-
- News
- Books
Featured Books
- smt007 Magazine
Latest Issues
Current IssueIPC 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.
The Cost of Rework
In this issue, we investigate rework's current state of the art. What are the root causes and how are they resolved? What is the financial impact of rework, and is it possible to eliminate it entirely without sacrificing your yields?
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - smt007 Magazine
STEP 9: Test & Inspection
December 31, 1969 |Estimated reading time: 6 minutes
Fast-evolving technology on circuit boards and quick-turn production requirements place new demands on circuit board test strategies and platforms. A dramatic shift in the fault spectrum, new test technologies, and the rise of in-system programming (ISP) at in-circuit test pose major challenges to a test strategy based on traditional “big-iron” test platforms.
In-circuit test (ICT) has been around for a long time, but remains a cost-efficient way to ensure that the manufacturing process produces high-quality PCB assemblies (PCBAs). ICT has been around long enough that test managers consider it a “solved problem.” But the variables that determined test strategies a few years ago are strikingly different now, making ICT an unsolved problem. Some challenges include:
- Devices have evolved from digital ICs to system-on-chip (SOC) and emerging system-in-package (SIP) technologies.
- New component technologies, package types, and assembly processes have dramatically shifted the fault spectrum-the typical distribution of PCB defects.
- Test technologies such as vectorless test and boundary scan are routine.
- ICT is the preferred point in the assembly process to program ISP parts.
- Growing product mix, shorter production runs, and faster time-to-volume requirements put ICT bed-of-nails (BON) fixture build and test programming on the critical path.
- Today’s electronics are built in geographical locations and with resource constraints that were unimaginable when ICT was declared a “solved problem.”
Despite these changes, the majority of OEMs and EMS providers continue to rely on ICTs from the 1990s. These big-iron testers were marketed and priced as upgradeable platforms that could handle evolving test requirements indefinitely. Consequently, the mind set is that these testers can be used indefinitely, and changing ICT strategies or equipment would be needlessly risky.
Most in-circuit testers are used as glorified manufacturing defect analyzers, testing for shorts and guarded analog measurements, with perhaps a few rudimentary digital tests added. Big-iron testers are optimized around defects that used to occur, rather than those that do. Therefore, it’s pointless to write digital vector tests for modern ICs because chips rarely fail.
High fixture and test programming costs cause test managers to skip significant portions of ICT. Shrinking board geometries reduce access to circuit nodes. There’s rarely enough time during production ramp-up for thorough preparation and validation on big-iron testers whose channel assignment and nail-mapping software complicate the programming process.
Programming and maintaining complex test systems featuring unintuitive software is expensive when trained personnel are available; it is impossible when they are not. Test managers also commonly skip integration of boundary-scan test or ISP on the tester because of cost or difficulty. Adding ISP programming usually requires specialized circuitry, arcane programming, and expensive channel card add-ons. Because most in-circuit testers can program only one chip at a time, lengthy ISP cycles can create production bottlenecks. Even when upgrading for boundary-scan or ISP is feasible, many testers are no longer in production.
The fact that many users pay for ICT service-and-applications contracts is more evidence that the ICT problem is far from solved. We’re left with the paradox that, as boards have become increasingly dense and complex, ICT has grown more abbreviated. Sticking to what worked in the 1990s is inadequate for the test challenges of today’s smaller, denser boards packed with high-pin-count chips, and even more inadequate for technologies such as direct die attachment.
Changing Your Test Strategy
A test manager may assert that ICT is a solved problem and the tester is good enough. An ICT whose complexities make it increasingly difficult to apply against today’s fault spectrum, expensive to fixture and program, and costly to upgrade, is not a solved problem. It is an overlooked opportunity.
Figure 1. Shifted fault spectrum. Big-iron ICTs were optimized around the fault spectrum that predominated in the mid- to late-1990s.
However, it also is an ignored risk. Foregoing boundary-scan because of cost or complexity means lower test coverage and more costly escapes to the next stage. Insufficient ISP speeds at ICT cause production bottlenecks or lead to placing pre-programmed chips on the board, resulting in lost flexibility to handle last-minute code changes or add serial numbers and higher overall costs in high-volume applications. These logistic and financial risks rise when ICT remains unexamined. Most companies that re-evaluate ICT strategies and equipment generally change both.
Figure 2. A modern in-circuit platform architecture provides flexible and straightforward integration of a variety of test and ISP capabilities.
Identifying ICT RequirementsOnce a test manager acknowledges that ICT deserves a fresh look, the question is what is next? The challenge is to select a platform that can test complex boards in a fast-changing, resource-constrained production environment, while providing economic value.
Although big-iron ICT promised expandability and upgradeability, most test managers acknowledge that time and budget constraints conspire against upgrading. Three criteria are suggested to define the most suitable ICT platform:
1. Basic ICT focuses on ease-of-use, productivity, and reliability.
- ICT should match the modern fault spectrum. Low digital IC functional failure rates lead to eliminating digital vector test hardware and software overhead in favor of vectorless test.
- A sensible, test-oriented software architecture that centers on a single test program file eliminates the all-too-numerous files required by big-iron testers. This simplifies documentation and updating.
- An ICT stimulus and measurement system implemented without switch multiplexing conflicts simplifies and speeds fixture creation and test generation. Eliminating the possibility of node-assignment conflicts lets users design and build the BON fixture ahead of time, without having to run nail-assignment software.
- Software should include simple-to-use, yet meaningful tools such as real-time trends, statistical process control (SPC) analysis, and production reports that speed test-program validation.
2. Implementing new test and programming technologies is straightforward.
- Hardware and software platforms are ready to add high-speed, high-throughput serial ISP capability without requiring expensive new channel cards or other specialized equipment and convoluted software.
- ISP capability installed in the ICT fixture, rather than the tester, simplifies integration and ensures high signal integrity at high programming speeds. Unified hardware architecture also eliminates problems associated with multi-vendor dongle-type solutions.
- Limited functional testing at ICT improves next-stage yield. The tester platform should allow implementation of moderate levels of functional test, such as voltage or timing measurements, without adding expensive VXI or rack-mounted instrumentation, which drives up cost and complexity.
- Fast-paced production environments require features that speed test fixture and program changeover. The tester should accommodate simple press-down, rather than vacuum, fixtures so that BON fixtures can be built compactly and inexpensively.
- The era of proprietary tester code for boundary-scan testing has passed. Boundary-scan development and programming tools now are widely available from a variety of vendors. The test platform should be built for direct integration of boundary-scan hardware and software tools.
- Vectorless test has become the standard approach to locating opens around IC packages. The tester platform - especially its software - must accommodate vectorless test tools seamlessly.
3. The platform is simple to deploy globally.
- As electronics are built in more places globally, relying on the test equipment vendor for installation, service, and support is impractical and expensive. The tester must be compact, easy-to-ship, and able to be installed and maintained by the user.
- The test system provides stable, repeatable measurements that are chassis-independent. Every test program should run equally well on all testers.
- The test system must be rugged and simple to repair, maintain, and modify in any locale by local personnel.
- The test executive is based on a de-facto standard operating system that makes it easy to add peripherals such as bar-code scanners, and to integrate the tester into factory networks for data collection.
Conclusion
It’s risky to leave ICT as a static element in a dynamic world of fast-changing technology, shrinking test engineering resources, and escalating time and cost pressures. Reduced test quality and higher costs are the only possible outcomes. Like the products they test, modern ICT platforms should take advantage of technology advances to reduce cost and improve productivity and ease of use. ICT requirements do not call for expensive, time-consuming, channel multiplexing, vector test and proprietary add-on hardware and software that characterize big-iron ICT. Instead, they need designed-in flexibility that helps the user mix-and-match ICT with functional test, boundary-scan, and high-throughput ISP. Above all, the platform must meet resource-constrained realities by offering low acquisition, operating, application and support costs, as well as the ability to be deployed and re-deployed globally without over-reliance on an equipment vendor.
John VanNewkirk, president, CheckSum, may be contacted at (360) 435-5510; john.vannewkirk@checksum.com.