch incoming material-related problems.
The third source of defects is the assembly process and equipment used in manufacturing. Let us look at the various manufacturing processes where the defects could come from. Before reflow, they could come from solder paste printing or placement machines. After reflow, they could come from the reflow oven itself. There are AOI systems on the market to perform inspection after any of these processes. If you have the budget, you can put an AOI system after each manufacturing process & after printing, placement and reflow. However, this is not the way to achieve the lowest total cost.
AOI After Printing. To settle on the most cost-effective AOI strategy, ask the following general questions: “Which manufacturing process is easier to set up and control without AOI machines and which is not?” Also, do you want to prevent the problem from occurring or do you want to take the traditional approach and wait until the end of the line, inspect the defects and correct them before shipping the product to the customer?
There are only three major process steps in the SMT line & paste printing, component placement and reflow. First, let us look at the printing process. Much had been written, illustrating the import of the printing process. Fortunately, it is the easiest process to control. The solder paste volume in paste printing is an important source of defects but easiest to prevent without spending money on AOI. This is because there are only two kinds of defects in a board that really matter & opens and shorts or bridges. In my view, it is better to have shorts than opens because opens may escape, but shorts always will be caught. A board with bridges will never pass functional test, or any test. Also, shorts or bridges are easier to fix than opens. By designing stencil aperture correctly, and using a reasonable quality printer, you can control paste volume if you focus on getting about eight times more bridges than opens. This strategy also will prevent insufficient and open joints. Monitor shorts to open ratio and keep it at six or more.
There are only three major process steps in the SMT line & paste printing, component placement and reflow. First, let us look at the printing process. Much had been written, illustrating the import of the printing process. Fortunately, it is the easiest process to control. The solder paste volume in paste printing is an important source of defects but easiest to prevent without spending money on AOI. This is because there are only two kinds of defects in a board that really matter & opens and shorts or bridges. In my view, it is better to have shorts than opens because opens may escape, but shorts always will be caught. A board with bridges will never pass functional test, or any test. Also, shorts or bridges are easier to fix than opens. By designing stencil aperture correctly, and using a reasonable quality printer, you can control paste volume if you focus on getting about eight times more bridges than opens. This strategy also will prevent insufficient and open joints. Monitor shorts to open ratio and keep it at six or more.
AOI After Reflow. The reflow oven itself can cause defects. However, if you have taken the time to develop a unique profile for each product or product group, you can be sure that the reflow oven is not going to be your major concern. Also, AOI after reflow may have excessive false accepts and rejects. For example, it is difficult even for humans to decipher the subtle differences between cold solder, dewetting or non-wetting. How can an AOI system be expected to be accurate? Even if it is accurate, keep in mind that solder joint inspection after reflow is an after the fact step. It is too late to rework the defects you find at this stage. It not only adds to the total cost but adversely impacts solder joint reliability.
AOI After Placement. The third major source of defects is component placement. Placement machines have many fast, moving parts. Hence, they are the major source of problems. Additionally, they take long time to set up. It is not uncommon to incorrectly place a wrong reel, tube or tray, especially if the components look alike. But there is good news. Even inexpensive component placement verification AOI machines are effective in finding simple but common mistakes like wrong parts, part presence or absence, or wrong orientation or polarity. The AOI after placement machine is effective in performing complete first article inspection to make sure all feeders are set up correctly, the nozzles have not worn out and the vision system is working properly to place parts correctly. Also keep in mind that simple AOI systems for component verification are relatively inexpensive compared to AOI systems designed to inspect everything.
Because there generally are two or more pick-and-place machines in an SMT line, where should the AOI machine be placed to find the placement errors? If your AOI budget is limited, it may be better to place the system after the final placement machine to catch all placement errors. Additionally, because the last machine in the line places the fine-pitch components, it is important to catch all the fine-pitch-related problems. On a typical board, about half the defects are related to fine-pitch components even though there are only limited numbers of such components on a board.
AOI After Placement. The third major source of defects is component placement. Placement machines have many fast, moving parts. Hence, they are the major source of problems. Additionally, they take long time to set up. It is not uncommon to incorrectly place a wrong reel, tube or tray, especially if the components look alike. But there is good news. Even inexpensive component placement verification AOI machines are effective in finding simple but common mistakes like wrong parts, part presence or absence, or wrong orientation or polarity. The AOI after placement machine is effective in performing complete first article inspection to make sure all feeders are set up correctly, the nozzles have not worn out and the vision system is working properly to place parts correctly. Also keep in mind that simple AOI systems for component verification are relatively inexpensive compared to AOI systems designed to inspect everything.
Because there generally are two or more pick-and-place machines in an SMT line, where should the AOI machine be placed to find the placement errors? If your AOI budget is limited, it may be better to place the system after the final placement machine to catch all placement errors. Additionally, because the last machine in the line places the fine-pitch components, it is important to catch all the fine-pitch-related problems. On a typical board, about half the defects are related to fine-pitch components even though there are only limited numbers of such components on a board.
So where do we go from here? I have nothing against placing an AOI machine after each manufacturing process; but this is not the way to reduce the total cost. In deciding whether to place the AOI system after printing, placement or reflow to minimize cost, an AOI after placement machine is the most cost-effective approach.
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. and founder of the Ray Prasad Consultancy Group. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax (503) 297-0330; Web site: http://www.rayprasad.com/
PULL QUOTE: “In deciding whether to place the AOI system after printing, placement or reflow to minimize cost, an AOI after placement machine is the most cost-effective approach.”
By Ray Prasad
August 2001
Mass rework? Automated? Why would anyone need to do mass rework? Are they totally new to SMT? Haven't companies figured out ways to control their process? After all, the technology has been in very high-volume use since the mid-1980s.
Do we still need to do rework?
In an ideal world there would be no rework. Welcome to the real world. We have been doing through-hole assembly since the mid-1950s. We don't have zero-defect in through-hole assemblies yet. We will never have zero-defect in any assembly - through-hole or SMT, as long as we need humans to do things.
Whatever the reasons, less than 10 percent of companies have first pass yields of more than 90 percent. To put it differently, 90 percent of companies do excessive rework.
What are the reasons for such poor performance? There are three major causes of defects: design for manufacturing (DFM), quality of incoming materials (i.e., adhesive, solder paste, boards and components) and in-house SMT infrastructure (i.e., equipment set up, thermal profile, documentation and training) at all levels in the organization. Training is very important. Nobody gets up in the morning and says, "I am going to screw up three things
at work today."
With the major trend toward outsourcing today, the contract manufacturer (CM) controls less than two thirds of defect causes because, in most cases, the OEM designs the board and is responsible for DFM. Typically even then, the OEM tells the CM who his suppliers are for the component and boards. If the CM is working on a consignment basis, in which all parts and boards are supplied by the OEM, he has control only over one third of the causes of defects. So while the CM generally has no or very little control over DFM and the quality of incoming material, he bears 100 percent responsibility (blame, actually) for the defects.
I think we all can agree that rework is never going to go away. We can minimize it but we cannot eliminate it. It is important to reconcile with this reality so that resources can be put in place to address it. Most people probably agree that we have to deal with rework. But mass rework? That is another matter entirely.
Why Mass Rework ?
I have been in the SMT industry for some time. I have yet to visit a company where mass rework has not been needed. The fortunate thing is that it happens only occasionally. When it does happen, however, it is a major crisis. For example, say 36 components (same component) have to be replaced on each board and you have to rework 5,000 boards. How and why do such things happen? Very simply, it could be that a wrong reel with a different part number was placed with the same types and number of leads. But what is the excuse for building such large quantity assemblies?
One reason the SMT process is so pervasive is that it is very high-speed and, hence, is natural for automation. But also keep in mind that if anything, no matter how trivial, goes wrong it takes almost no time to build thousands of assemblies that must be scrapped before the problem is discovered. If this has not happened to you, you are in denial, extremely lucky or have achieved nirvana.
If you do deal with high-volume rework, you are in big trouble with existing technology because it takes 12 minutes to one hour to remove and replace one component.
That is only one to five components per hour. For 100,000 components, assuming 12 minutes per component, that is 10 years! This is when you need an automated rework system.
Integrating Automated Inspection and Rework
In the May 2001 issue, there is an article by Fred Schlieper1 discussing benefits of integrating inspection and rework. While this article focuses on integrating X-ray and rework systems, the concept is equally valid for integrating automated optical inspection (AOI) with automated rework systems. The inspection system finds the problems and the rework system fixes them automatically. This creates an interesting opportunity for both the suppliers of these machines and the end customer. With the integrated system, problems are fixed with gathered data and corrective action is implemented to prevent future repetition.
There is one minor problem with automated rework systems. Does such a system exist? Before I answer this question, let us agree on the definition of automated rework systems.
In a fully automated rework system, all rework process steps - defective part removal, solder paste or flux application, vision-assisted precision placement of small and large components from tape-and-reel feeders or trays, and reflow without overheating the part being removed or reflowing neighboring components - must be accomplished automatically, without human intervention. And it also would be helpful if the board does not warp because of intense local heating. A tall order, indeed. Does such a system exit?
Automated Rework Systems
There are many rework systems on the market today. They vary in cost from a few hundred dollars to $160,000. These systems are either manual or semiautomated. When I say semi-automated, I basically mean component placement either is totally manual or the operator gets some help in aligning the part for placement. And solder paste dispensing or printing is done manually with a syringe or mini-stencil. A mini-stencil is needed for each part type.
To fully automate rework, not a single machine but an automated assembly system that can selectively dispense solder paste, place components and reflow without heating neighboring components is needed. The system also needs to be fairly fast to assemble quickly and automatically to meet the mass rework requirements. To provide a real relief,
it must be able to selectively assemble many part types without the problem (and the cost that can add up quickly) of changing nozzles and stencils for each part type. There is only one system on the market (at least for now) that can perform all the functions of rework - Spark 400® from BeamWorks.
Before I briefly describe this machine, let me make a public disclosure to avoid any conflict of interest. I have significant interest in BeamWorks. I have been working with BeamWorks since its inception about three years ago and currently I play a major role in its operation. I also am an investor in this company.
To fully automate rework, not a single machine but an automated assembly system that can selectively dispense solder paste, place components and reflow without heating neighboring components is needed. The system also needs to be fairly fast to assemble quickly and automatically to meet the mass rework requirements. To provide a real relief,
it must be able to selectively assemble many part types without the problem (and the cost that can add up quickly) of changing nozzles and stencils for each part type. There is only one system on the market (at least for now) that can perform all the functions of rework - Spark 400® from BeamWorks.
Before I briefly describe this machine, let me make a public disclosure to avoid any conflict of interest. I have significant interest in BeamWorks. I have been working with BeamWorks since its inception about three years ago and currently I play a major role in its operation. I also am an investor in this company.
Spark 400® from BeamWorks is an automated rework system. It performs all the functions of a rework operation - component removal, solder paste dispensing, component placement and laser soldering. Soldering and inspection functions are performed concurrently to provide real-time feedback on solder joint quality. It tailors heat input not only to a specific component but also to a specific lead. In other words, if a certain lead of a given component is connected to power or ground plane, the laser soldering time is longer for that lead than the adjacent lead that is not connected to any heat sink. Think of it as an assembly system that uses a unique solder profile not for a board but rather an individual lead.
Because the machine dispenses paste (or flux as is generally the case because enough solder is left on the pad after component removal), there is no need to spend money on mini-stencils for each component. Paste or flux dispensing, as the case may be, is done either as dots or a thin strip of paste or flux. Then, using the same computer-aided design (CAD) data, the system places components and solders them using four diode lasers.
An important benefit is that the laser does not reflow solder joints of adjacent components even if they are only 0.020" away and does not need any nozzles. Because the board is not heated (only the solder pads are), board warpage is a non-issue. Also, because the heat source for reflow is a diode laser, the moisture-sensitive components already assembled (or to be assembled) do not require baking. This is very important when you want to selectively assemble only certain components.
Having such a system makes rework not the dreaded operation it is thought to be.
Because all operations are performed without human intervention, quality is consistent. Because reflow time is in milliseconds as opposed to minutes as in conventional systems, the intermetallic thickness is less than a micron (compared to 5 mm in conventional processes). Thin intermetallic and complete independence from human variables improve solder joint quality - one key rework concern.
SMT
Reference
1Schlieper, Fred, "Integration of X-ray Inspection and Rework Improves SMT Yields,"
SMT, May 2001, p. 60-64.
SMT, May 2001, p. 60-64.
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure.
Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax (503) 297-0330
Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax (503) 297-0330
Web site: http://www.rayprasad.com/.
Pull Quote: "To fully automate rework, not a single machine but an automated assembly system that can selectively dispense solder paste, place components and reflow without heating neighboring components is needed"
By Ray Prasad
September 2001
September 2001
Despite all effort put into design, process and material control, there always will be some assemblies that fail to meet requirements and must be repaired. This is true in every company. Even with conventional through hole assemblies, which have been around since the 1950s, every company performs some level of repair. Surface mounting is no exception. There is no reason to think that the need for rework/repair will ever go away completely.
In this column, I compare and contrast some of the rework processes for surface mount repair. Before I delve into the details of different rework processes, let me first highlight some major issues, not necessarily in the order of importance, that we need to be aware of in surface mount repair.
In this column, I compare and contrast some of the rework processes for surface mount repair. Before I delve into the details of different rework processes, let me first highlight some major issues, not necessarily in the order of importance, that we need to be aware of in surface mount repair.
Surface Mount Repair Issues
A key issue is to know, on a real-time basis if possible, what is happening to the temperatures of the component and board. This means that an automatic thermal management system is needed that not only monitors package temperature but also controls it within predefined thermal profile parameters.
Interpackage spacing is decreasing constantly. Even if companies have some design for manufacture (DFM) guidelines for interpackage spacing, those on the front line of manufacturing know very well that DFM guidelines are not always followed. So, using mini-stencils to print solder paste is becoming more difficult. Also, because a mini-stencil is needed for each size and type of part, it not only slows down the process but also quickly adds to the cost of repair.
Mini-stencil is not the only issue with ever-decreasing interpackage spacing. Using different hot air nozzles for each size and type of part being removed also adds to the cost and complexity of rework. Additionally, the potential for melting solder joints of neighboring components is a serious concern. In addition to increased intermetallic thickness because of unnecessary reflow, which weakens the solder joints, the boards must be baked before rework, increasing cycle time.
Throughput in rework is very important. Unfortunately, the typical time to remove and replace a simple component like a 32-pin plastic leaded chip carrier (PLCC) can take more than 12 minutes in the conventional process. Ball grid arrays (BGA) and some of the larger components can take at least 20 minutes per component. And some components can take more than an hour for each component. As discussed last month (“Mass Rework? Automated? You must be kidding!,” SMT Magazine, August 2001, p. XX-XX), this can be quite a bottleneck if you have to do any major rework.
Another important issue in rework is board warpage. Warpage partly is due to intense local heating for a relatively long time necessary to remove the component.
For surface mount repair, three rework processes are used today: conductive tools, hot air and laser. Conductive rework tools are the least expensive. Hot air is the most common. The up-and-coming rework process for surface mount repair is laser. Now, let us see how these processes address some of the major rework issues just highlighted.
Interpackage spacing is decreasing constantly. Even if companies have some design for manufacture (DFM) guidelines for interpackage spacing, those on the front line of manufacturing know very well that DFM guidelines are not always followed. So, using mini-stencils to print solder paste is becoming more difficult. Also, because a mini-stencil is needed for each size and type of part, it not only slows down the process but also quickly adds to the cost of repair.
Mini-stencil is not the only issue with ever-decreasing interpackage spacing. Using different hot air nozzles for each size and type of part being removed also adds to the cost and complexity of rework. Additionally, the potential for melting solder joints of neighboring components is a serious concern. In addition to increased intermetallic thickness because of unnecessary reflow, which weakens the solder joints, the boards must be baked before rework, increasing cycle time.
Throughput in rework is very important. Unfortunately, the typical time to remove and replace a simple component like a 32-pin plastic leaded chip carrier (PLCC) can take more than 12 minutes in the conventional process. Ball grid arrays (BGA) and some of the larger components can take at least 20 minutes per component. And some components can take more than an hour for each component. As discussed last month (“Mass Rework? Automated? You must be kidding!,” SMT Magazine, August 2001, p. XX-XX), this can be quite a bottleneck if you have to do any major rework.
Another important issue in rework is board warpage. Warpage partly is due to intense local heating for a relatively long time necessary to remove the component.
For surface mount repair, three rework processes are used today: conductive tools, hot air and laser. Conductive rework tools are the least expensive. Hot air is the most common. The up-and-coming rework process for surface mount repair is laser. Now, let us see how these processes address some of the major rework issues just highlighted.
Conductive Tools for Surface Mount Repair
Conductive tools such as soldering iron tip attachments are used in conjunction with the soldering iron for desoldering components from the substrate. The tip attachment is shaped such that all the solder joints on a particular device are heated simultaneously, and the part is lifted by the tool itself or by other aids such as tweezers.
Attachment size and shape will vary with the size and shape of the part to be removed. Thus, chips and small-outline integrated circuits have a heating attachment on two sides of the package, but PLCCs need an attachment on all four sides. Using the appropriate tip, heat is applied to the surface mount component to be desoldered until all solder melts.
The component then is removed with a twisting motion. There is no preheating mechanism unless an external system is used to preheat the boards.
Conductive tools do increase the potential for lifting pads and board damage. The rework process is very slow and requires the largest interpackage spacing. This method does not address the mini-stencil issue but does not melt solder joints of neighboring components.
Attachment size and shape will vary with the size and shape of the part to be removed. Thus, chips and small-outline integrated circuits have a heating attachment on two sides of the package, but PLCCs need an attachment on all four sides. Using the appropriate tip, heat is applied to the surface mount component to be desoldered until all solder melts.
The component then is removed with a twisting motion. There is no preheating mechanism unless an external system is used to preheat the boards.
Conductive tools do increase the potential for lifting pads and board damage. The rework process is very slow and requires the largest interpackage spacing. This method does not address the mini-stencil issue but does not melt solder joints of neighboring components.
Hot Air Systems for Surface Mount Repair
Hot air systems are either totally manual or semi-automated. Essentially, they are sophisticated hair dryers that blow hot air on the part to be reworked. The part is pulled away from the board when the solder on all joints is molten. The hot air usually is directed on the leads by a nozzle designed specifically for that component. The package body is heated by the hot air impinging on the package and conduction within the package. Initially, the package is preheated with the nozzle some distance away (typically 1" or more) from the package body. Then the nozzle is lowered to a point just above the package body and lead temperature increases sharply until it reaches a peak. During this process of blowing hot air, the solder joints of neighboring components even half an inch away can reflow, an unwanted and undesirable result.
After the component is removed, paste application for reattachment is a most difficult and time-consuming process. Typically, a mini-stencil is used to apply the paste. Both hot air nozzles and mini-stencils are needed for each type and size of part being reworked. Both these items require sufficient interpackage spacing for rework.
After the component is removed, paste application for reattachment is a most difficult and time-consuming process. Typically, a mini-stencil is used to apply the paste. Both hot air nozzles and mini-stencils are needed for each type and size of part being reworked. Both these items require sufficient interpackage spacing for rework.
Laser Systems for Surface Mount Repair
Up-and-coming repair systems use from one to four lasers. Some of the laser systems are limited to reworking only peripheral components,
in which the leads are in the laser’s line of sight. However, the higher end laser systems that use multiple diode lasers can rework both peripheral and array type packages such as BGAs, chip scale packages (CSP) and flip chips by rapidly scanning top of package surfaces. This causes BGA/CSP/flip chip ball reflow underneath by conduction through the package, as is the case in hot air rework. Some of these high-end systems also have a built-in automated thermal management capability to monitor and control package temperatures within the specified limits to prevent overheating.
Unlike hot air systems, with their built-in temperature monitoring systems, the higher end laser rework systems can remove components, dispense paste, place and solder components automatically without nozzles or mini-stencils.
in which the leads are in the laser’s line of sight. However, the higher end laser systems that use multiple diode lasers can rework both peripheral and array type packages such as BGAs, chip scale packages (CSP) and flip chips by rapidly scanning top of package surfaces. This causes BGA/CSP/flip chip ball reflow underneath by conduction through the package, as is the case in hot air rework. Some of these high-end systems also have a built-in automated thermal management capability to monitor and control package temperatures within the specified limits to prevent overheating.
Unlike hot air systems, with their built-in temperature monitoring systems, the higher end laser rework systems can remove components, dispense paste, place and solder components automatically without nozzles or mini-stencils.
Even though there are various laser systems on the market for rework, to avoid conflict of interest, I want to disclose that I have financial interest in one of the higher end laser rework systems made by BeamWorks. I am not only an investor in this company, but I also play a major role in its operation as my bio at the end of this column clearly indicates.
Hot Air or Laser?
Interpackage spacing on surface mount boards is much smaller than that encountered on conventional boards. Because the laser beam is very narrow, components even 0.020" (20 mil) away do not experience any heat. Laser systems heat the package without melting the solder joints of neighboring components while keeping the board cool enough to touch.
With hot air systems, melting solder joints of neighboring components is a serious concern. Using correct hot air nozzles and shielding adjacent components can minimize this problem. However, hot air nozzles are needed for each type and size of component. This adds to the cost and complexity of repair. And to shield adjacent components, adequate interpackage spacing is necessary, which is becoming difficult to come by because designers constantly reduce interpackage spacing as they try to pack more components into ever-smaller electronics devices. Even with shields, there is no way to completely prevent reflow of neighboring components because some air does leak out.
Adjacent component reflow is not the only concern with hot air. The entire board must be baked for up to 24 hours before rework to prevent popcorning of both the part being reworked and neighboring components. This adds to the cycle time of rework. Because lasers only heat the lead, baking the board before rework is not necessary unless a part needs to be salvaged. This generally is not a major concern because most companies scrap reworked parts.
With hot air systems, melting solder joints of neighboring components is a serious concern. Using correct hot air nozzles and shielding adjacent components can minimize this problem. However, hot air nozzles are needed for each type and size of component. This adds to the cost and complexity of repair. And to shield adjacent components, adequate interpackage spacing is necessary, which is becoming difficult to come by because designers constantly reduce interpackage spacing as they try to pack more components into ever-smaller electronics devices. Even with shields, there is no way to completely prevent reflow of neighboring components because some air does leak out.
Adjacent component reflow is not the only concern with hot air. The entire board must be baked for up to 24 hours before rework to prevent popcorning of both the part being reworked and neighboring components. This adds to the cycle time of rework. Because lasers only heat the lead, baking the board before rework is not necessary unless a part needs to be salvaged. This generally is not a major concern because most companies scrap reworked parts.
SMT
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is President of BeamWorks Inc. and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax: (503) 297-0330;
e-mail: smtsolver@rayprasad.com; Web site: http://www.rayprasad.com/.
By Ray Prasad
February 2001
February 2001
From the title of this column, you might find it strange that someone with a reasonable understanding of the SMT industry would even suggest that quality and delivery could improve by building assemblies with missing components. Of course, it is always better to have all components on hand before assembly but there are times when this is not the case and one must deal with missing component issues.
For a change, this column is not about materials and process technology but about production planning and scheduling issues as they relate to missing components. This month's column will discuss how to reduce cycle time and improve quality by building assemblies even with missing components. This unusual recommendation is based on a recently available innovative assembly system.
In our industry there are two key manufacturing concerns - quality and delivery.
This is especially true in contract manufacturing (CM). Problems arise either because the product is late or there is some complaint about product quality.
Sometimes there are complaints about both quality and delivery.
There are many reasons for quality and delivery problems. It depends on whether you are the OEM or the subcontractor. In an OEM's view, the quality and delivery problems primarily are because of poor process control and production planning by the subcontractor.
On the other hand, if you are the subcontractor, it really is not your fault. The quality problems are because of the poor design for manufacturing (DFM) used by the OEM.
And the reason the subcontractor has a hard time delivering products on time is because of repeated engineering change notices (ECN) and poor product volume forecasting by the OEM. This makes it difficult to order enough parts in time without incurring excessive inventory carrying cost. And being restricted to buying parts, especially long lead parts, only from suppliers approved by the OEM only compounds the problem.
In my consulting practice, I find that both the OEMs and the subcontractors are partially right and the truth lies somewhere in between. However, a main reason for delivery problems is component availability before assembly. Very often there are missing components before assembly.
Note that quality and delivery problems are not unique to subcontractors.
OEMs building in-house boards face the same problems. So, as the production manager responsible for shipping the product, whether an OEM or a subcontractor, what options are available if components are missing? Should boards be partially assembled or should assembly wait until all components arrive?
Waiting is a gamble because it is unknown when those missing components will arrive. Instead, maybe the production manager will partially build the assemblies with missing parts and then hand solder them when the missing components arrive. But this option does not provide relief either because hand soldering will not only increase cost but also will reduce product quality. So, as a production manager, there really is not a good option to keep customers satisfied.
Of course, it would be better if all the components were available and no one had to deal with missing components. But let's be realistic. As the people on the front line responsible for manufacturing know, no matter how perfect production planning and component ordering systems, there are times when components will not arrive as scheduled. This is when the lack of options comes into play. That is, until now. This brings me back to where I started how to build assemblies with missing components while reducing cycle time and improving quality.
The key reason to wait for all components before proceeding with the assembly is the need to print solder paste for all components and solder them simultaneously in the convection reflow oven. If components, which happen to be fine-pitch components with high pin counts in a high-volume line, were missing, numerous skilled operators would be needed to meet the production demand. Even then, rework would be high, compounding quality and delivery problems.
For a change, this column is not about materials and process technology but about production planning and scheduling issues as they relate to missing components. This month's column will discuss how to reduce cycle time and improve quality by building assemblies even with missing components. This unusual recommendation is based on a recently available innovative assembly system.
In our industry there are two key manufacturing concerns - quality and delivery.
This is especially true in contract manufacturing (CM). Problems arise either because the product is late or there is some complaint about product quality.
Sometimes there are complaints about both quality and delivery.
There are many reasons for quality and delivery problems. It depends on whether you are the OEM or the subcontractor. In an OEM's view, the quality and delivery problems primarily are because of poor process control and production planning by the subcontractor.
On the other hand, if you are the subcontractor, it really is not your fault. The quality problems are because of the poor design for manufacturing (DFM) used by the OEM.
And the reason the subcontractor has a hard time delivering products on time is because of repeated engineering change notices (ECN) and poor product volume forecasting by the OEM. This makes it difficult to order enough parts in time without incurring excessive inventory carrying cost. And being restricted to buying parts, especially long lead parts, only from suppliers approved by the OEM only compounds the problem.
In my consulting practice, I find that both the OEMs and the subcontractors are partially right and the truth lies somewhere in between. However, a main reason for delivery problems is component availability before assembly. Very often there are missing components before assembly.
Note that quality and delivery problems are not unique to subcontractors.
OEMs building in-house boards face the same problems. So, as the production manager responsible for shipping the product, whether an OEM or a subcontractor, what options are available if components are missing? Should boards be partially assembled or should assembly wait until all components arrive?
Waiting is a gamble because it is unknown when those missing components will arrive. Instead, maybe the production manager will partially build the assemblies with missing parts and then hand solder them when the missing components arrive. But this option does not provide relief either because hand soldering will not only increase cost but also will reduce product quality. So, as a production manager, there really is not a good option to keep customers satisfied.
Of course, it would be better if all the components were available and no one had to deal with missing components. But let's be realistic. As the people on the front line responsible for manufacturing know, no matter how perfect production planning and component ordering systems, there are times when components will not arrive as scheduled. This is when the lack of options comes into play. That is, until now. This brings me back to where I started how to build assemblies with missing components while reducing cycle time and improving quality.
The key reason to wait for all components before proceeding with the assembly is the need to print solder paste for all components and solder them simultaneously in the convection reflow oven. If components, which happen to be fine-pitch components with high pin counts in a high-volume line, were missing, numerous skilled operators would be needed to meet the production demand. Even then, rework would be high, compounding quality and delivery problems.
Addressing Industry Concern
What is needed is an automated selective assembly system (not a single machine but an assembly system) that can selectively dispense solder paste, place components and reflow solder the missing components without impacting already assembled components. The system also must automatically assemble and be fairly fast to meet the needs of
a high-volume line. Additionally, it must selectively assemble many part types because typically there are many components missing. Such a machine to selectively assemble surface mount components did not exist. That is, until last month.
Those who attended APEX last month probably saw the automated selective assembly system Spark 400 by BeamWorks. Before I briefly describe this machine, let me make a public disclosure to avoid any conflict of interest. I have significant interest in BeamWorks. I have been working with the company since its inception about three years ago and currently play a major role in its operation.
Spark 400 from BeamWorks is an automated selective assembly system that performs all functions of a typical SMT line: solder paste dispensing, component placement, laser soldering and inspection. Soldering and inspection functions are performed concurrently to provide real-time feedback on solder joint quality. This is the only system I know that automatically stops soldering when desired solder joint quality is achieved. It tailors heat input to a specific component as well as to a specific lead. If a certain lead of a given component is connected to power or ground plane, the laser soldering time is longer for that lead than the adjacent lead not connected to a heat sink.
Because the machine dispenses paste, a partially assembled board does not pose any problems as it would if a stencil were used. It does not require separate programming for dispensing, placement and soldering. Paste dispensing is done either as dots (as would be the case for array packages) or a thin strip of paste (as would be the case for peripheral components). Then, using the same computer-aided design (CAD) data, it places components and quickly solders them using four diode lasers. What is really important is that the selective assembly of missing surface mount components does not affect in any way the components that already have been assembled on the conventional SMT line. Also, because the heat source for reflow is a diode laser, the moisture-sensitive components already assembled (or to be assembled) do not require baking.
This is important to selectively assemble only certain components.
Having such a system makes production planning a lot easier for the production manager responsible for meeting delivery dates. For example, the production plan does not have to be reshuffled for the SMT line if some components do not make it on time because assembly can proceed with one or more missing components. Additionally, this machine makes the best use of the expensive equipment in the SMT line. Any time the line stops because of last minute changes in production scheduling, due to missing components, can be very expensive. Proper machine utilization is key to making money, especially in a very competitive industry with low profit margins.
The BeamWorks machine provides flexibility to the production manager. When those missing components arrive, they can be soldered off-line or in-line without impacting the current production plan. This will improve delivery because partially assembled boards with missing components can be assembled quickly and automatically. Additionally, the delivery dates of other products with or without missing components will be improved because they will not be affected by the chain reaction of missing components in one product impacting the delivery dates of other products.
Missing components have been one reason for quality and delivery problems in manufacturing. Using the selective automated assembly system from BeamWorks provides a cost-effective system for addressing this common industry problem.
a high-volume line. Additionally, it must selectively assemble many part types because typically there are many components missing. Such a machine to selectively assemble surface mount components did not exist. That is, until last month.
Those who attended APEX last month probably saw the automated selective assembly system Spark 400 by BeamWorks. Before I briefly describe this machine, let me make a public disclosure to avoid any conflict of interest. I have significant interest in BeamWorks. I have been working with the company since its inception about three years ago and currently play a major role in its operation.
Spark 400 from BeamWorks is an automated selective assembly system that performs all functions of a typical SMT line: solder paste dispensing, component placement, laser soldering and inspection. Soldering and inspection functions are performed concurrently to provide real-time feedback on solder joint quality. This is the only system I know that automatically stops soldering when desired solder joint quality is achieved. It tailors heat input to a specific component as well as to a specific lead. If a certain lead of a given component is connected to power or ground plane, the laser soldering time is longer for that lead than the adjacent lead not connected to a heat sink.
Because the machine dispenses paste, a partially assembled board does not pose any problems as it would if a stencil were used. It does not require separate programming for dispensing, placement and soldering. Paste dispensing is done either as dots (as would be the case for array packages) or a thin strip of paste (as would be the case for peripheral components). Then, using the same computer-aided design (CAD) data, it places components and quickly solders them using four diode lasers. What is really important is that the selective assembly of missing surface mount components does not affect in any way the components that already have been assembled on the conventional SMT line. Also, because the heat source for reflow is a diode laser, the moisture-sensitive components already assembled (or to be assembled) do not require baking.
This is important to selectively assemble only certain components.
Having such a system makes production planning a lot easier for the production manager responsible for meeting delivery dates. For example, the production plan does not have to be reshuffled for the SMT line if some components do not make it on time because assembly can proceed with one or more missing components. Additionally, this machine makes the best use of the expensive equipment in the SMT line. Any time the line stops because of last minute changes in production scheduling, due to missing components, can be very expensive. Proper machine utilization is key to making money, especially in a very competitive industry with low profit margins.
The BeamWorks machine provides flexibility to the production manager. When those missing components arrive, they can be soldered off-line or in-line without impacting the current production plan. This will improve delivery because partially assembled boards with missing components can be assembled quickly and automatically. Additionally, the delivery dates of other products with or without missing components will be improved because they will not be affected by the chain reaction of missing components in one product impacting the delivery dates of other products.
Missing components have been one reason for quality and delivery problems in manufacturing. Using the selective automated assembly system from BeamWorks provides a cost-effective system for addressing this common industry problem.
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax (503) 297-0330;
Web site: http://www.rayprasad.com/
By Ray Prasad
(January 2001 SMT Magazine)
Let me first take this rare, rather once in a lifetime, opportunity to wish all of you, the readers of this column, a Happy New Millennium, the start of the 21st century. This also is the time for many of you to think about Sunny San Diego. I am not talking about the boardwalk and the sandy beaches, although they are certainly very pleasant, but APEX, the premier trade show of our industry. Last year’s APEX show in Long Beach was a resounding success by any measure even though it was its first year. The early indications are that it will be an even better trade show this year.
People come to trade shows for many reasons: to attend professional tutorials and workshops, to participate in technical sessions, and to see the latest in assembly technology and equipment. Numerous suppliers will introduce new equipment in paste printing, pick-and-place, reflow soldering, inspection, and rework. In this column, I will preview a laser assembly machine (not just a laser soldering machine) made by BeamWorks that performs all assembly functions (no, it does not make coffee).
Before I describe this machine, let me make a public disclosure to avoid any conflict of interest. I have consulted for BeamWorks since the company’s inception about three years ago and now have significant involvement in the company’s operation. Rest assured, however, that as in my past columns, I will give a balanced technical opinion whether I am working with a given company or not.
(January 2001 SMT Magazine)
Let me first take this rare, rather once in a lifetime, opportunity to wish all of you, the readers of this column, a Happy New Millennium, the start of the 21st century. This also is the time for many of you to think about Sunny San Diego. I am not talking about the boardwalk and the sandy beaches, although they are certainly very pleasant, but APEX, the premier trade show of our industry. Last year’s APEX show in Long Beach was a resounding success by any measure even though it was its first year. The early indications are that it will be an even better trade show this year.
People come to trade shows for many reasons: to attend professional tutorials and workshops, to participate in technical sessions, and to see the latest in assembly technology and equipment. Numerous suppliers will introduce new equipment in paste printing, pick-and-place, reflow soldering, inspection, and rework. In this column, I will preview a laser assembly machine (not just a laser soldering machine) made by BeamWorks that performs all assembly functions (no, it does not make coffee).
Before I describe this machine, let me make a public disclosure to avoid any conflict of interest. I have consulted for BeamWorks since the company’s inception about three years ago and now have significant involvement in the company’s operation. Rest assured, however, that as in my past columns, I will give a balanced technical opinion whether I am working with a given company or not.
Addressing Concerns
This system is attempting to address some key technical concerns in our industry. The first concern is selective assembly of through-hole components such as sockets and connectors. These problems will not dissipate because through-hole components will remain in the SMT industry for many years. There are many reasons for this, but that is content for another column.
In a mixed assembly, through-hole components are soldered selectively by three methods:
• Wave soldering using a specially designed fixture
• Paste-in-hole process
• Hand soldering.
When selective wave soldering and paste-in-hole process are not technically feasible, as typically is the case, hand soldering is used as the “last resort” process. Hand soldering is not only slow and expensive, but also operator-dependent, resulting in inconsistent solder joint quality. Additionally, hand soldered joints are susceptible to external and internal damage. The internal damage is a serious concern because the solder joint appears fine on the board’s surface but would be highly unreliable even if it passed electrical tests.
Through-hole components are not the only drivers for selective soldering. With increased cell phone use, selective RF shield soldering is becoming very common. While some RF shields can be reflow soldered in convection ovens, many still need to be soldered by hand. Even shields that can be soldered in convection ovens require hand soldering for removal to allow rework of components shielded by them. Additionally, there are many surface mount connectors, especially those with leads on each side of the board, that need to be selectively soldered.
Furthermore, moisture-sensitive devices must be baked before soldering to prevent package cracking, commonly referred to as popcorning. Reworking an expensive device that must be salvaged is especially painful. To salvage an individual component, the entire assembly must be baked for 24 hours to seven days. The potential for localized board damage is increased greatly when removing large components such as fine pitch, plastic leaded chip carriers (PLCC), ball grid arrays (BGA) and small components such as chip scale packages (CSP). This problem will get worse before it gets better when (and if) lead-free solders must be used because most have very high melting points.
In a mixed assembly, through-hole components are soldered selectively by three methods:
• Wave soldering using a specially designed fixture
• Paste-in-hole process
• Hand soldering.
When selective wave soldering and paste-in-hole process are not technically feasible, as typically is the case, hand soldering is used as the “last resort” process. Hand soldering is not only slow and expensive, but also operator-dependent, resulting in inconsistent solder joint quality. Additionally, hand soldered joints are susceptible to external and internal damage. The internal damage is a serious concern because the solder joint appears fine on the board’s surface but would be highly unreliable even if it passed electrical tests.
Through-hole components are not the only drivers for selective soldering. With increased cell phone use, selective RF shield soldering is becoming very common. While some RF shields can be reflow soldered in convection ovens, many still need to be soldered by hand. Even shields that can be soldered in convection ovens require hand soldering for removal to allow rework of components shielded by them. Additionally, there are many surface mount connectors, especially those with leads on each side of the board, that need to be selectively soldered.
Furthermore, moisture-sensitive devices must be baked before soldering to prevent package cracking, commonly referred to as popcorning. Reworking an expensive device that must be salvaged is especially painful. To salvage an individual component, the entire assembly must be baked for 24 hours to seven days. The potential for localized board damage is increased greatly when removing large components such as fine pitch, plastic leaded chip carriers (PLCC), ball grid arrays (BGA) and small components such as chip scale packages (CSP). This problem will get worse before it gets better when (and if) lead-free solders must be used because most have very high melting points.
Laser Assembly Machine
The key concerns in today’s electronics assembly are selective assembly of through-hole components, missing and moisture-sensitive surface mount components, and lead-free soldering, and soldering of RF shields. The BeamWorks laser assembly machine’s ability to selectively assemble any component addresses these technical concerns. It not only is a laser soldering machine but also an assembly machine, performing all the assembly functions from paste dispensing to inspection and rework. This machine dispenses solder paste with positive force displacement to accurately control paste volume. Then various pick-up heads pick up components from as small as 0402 to as large as BGAs and fine pitch. Multiple diode lasers selectively or simultaneously solder in milliseconds. Because laser cannot see BGA, CSP and flip chip balls, the laser beam width is increased to simulate an oven profile to reflow hidden solder balls. The system also can perform rework and inspection functions. The need for programming is eliminated because computer-aided design (CAD) data is used for all assembly functions: paste dispensing, placement, soldering and rework. Components such as fine pitch do not have to be baked because the package does not get hot, only the leads do.
The machine can be used for both high-volume and low-volume applications. For example, if it is used only as a soldering machine (and I do not see why it should be), multiple lasers can keep up with the high-volume pick-and-place machine in the line. Also, it can be used to selectively assemble components in a high-volume line. Examples include missing components in an already assembled board or an RF shield in a high-volume line. In low-volume applications, it can assemble all components in quick-turn applications. If the machine is used for every assembly processes; however, it naturally will be slower. In prototype and pilot builds, volume is not a concern, but quick turn around time is.
While this machine can be used for low-volume, quick-turn applications, it is well suited for unique applications where conventional processes either are not technically feasible or are too expensive. Additionally, because reflow time for each joint can be individually programmed, the heat input for each joint can be precisely tailored to meet the needs of joints connected to power and ground planes or heat sinks.
Because this machine measures soldering time in milliseconds (as opposed to seconds), the intermetallic thickness is under 1 µm (or less than 40 µ". Compare this to 500 µ" or more in other soldering methods), greatly improving solder joint reliability.
This machine is not a panacea for all our ills – but nothing is. It does, however, meet some unique needs and, while it cannot be used to replace existing equipment, it can be used as a supplement.
Those who would like to see the machine in operation can do so at APEX 2001 where it will be unveiled to the world for the first time.
Hope to see you in sunny San Diego!
The machine can be used for both high-volume and low-volume applications. For example, if it is used only as a soldering machine (and I do not see why it should be), multiple lasers can keep up with the high-volume pick-and-place machine in the line. Also, it can be used to selectively assemble components in a high-volume line. Examples include missing components in an already assembled board or an RF shield in a high-volume line. In low-volume applications, it can assemble all components in quick-turn applications. If the machine is used for every assembly processes; however, it naturally will be slower. In prototype and pilot builds, volume is not a concern, but quick turn around time is.
While this machine can be used for low-volume, quick-turn applications, it is well suited for unique applications where conventional processes either are not technically feasible or are too expensive. Additionally, because reflow time for each joint can be individually programmed, the heat input for each joint can be precisely tailored to meet the needs of joints connected to power and ground planes or heat sinks.
Because this machine measures soldering time in milliseconds (as opposed to seconds), the intermetallic thickness is under 1 µm (or less than 40 µ". Compare this to 500 µ" or more in other soldering methods), greatly improving solder joint reliability.
This machine is not a panacea for all our ills – but nothing is. It does, however, meet some unique needs and, while it cannot be used to replace existing equipment, it can be used as a supplement.
Those who would like to see the machine in operation can do so at APEX 2001 where it will be unveiled to the world for the first time.
Hope to see you in sunny San Diego!
SMT
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. in Portland, OR and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax: (503) 297-0330;
Web site: http://www.rayprasad.com/.
Pull Quote: “The key concerns in today’s electronics assembly are selective assembly of through-hole components, missing and moisture-sensitive surface mount components, lead-free soldering, and soldering of RF shields. The BeamWorks laser assembly machine’s ability to selectively assemble any component addresses these technical issues.”
By Ray Prasad
(February 1995 CMPONENT.SMT )
(February 1995 CMPONENT.SMT )
Although SMT is widely used in commercial and military applications (almost 90 percent of the components on PCBs are surface mount), many are still using through-hole components. However, through-hole and surface mount components on the same board increases complexity, size and cost. As most manufacturers know, having even a few through-hole components on a board requires conventional insertion, soldering, cleaning and rework processes in addition to surface mount process steps. Also, through-hole components add to pin inductance and capacitance. There are some reasons why designers still use through-hole packages.
The reasons for using through-hole components vary for different devices such as sockets, connectors and microprocessors. Sockets and connectors will continue to be more widely used in through-hole than in surface mount because of higher mechanical strength requirements. A through-hole joint provides much more mechanical integrity than a surface mount joint. Larger devices, such as connectors, need a great deal of mechanical integrity.
There are various packaging options for microprocessors. As the processor frequency increases, so do demands on electrical and thermal package performance. Increased electrical performance requires increased package layers and pin count.
The options for packaging microprocessors in surface mount are fine-pitch (0.65 and 0.5 mm or lower) surface mount package, TAB and BGA. With these packages, there is considerable yield loss during package handling and test due to the fragility of fine-pitch leads. Most defects during board assembly are experienced in fine-pitch devices for this reason. As the lead-pitch is reduced, defects increase. For this reason, the PowerPC in ultrafine-pitch (0.4 mm pitch) and Intel Pentium in TAB (0.25 mm pitch for notebook computers) may not become widely accepted by the user community. Currently, high volume manufacturing capability for bare boards and assembly for ultra fine-pitch packages does not exist in most companies. These packages are too complex for widespread usage.
Despite some inherent problems such as inspection and repair, BGA offers advantages over fine-pitch or TAB. Its rigid solder bumps solve the problems caused by the fragile leads of fine-pitch packages, and with an overall yield improvement. Instead of spending time and money on a fine-pitch infrastructure, resources might be better spent on solving the issues of a promising package like BGA.
In through-hole, PGA is most widely used for microprocessor packaging. The main disadvantage of PGA is its larger size. However, for high wattage devices, board or system space requirement is determined by the heat sink size. Hence the larger size of PGA really does not matter since large heat sinks are needed in every package. In addition, a ceramic PGA easily incorporates electrical (in-package capacitance) and thermal enhancements required by high performance devices. In-package capacitance is difficult to incorporate in a surface mount package.
Unlike ultra-fine-pitch and BGA packages, PGA has been used for a long time. Also, PGA can be reliably socketed (and soldered). For various reasons, socketing is preferred by OEMs and users. There is a strong need for socketing the most expensive component in a typical system. For example, socketing allows OEMs to design a common mother board that will accept all different frequencies of a given microprocessor with common pinout. Socketing allows upgrades by end users who do not have to buy a whole new system in order to take advantage of higher performance of future processors. For example, users can easily upgrade to a higher frequency processor if it is socketed as is the case for Intel 486SX, DX 33 and DX266. A socket is also handy if a processor needs to be replaced because of any other reason as was the case with Intel Pentium.
Perhaps the main reason for using PGA is that users of this package type enjoy the advantage of lower import duty on motherboards assembled overseas. Since import duty is generally based on value of products being imported, lower duty is levied on a board with sockets for processors. For these reasons, many users even go to the trouble of converting fine-pitch devices into PGAs by using interposer cards.
With the recent passage of GATT, the tariff on imports will be reduced. Although GATT may reduce this incentive for using socketed parts, some incentive will remain since the tariff will not be entirely eliminated. Besides, OEMs do not necessarily like the idea of tying up millions of dollars worth of inventory for any longer period than required. Socketing allows them the freedom to purchase the processor just before shipping to the end customers.
All this adds up to a strong case for continued use of PGAs along with predominantly surface mount packages on the same board. Surface mount packages, including BGA and fine pitch, will not be as widely used as the old stand by through-hole PGA for high pin count, high performance applications such as microprocessors for some time. The continued use of PGAs, through-hole sockets and connectors means that the industry will have to wrestle with mix-and-match assemblies for the foreseeable future. So, even though surface mount is the wave of the future, do not get rid of your hand soldering, point-to-point soldering or wave soldering machines yet.
RAY P. PRASAD is an SMT Editorial Advisory Board member and author of the textbook Surface Mount Technology: Principles and Practice. Additionally, he is president of BeamWorks Inc. in Portland, OR and founder of the Ray Prasad Consultancy Group, which specializes in helping companies establish strong internal SMT infrastructure. Contact him at P.O. Box 219179, Portland, OR 97225; (503) 297-5898 or (503) 646-3224; Fax: (503) 297-0330
Web site: http://www.rayprasad.com/.
PULL QUOTE
“The reasons for using through-hole components vary for different devices.”
“The reasons for using through-hole components vary for different devices.”
