-
- 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
Step 6: Component Placement
December 31, 1969 |Estimated reading time: 8 minutes
Challenges surrounding component placement include asset utilization, production yield, changeover times, floor space constraints, total cost of operation, and new and emerging technologies. This article explores some of the characteristics of various equipment sets that affect component placement.
By Jeffrey L Timms
Over the past 10 years, SMT placement solutions have subdivided into four categories:
- Traditional turret-style chip placers, which are large and reliable but quite inflexible
- Extremely high-volume and inflexible multiple spindle pipette machines
- Accurate but slow gantry-style flexible machines (SMC and odd-form)
- Hard-tooled mass placement solutions.
High volume does not need to mean inflexible. Small, flexible work cells do not necessarily mean low volume. Mixed supplier "best in class" lines do not mean best business practices and results. Today's advanced technologies of flip chip and CSP are becoming mainstream, making the niche-oriented suppliers who propagate a highly specialized approach to the businesses rather impractical. An equipment end-user does not need to live with the configuration of an assembly line simply because that is the way it was purchased and it is too difficult (or expensive) to change. High accuracy does not mean slow tact times. How much accuracy does one really need to provide the component handling characteristics necessary to achieve an acceptable yield?
Figure 1. Speeds reduce on a turret-style chipshooter as feeder size increases.
null
Users have been driven by advanced package technologies, manufacturing operational strategies that impact support logistics and infrastructure requirements, and general economic factors such as total cost of operation, cost of rework due to poor first pass yield, and opportunity cost associated with time-to-market of new products. When combined with the fact that capacity planning has been almost nonexistent and utilization levels typically are less than 40 percent, it is time for a change.
Throughput and Quality Effects
Turret-style chipshooters have inherent problems related to speed and accuracy based on the fundamentals of machine design. As a function of component size, the turret speed and, therefore, placement rate must slow down to accommodate various component and feeder configurations. Though Figure 1 looks troubling, it may be worse than what is represented. The misleading aspect of this representation is in the fact that if a single large component is present on the turret, all other components must be slowed as well. If not, device placement reliability will suffer. Essentially, any other components on the turret are gated by the processing speed of the slowest component. This is further exacerbated by multiple index scenarios required on large-pitch mechanically actuated tape feeders, indicated by the extreme valleys in Figure 1. It also is important to keep in mind that a traditional chipshooter is limited primarily to tape or bulk component input, a critical aspect of flexible placement.
When combined with the peculiarities of whatever fine-pitch solution one has chosen, line balancing can become a significant challenge. Deciding which components to put on which machine and is a key struggle if one is to achieve a truly optimized and balanced line. Combine this with multiple products, and the problem has worsened dramatically. In short, placement speed is component size-dependent and, as such, performance characterization will be non-linear with respect to the relationship between component size and placement speed. This will create significant line balance and optimization complexities.
Turret-style chipshooters also can create component pick-up and placement accuracy issues over the longer term, as assemblies begin to wear. As lead screw and cam surfaces begin to wear in (or wear out), the relationship of tolerances between X-Y table, turret head, theta rotators, carriage lead screws, fiducial recognition capabilities, etc. stack up in a manner that makes target placement or pick-up points less definitive. As a result, pick-up and placement accuracy deteriorates (Figure 2). This is not necessarily a problem on larger components, but when addressing the placement needs of 0402, 0201 and even 01005 passive devices, these relationships become showstoppers.
Figure 2. Old vs. new chipshooter accuracy.
null
Gantry placement solutions, on the other hand, have unique issues. Though not entirely the same as the turret machine, they do have their own version of a derate scenario. To achieve the best possible tact times, one must have a configuration that encourages a "gang pick" scenario. In other words, feeders and components must be located such that the spacing of the pick-up nozzles aligns exactly with the spacing of feeders. One can imagine the impact that this will have on the optimization process and the machine's inherent ability to handle multiple products with a single set-up. As the number of heads and nozzles increase, this characteristic creates even more difficulty in the optimization and line balancing process. Lastly, feeder performance and the number of required camera fields of view for optical component centering will have additional impact on the process. In such a scenario, multiple feeders often are required for the same component part number to realize optimal performance, which will reduce the number of unique component inputs and increase end-user cost significantly.
Pick accuracy is affected by the mechanics of a gantry-style machine. Though in a slow single-pick mode one can achieve fine results, once a product is transferred to the manufacturing floor the inherent limitations of traditional gantry pipette style machines become obvious. Pick accuracy becomes a function of accumulated tolerances and starts with the location of the feeders. In a traditional turret machine, this is influenced by the movement and tolerance of the feeder carriage. In a gantry machine, accurate pick-up when in production speed gang pick mode is influenced by six key factors starting with repeatability of tape location and repeatability of feeder position.
Figure 3. Typical gantry-style placer performance.
null
The next key consideration of pick accuracy is component location within the tape pocket. This is not much of an issue on larger components such as 0603 style but on 0402, 0201 or 01005 devices, the tolerance may be approaching 50 percent of the component body width, requiring different techniques to ensure adequate pickup. The fourth factor affecting pick-up reliability on traditional gantry style machines is the run-out of the spindle itself. The more spindles within the head, the more difficult it is to manage this problem.
The next area of concern affecting pick-up reliability is the parallelism of the head to the feeder pick-up center line. The last key area of concern for pick reliability is related to limitations of unique component types one can place on the machine. This effect can limit machine flexibility by reducing the number of unique components on the machine by up to 60 percent.
null
In spite of the drawbacks, gantry-style machines can offer solid performance in the areas of versatility and placement accuracy. The fact that the circuit board does not move in the X and Y axes during placement minimizes the chance that large fine-pitch or bare die, microBGA or CSP-type devices will be dislodged during the placement cycle. On gantry-style pick-and-place machines, such attributes may be overshadowed by this machine type's optimization and line balancing limitations imposed by "gang pick" scenarios, multiple camera options, multiple head options, multiple required feeder types and more. All these factors must be considered to obtain optimum throughputs. This scenario is further complicated by the complexities created by incompatibilities between software packages of up- and downline machines.
Collect-and-place machine types combine the best of both turret- and gantry-style machines. If done correctly, collect-and-place style modules will exhibit no negative performance characteristics as illustrated in Figure 3 while also providing the highest possible pick-up accuracy, unique component input, placement accuracy and control, odd-form/end-of-line flexibility, and outbound product quality. Additionally, collect-and-place modules have the advantage of providing modular, incremental units of capacity that offer no performance degradation across the full range of component types. The tact time available from today's highly accurate collect-and-place modular solutions ranges from 10 to 60K Cph from a single module. When this level of scalability is combined with the fact that such modular solutions are typically lightweight and easily moved around within the end user's factory without recalibration, factory capacity reconfiguration becomes a reality. On-line flexibility can be further enhanced by implementing multifunction feeders that minimize the total number of feeders required to perform multiple machine set-ups, and ultimately reduce end-user cost (Figure 4).
On a collect-and-place head, often referred to as a revolver head, one key factor influencing the rate at which components can be placed is the turret's diameter on the placement head and the resulting centrifugal force exerted on the component being placed. The larger the diameter, the slower the head must rotate to prevent any negative affects of centrifugal force. It is for this reason that the weight of larger components becomes such a negative influence on larger, traditional chipshooter turret head machines.
Figure 4. Typical collect-and-place performance characteristics.
null
The same factor of size also reduces the mechanical amplification of any tolerance problems associated, thus improving both pick-up and placement accuracies. When combined with the ability to pick a "head-load" of components from a single feeder without slowing, one now can put as many component types on the machine as one has feeder slots without worry of derate.
Fine-pitch placement and advanced technology applications also are served well with auxiliary pick-and-place head, mounted opposite a collect-and-place head, such that grippers, customized pick-up nozzles and extreme placement control can be achieved without encroaching on any of the benefits of a collect-and-place principle machine. The key to this is placing component capacity on one's machine where it is needed and having modules capable of placing the widest variety of components with the highest degree of overlap between modules possible.
Conclusion
Component handling affects asset utilization, production yield, changeover times, floor space constraints, total cost of operation and emerging technologies. State-of-the-art collect-and-place solutions may be an alternative worth considering.
Jeffrey L. Timms, vice president of sales and service, Americas, may be contacted at Siemens Dematic Electronics Assembly Systems Inc., 3140 Northwoods Pkwy., Suite 300, Norcross, GA 30071; (770) 797-3060; Fax: (770) 797-3432; E-mail: jeffrey.timms@siemens.com.