Newsroom
NEMI Convergence Project Closes In on Single Standard for CAD/CAM Data Exchange
Press contact at bottom of page
APEX (San Diego, Calif.) - January 21, 2002 — Recommendations for a single standard to be used in the exchange of CAD/CAM data for printed circuit board fabrication and assembly are being presented publicly for the first time at the APEX® conference in San Diego this week. The proposed solution uses Valor Computerized System's ODB++(X) data format as a starting point and integrates strengths of the IPC GenCAM® specifications to create a "converged" standard format that will be in the public domain.
This recommendation is the result of a year-long industry effort, coordinated by the National Electronics Manufacturing Initiative's (NEMI's) Data Exchange Convergence Project, and involving approximately 40 companies that broadly represent the electronics manufacturing supply chain.
The project's objective is to lower the industry-wide cost of design exchange. By consolidating existing CAD/CAM exchange formats and standards, the group intends to promote development of software solutions built on a single, consensus-based exchange format that can be widely adopted in a timely manner. The resulting specifications will define formats for the interchange of design, manufacturing, test, fixture and inspection information, and will be controlled by industry through the open standards process of IPC-Association Connecting Electronics Industries®.
"This project involved companies from all relevant segments of the electronics manufacturing supply chain," said Jim McElroy, executive director and CEO of NEMI. "We received input from all of the top five EMS providers, most of the solutions providers active in this area, many of the top OEMs, plus numerous PWB fabricators. Their efforts leveraged the substantial body of work already within the GenCAM (IPC-2510) and ODB++ exchange specifications to create a data exchange solution that will enable the electronics manufacturing industry to reliably and efficiently exchange CAD/CAM data in today's distributed environment."
"Without an industry accepted test methodology, component suppliers have no way to assure their customers of the reliability of their products," said Swaminath Prasad. "They sorely need some means of ensuring that a given lead material and plating combination is reliable. For the long term, we need a complete understanding of why this phenomenon occurs and how to control it."
Recommendation Development Process
Project efforts were organized into two groups: the Convergence Technical Team (CTT), led by Bill Honeycutt of Motorola and John Minchella of Celestica, and the Convergence Management Team (CMT), headed by Henry Jurgens of Celestica and Dana Korf of Sanmina-SCI. The CMT defined the scope of, and developed guidelines for, the project. They also developed a list of key industry leaders in all segments of the supply chain from which to solicit input, and developed the decision process that served as the basis for making a recommendation.
The CTT began its work by mapping the existing capabilities of GenCAM and ODB++ to provide a broader understanding of functionality. In parallel, a sub-group generated a list of user requirements that any future standard should support. This list covered PWB fabrication, PWB assembly, test and inspection, and other aspects of the product realization process. After considering several approaches for creating an integrated starting point that would capture the best of both GenCAM and ODB++, the group settled on an approach that "started at each end" and worked toward a common goal. They organized into two groups - one using GenCAM as a base and the other using ODB++(X). Based on user requirements, missing features were identified, along with the approach to resolve those items (pulling from the other format in most cases). The end result was two alternative proposals for industry consideration.
Industry Feedback and Vote
The Convergence Project used written surveys to solicit initial industry input on the two proposals - first, from solution providers and, later, from users. The surveys also asked respondents to project how long it would take to implement a converged standard, to identify any major gaps in the functionality of the two proposed formats, to identify barriers to implementation, and whether the company would support a converged standard and the IPC standards process.
Users saw availability of industry-proven software solutions as a major driver behind how quickly a converged standard could be adopted and implemented. Solution providers noted several factors that would drive adoption schedules: time to develop/verify products compatible with a converged standard, level of customer pull/demand for products that meet the standard, and dependencies on other solution providers (e.g., interfaces to CAD tools).
The primary issues raised in review of the two technical proposals were:
- Hierarchy vs. flat data structure
- Use of pad stacks vs. layer descriptions
- Ability to partially populate data sets
- Ease/file size of back annotation to CAD data
- Synchronization with PDX/BOM/AVL
- Library support (e.g., for component data)
- Ability to handle test data and drawings (e.g., schematics)
Among both groups (users and solution providers) the majority of respondents indicated a willingness to participate in the IPC standards process. (summaries of survey results.)
Once the initial feedback was tallied and reviewed, the CMT conducted an industry voting process to determine which technical proposal to recommend to initiate the IPC standards process. Respondents voted only on the two proposed solutions, not on any existing implementation. There was no appreciable difference between the results of the initial surveys and the final vote. (list of companies participating in the final vote.)
The EMS and PWB fabricator segments were unanimous in their preference for ODB++(X) as the starting point. The OEM segment showed a solid majority (69%) for the same preference. While there is a clear preference for ODB++(X) in the overall tally (73%), it is important to note that the solution provider segment showed a preference (59%) for GenCAM as the starting point.
The decision process provided for weighting of votes based on size of company (as a rough indicator of industry influence). As Table 1 indicates, there was little difference between results of raw votes vs. weighted votes.
Table 1. Raw and Weighted Results of Final Industry Vote
| Market Segment |
GenCAM Starting Point (raw vote) |
ODB Starting Point (raw vote) |
GenCAM Starting Point (weighted) |
ODB Starting Point (weighted) |
| EMS |
0 |
8
|
0 |
13 |
| OEM |
5 |
12 |
10 |
22 |
| PWB fab. |
0 |
8 |
0 |
13 |
Solution Prov.
|
8 |
5 |
10 |
7 |
| Totals => |
13 (28%) |
33 (72%) |
20 (27%) |
55 (73%) |
"We believe that this converged standard has the potential to achieve much greater industry acceptance than any alternative available today," said Dr. Vincent DePalma, Solectron vice president, technology and new product introduction, and a member of the NEMI board of directors. "This standard is a big step forward in providing a framework for the vast array of software tools required for collaborative design, test and assembly. Solutions that meet the full set of industry requirements have the best chance to displace today's outdated approaches."
Recommendations
The Convergence Project is taking the following course of action to move forward on development of a converged standard:
- An IPC PIN (project initiation notification) form has been submitted to formally request initiation of a standards committee by IPC. This committee will use the industry preferred proposal as a basis for a converged standard.
- A joint CMT/CTT meeting has been held to discuss the path forward and to request participation in the standards committee.
- Focus efforts of the NEMI Convergence Project on recruiting all companies that participated in the industry feedback and voting stages to participate in the standards process (in order to ensure broad industry involvement) and on creating industry acceptance.
As we move ahead, NEMI will encourage early adoption and implementation through tool development, partnerships with other consortia, and demonstrations," said McElroy. User support will also be solicited to help ensure that solution provider investments are timely and of high return. This effort cannot be successful without strong participation of all four of the industry sectors identified. In the end, a standard can only be effective if both users and solution providers embrace it."
About NEMI
The National Electronics Manufacturing Initiative's mission is to facilitate leadership of the North American electronics manufacturing supply chain. Based in Herndon, Va., the industry-led consortium is made up of more than 60 electronics equipment manufacturers, suppliers, industry associations and consortia, government agencies and universities.
NEMI roadmaps the needs of the North American electronics industry, identifies gaps in the technology infrastructure, establishes implementation projects to eliminate these gaps (both business and technical), and stimulates standards activities to speed the introduction of new technologies. The consortium also works with government, universities and other funding agencies to set priorities for future industry needs and R&D initiatives. For additional information about NEMI, visit http://www.nemi.org/.
###
ODB++(X) is the XML version of ODB++
GenCAM stands for Generic Computer-Aided Manufacturing; the XML version of this format is also being used.
For further information:
Cynthia Williams
phone: 207-871-1260
cwilliams@nemi.org