SMT Trends & Technologies: Optimizing the Flexibility of Your SMT Line


Reading time ( words)

Editor's Note: This article originally appeared in the February 2012 issue of SMT Magazine.

During the recent holiday season I was overwhelmed by the number of products available in an electronics shop. Obviously, buyers exist for each one of these products. Being a technician, I wondered how it was possible to manufacture these different products at affordable prices.

For example, the website of a well-known Asian cell phone brand shows that their product range covers over 130 different models. The average lifespan per cell phone type is approximately 18 months and an average of seven new cell phone types are introduced each month. To avoid risky stocks of unwanted or obsolete electronic end-products, EMS companies need to manufacture all product types on a daily basis. This means multiple changeovers per SMT assembly line per day.

But that’s not all: A huge variation in product quantities will occur over the year. Most people like to find a new cell phone, tablet, gaming system, or computer under their Christmas tree. Many will want to replace their old TV before the Olympic Games in London begin. This customer buying behavior leads to seasonality in manufacturing volumes. Flexibility is the key to solving this production variety.

Three major types of flexibility exist:

  • Type/mix flexibility: Adapting to different types of products over time.
  • Volume flexibility: Adapting to different quantities of products over time.
  • Ramp-up flexibility: Adapting to new product introductions (NPI) over time.

In reality, all three types coincide in the operational flexibility of the SMT flow line. Let’s have a closer look at the changeover activities needed in an SMD pick-and-place machine. First, the PCB transport system settings need adjusting: Board width, board thickness, and PCB support pin positioning. Second, board content settings must be adjusted (feeder setup and nozzles). Third, pick-and-place process-related settings need to be adjusted (pick-and-place program, component vision files, and board alignment settings). Finally, the “old” boards should leave the flow line while the “new” boards enter (after finalizing PCB transport system settings) and the board quality of first “new” boards coming out of the system should be inspected to avoid scrap and rework.

As a result of these actions, the system will be down, or it will work at reduced speed during changeover, reducing the throughput of the SMT flow line. As a result, the capacity of the pick-and-place machines should be increased to meet the quantity of boards needed. For example, the Asian cell phone manufacturer mentioned previously sold approximately 300 million cell phones in 2011. Suppose it manufactures its phones in five plants around the world using SMT flow lines with a placement capacity of 200,000 components per hour, per line. With an average of 400 components per cell phone, 75 flow lines are needed worldwide. This means an average of six changeovers per day, per line. If an average product changeover takes 20 minutes, a total of 6 x 20 = 120 minutes of production time will be lost per day. Throughput will be reduced from 12,000 to 11,000 boards a day. To compensate for this output loss, the installed line capacity needs to be increased by 9% to 218,000 components per hour! Obviously, flexibility has its price.

What can be done to reduce such changeover losses? Most changeover activities can be executed while the machine keeps running: Tape loading into feeders, feeder set-up in trolleys, and placement program preparation. Other changeover activities can be performed only while the machine is stopped: PCB transport system width adjustment, feeder trolley exchange, and nozzle exchange. However, we should ask ourselves: Is it possible to eliminate machine stoppage and perform these changeover activities while the machine is running?

Careful selection of the pick-and-place machine concept can solve this problem. Criteria to bear in mind are:

  • Are sufficient feeder lanes available at the flow line to support multiple product set-ups?
  • Is feeder exchange possible while the machine continues production?
  • Is nozzle exchange possible “on-the-fly?”
  • Can transport width adjustment be performed in small segments (one board length), safeguarding production in most of the flow line?
  • Is instant placement program changeover possible?
  • Can line set-up be verified off-line?

Well selected pick-and-place machines will reduce changeover time to limit output loss. Compare this with a Formula 1 pit stop, where all kinds of measures are taken to minimize time loss!

Figure 1: The fastest changeover: A Formula 1 pit stop.

Let’s return to the Asian cell phone manufacturer. Suppose the company had installed 30 flow lines per plant, each with a capacity of 100,000 components per hour. This would reduce the number of changeovers needed per line to three, reducing output loss (one hour’s loss per 24 hours = 4%). Installing a dual-lane transport system would also give this advantage. As an additional benefit, production would continue on one production lane while changeover actions are performed on the other. This would result in almost zero output loss!

Want to produce a variety of products each day? Flexibility requirements are growing in importance. Careful selection of a pick-and-place machine concept can minimize changeover output losses. It is better to install multiple, small-capacity lines rather than a few large-capacity lines. Dual-lane machines will further reduce changeover output loss.

In addition to playing the clarinet in two bands, Assembléon’s Sjef van Gastel has another passion: SMT. He has been with the company since its start-up as a Philips division in 1979. As the current Manager for Advanced Development, he combines his experience as systems architect and machine designer to explore technical and business opportunities from emerging technologies. van Gastel holds many patents and is a frequent speaker at international conferences related to SMT. He is also the author of “Fundamentals of SMD Assembly,” which has become a standard piece of literature in the industry.

Share



Copyright © 2015 I-Connect007. All rights reserved.