A new, broad industry consortium drives for standardization across design, fabrication, assembly and test.

PCBs have changed significantly over the past three decades, yet to the surprise of many, we still commonly use 30-year-old ways of communicating design intent to manufacturing. These decades’ old data-communication formats were originally conceived to drive the emerging numerically controlled machines. Adoption of these formats was further driven by the growth of computer-aided design tools in the early 1980s and the need to move from reprographic photography artwork to digital data-driven manufacturing.

Many formats were created by hardware manufacturers themselves, or derived from broader standards, and over time became commonplace. Such numerically controlled equipment included photoplotters, drill/mill machines, assembly (insertion/pick-and-place), test, etc. The Gerber Scientific Corp., a photoplotter hardware manufacturer, created one of the industry’s best-known de facto standards, Gerber RS-274-D, which later evolved into RS-274X.¹ But Gerber data alone were not enough to produce a completed PCB assembly, so other data formats were created, again to drive specific numerically controlled manufacturing equipment.

Using multiple formats to fabricate, assemble and test a PCB is fraught with issues. For starters, there are data inconsistencies caused by the CAD software when producing all the different file formats, especially when data files are not from the same revision of a PCB design. Most formats in use today were designed to drive machines, not to provide a complete relational view of the design data. Over the past three decades, this has been compounded by the evolution of PCB design, manufacturing and test capabilities. Consider just a few of the technology advances since the humble Gerber came about: surface mount components, JTAG test circuits, BGAs, blind and buried vias, microvias, buildup layers, embedded components (discrete and active), and embedded waveguides. The list will continue to grow as designers struggle to address consumer desires for more (functionality, reliability) from less (size, power, weight, cost, etc.).

IPC, the trade group, has been aware of the challenges and dynamics of the PCB design and manufacturing segment and has been an ardent advocate for the replacement of Gerber for more than 30 years. Over that time, IPC has defined and published multiple data format specifications that vied to streamline the process of transferring data from design tools to numerically controlled manufacturing and test equipment. In the late 1990s, it began looking at a single open, nonproprietary, holistic definition for printed board manufacturing, assembly, inspection and testing. That project was known as “Generic Requirements for Implementation of Product Manufacturing Description Data and Transfer Methodology” or GenCAM.

GenCAM, which under the IPC taxonomy is also known as IPC-2511, was first published in early 2000. It was a revolutionary step beyond Gerber, and, as such, was a stretch for many to adopt. Instead, companies in the PCB supply chain took a passive approach to adoption, waiting for others to take the lead, while continuing to use Gerber despite the known costs incurred from its inadequacies.

In early 2001, iNEMI (the International National Electronics Manufacturing Initiative) stepped in to lead a broad, industry-wide project to define the definitive data exchange convergence specification. The goal was to enable accurate, efficient data exchange between designers and manufacturers of printed circuit boards (PCB) and assemblies using a single XLM-based data exchange format. From this effort, an IPC committee developed a new standard, IPC-2581, Generic Requirements for Printed Board Assembly Products Manufacturing Description Data and Transfer Methodology; it was released in March 2004.

A New Consortium

With IPC-2581, the industry finally has a global opportunity to replace Gerber and save millions of dollars wasted by its inadequacies. But as usual with data format specifications, nothing can be done until the supply chain produces, consumes and supports such a specification. This is where a new consortium of PCB design software (EDA) and supply chain companies fits in: to bring companies together to enable, facilitate and drive use of IPC-2581. The charter is simple: “To accelerate the adoption of IPC-2581 as an open, neutrally maintained global standard to encourage innovation, improve efficiency and reduce costs. The members of the consortium will openly support and promote the adoption and usage of IPC-2581 by enabling their products, offerings and services to import/export/consume IPC-2581.”

The consortium was created in June and includes the following founding members from across the PCB design/supply chain:

• OEMs: Harris, Ericsson, Fujitsu, nVidia.
• EMS/ODM: Sanmina-SCI.
• Software tools: ADIVA, Cadence, Downstream Technologies, Zuken.
• Media: UP Media Group.

The consortium is open to any PCB design/supply chain company that is prepared to adopt the consortia goals/objectives and commit to a roadmap for IPC2581 adoption: a) publicly state that the company is a supporter of IPC2581, and b) actively support IPC2581 by either producing or consuming IPC2581 in the next 12 to 16 months.

Further, systems companies are asked to require suppliers to produce/consume IPC-2581 data within the next 12 months. EDA design tool companies are asked to output IPC-2581 from board designs. DfM verification companies must export IPC-2581, and fabricators, assemblers and test companies must import and consume IPC-2581.

Why is Cadence Design Systems involved, and to what degree? Cadence believes it is in the industry’s best interest that an open, public, neutrally maintained standard be adopted by all segments of the PCB design, fabrication, assembly and test supply chain. Cadence commits to develop and maintain IPC-2581 import and export from its Allegro PCB design software and stay current with the latest approved and published IPC-2581 specification. Cadence additionally commits to collaborating with all consortium members to ensure that Allegro-derived IPC-2581 data can be accurately and smoothly consumed by their technology(ies), methodology(ies) and process(es).

Whither ODB++? Before being acquired by Mentor Graphics, Valor Systems was in some ways a major contributor to IPC-2581. Indeed, Valor donated its ODB++X format, which was merged with IPC’s GenCAM to form the foundation of IPC-2581. With ODB++/ODB++X no longer in the open “pseudo standard” domain, the consortium feels compelled to ensure that an open, unbiased “living” standard is maintained, adopted and promoted across the global supply-chain.

References

1. Mike Buetow, “A Short History of Electronic Data Formats,” pcdandf.com, June 28, 2011.

Keith Felton is product management group director for Allegro and OrCAD products, and Hemant Shah is product management director for Allegro PCB products at Cadence Design Systems (cadence.com); shah@cadence.com.

Ed.: UP Media Group is actively supporting the effort to move users to an intelligent data transfer standard. This is our corporate position: For 20 years, UP Media Group has supported the development of an intelligent, robust format for electronics data transfer. We fully support the consortium’s effort to ensure a viable, supported and independent data transfer format that is driven by user needs.