-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueIntelligent Test and Inspection
Are you ready to explore the cutting-edge advancements shaping the electronics manufacturing industry? The May 2025 issue of SMT007 Magazine is packed with insights, innovations, and expert perspectives that you won’t want to miss.
Do You Have X-ray Vision?
Has X-ray’s time finally come in electronics manufacturing? Join us in this issue of SMT007 Magazine, where we answer this question and others to bring more efficiency to your bottom line.
IPC APEX EXPO 2025: A Preview
It’s that time again. If you’re going to Anaheim for IPC APEX EXPO 2025, we’ll see you there. In the meantime, consider this issue of SMT007 Magazine to be your golden ticket to planning the show.
- Articles
- Columns
Search Console
- Links
- Media kit
||| MENU - smt007 Magazine
Parallel Processing: Build It Right the First Time
December 31, 1969 |Estimated reading time: 7 minutes
As the electronics industry moves to smaller and more complex components, and the transition to lead-free materials is evident, the requirement to build it right the first time will be even more crucial to help achieve company financial goals.
By Joe Belmonte and Bob Boyes
There are several metrics used to quantify the performance of an electronics manufacturing line. The generally accepted metrics include process quality, expressed in defects per million defect opportunities; process uptime, which is the percentage of available production time that the process line was used to produce the product; units produced per a particular time period - usually called product cycle time, or throughput, and referred to as line velocity or pulse rate; and first pass yield (FPY). This is the measure of how many of the unit’s products pass the appropriate testing (in-circuit or functional) the first time they are tested.
These metrics seem like simple calculations. While the mathematics of each metric are simple, the SMT industry has no universal standard that states what should or should not be included in each calculation. Therefore, there are many variations from company to company, and even within divisions of the same company, on the content and measurement of these performance metrics.
In the case of FPY, some companies will test completed assemblies as soon as they reach their final operation without including inspection, touch-up or repair. This method provides a true measure of the process line’s performance and identifies the defects that are being experienced. Other companies calculate FPY based on all operations prior to in-circuit and functional testing. This calculation results in higher FPY numbers, but may obscure true line performance and prevents the identification of the defects that are being produced. This difference can cause cost variances reported through the financial system that are not reflected in the FPY metrics. It also can complicate comparing production line performance from one facility or company to another.
Process Throughput
Another common metric in the industry is process cycle time, or how many units are built per day. Generally, you can use the cycle time of each individual piece of process equipment to determine how much of a particular product you should theoretically be able to process each day. However, how many “good products” are produced is a more complex calculation.
The “bottom line” is the only metric that truly measures how well a process line is performing, or how many good products are built each day. Let’s assume that the definition of a good product is one that is built in an acceptable time period and passes all required testing the first time without any inspection, touch-up or repair. We define the rate of how many good products built each day as process throughput.
The focus of all electronic manufacturing operations should be to identify, evaluate and implement methods, procedures and tools that will optimize the amount of “good products” built daily. There is often a balance between product cycle time, or how long it takes a product to get from the beginning to the end of the manufacturing process, and the quality of that product. Producing defective product to achieve some required cycle time is not acceptable; neither is reducing the process speed to an unacceptable rate so as to minimize defects. Manufacturers must understand and optimize this balance.
An important way to reach a balance of cycle time and quality is to understand and optimize all aspects of the process by formal statistical studies. There is no substitute for good engineering work or a well-trained workforce. Process training must consist of more than how to operate a particular machine. Each manufacturing person must understand the entire process and what is required to obtain the best results for their portion of the process. There also is a great deal to be gained by enforcing process discipline - insuring that all policies and procedures are adhered to during each shift - and supporting and coaching process operators in process-improvement programs.
Equipment Features that Reduce Defects
The next area that should be examined to achieve maximum throughput are the tools. What equipment features are available to help maximize product throughput? There are many equipment features that can help to reduce process defects. For example, on sophisticated solder-paste printing equipment, there are 2-D and 3-D post-print inspection systems, stencil wiping and stencil vacuum-cleaning systems, automatic board-support placement systems, board-to-stencil alignment systems and other features that are available on the latest equipment models. Non-primary equipment functions, also called “overhead functions,” are aimed at reducing process defects and improving overall process throughput (Figure 1).
Figure 1. Typical SMT manufacturing line used for high-yield production.
All of these equipment features are effective for what they have been designed to do. The problem with them is that they operate in a serial mode. Therefore, while they are performing their overhead task, the equipment is not performing its primary function - printing solder paste onto a PCB. These equipment overhead features are required to produce as many good products as possible each day, especially when working with complex boards or in high-volume production environments. Because we are losing actual product build time needed for “overhead” equipment functions, we do not see the maximum cycle time benefits of each machine. Ideally, we would like our process equipment to perform some, if not all, of their overhead functions in parallel with the primary machine functions, or perform one or more of the overhead functions simultaneously.
Equipment that provides the necessary overhead functions in parallel will optimize both product cycle time and product throughput. If solder-paste printing equipment could clean a stencil while moving the next product into position and performing the visual-alignment function, we would have two overhead functions performed simultaneously, maximizing both cycle time and quality (Figure 2).
Figure 2. Advanced printer technology improves throughput and yield.
null
Reaching Parallel Processing Capabilities
What is required for solder paste printing equipment to achieve this parallel processing capability? In the past, equipment designers have focused on pure machine velocity. They designed machine functions to perform tasks quickly. Customers have evaluated equipment on cycle time specifications, so the focus became beating the competitors’ advertised cycle times. Little, if any, of the equipment design and development effort was assigned to maximize process throughput as we have defined it. A new equipment-design philosophy should be used when considering process throughput. It is nearly impossible to add parallel processing capabilities into an existing machine design. The equipment must be designed to allow multiple functions to occur simultaneously.
There are many process throughput advantages to equipment that provides parallel processing. Overhead functions that provide quality improvements, such as stencil cleaning and post-print inspection, can be performed more frequently without a reduction in process cycle time. The parallel process capability offers the best of both worlds - producing quality boards without a time penalty.
Equipment selection should include an analysis of process throughput. It is not sufficient to judge process equipment on cycle time alone. Just as it is inappropriate to purchase process equipment on price alone, it also is incorrect to purchase equipment on cycle time specifications alone. First, an industry standard for calculating and reporting machine cycle time should be developed. What seems like the best machine cycle time actually can be misleading upon review of all of the qualifying parameters in calculating that metric. Next, we must understand the true value of a particular piece of process equipment. Value in terms of process throughput should be the predominant factor in selecting process equipment - not price, cycle time or any other factor. Equipment that provides maximum overhead functionality with minimum impact on process cycle time, as well as maximum positive impact on process throughput, has a distinct advantage in maximizing an organization’s profits. As equipment manufacturers pay more attention to the total process throughput when developing new products, customers can enjoy the benefits by improving their overall production potential.
Conclusion
As the electronics industry moves to smaller and more complex components, and the transition to lead-free materials continues, the requirement to build it right the first time will be more critical to achieving a company’s financial goals. The ability to rework and repair these new product designs using newer components with lead-free materials is a significant challenge. Reworking lead-free products may take longer. It also may cause more damage to both components and boards. The most cost-effective solution is to build it right the first time. We must keep evaluating equipment and equipment features to maximize that probability.
Joe Belmonte, project manager, Advanced Process Group, Speedline Technologies may be contacted at (508) 698-7208; e-mail: jbelmonte@speedlinetech.com. Bob Boyes, product manager, MPM, Speedline Technologies may be contacted at (508) 541-6422; e-mail: bboyes@speedlinetech.com.