Ad
Ad
© Elmatica PCB | June 13, 2017

Battle of the formats – And how we are all loosing out on it

In this guest article, Andreas Lydersen, CTO at Elmatica, addresses some of the issues that might arise when designing and producing a PCB – languages and formats.
We humans are an interesting species. With time, we have developed several interfaces for communication, such as speech and writing. They are all based on languages – a common understanding of what is relayed through the interfaces. Linguists have traditionally divided language into the following five areas for purpose of study and analysis:
  1. Lexicon (vocabulary)
  2. Morphology (word structure)
  3. Phonology (sound system)
  4. Syntax (grammar)
  5. Discourse (ways to connect sentences and organise information)

In computer languages, the last four areas are what would be called the syntax. Where spoken languages are constantly moving and changing, computers are not as good as humans in understanding deviations, so the syntax needs to be set and agreed upon by everyone who uses it.

Let’s fast forward into when we needed to transmit data on the build of a printed circuit board. We already had an interface – the computer file. Now we needed a language. Brilliant people solved this, they created formats such as the Gerber format, which addressed all of the parts the linguists pointed out where needed. Voilá – we were able to transmit digital descriptions of printed circuit boards. The businesses took full advantage of this revolutionary method, and nobody would dream of transmitting these kind of descriptions in any other way than digitally today. As time went by, several formats appeared, and specialised interpreters with them. A whole eco system of formats now addresses the needs we have. Everything should be in perfect order – we have both the interface and the languages. But still it is not. In fact the business is spending vast amounts of money on understanding what is transmitted through the interfaces.

What went wrong?
Several things:

No common “lexicon”. As technology progressed, the need for the continuous development of word and descriptions falls behind and people starts communicating outside of both the language and the interface. Dialects appears. The need for a human brain to interpret what is meant follows.

The formats does not reflect the modus operandi. Formats are based on the need to communicate a printed circuit design from one party to one or more others. It does not take into account the fact that communication will go both ways. Questions will arise. Standards must be communicated. Defaults must be set. Company policies developed and maintained that directly influences the printed circuit boards. And most importantly – a printed circuit board is not the result of a design alone – it is a result of the combination of several layers of information from all these sources.

Ownership. When formats are owned by someone, they do not invite the trust and feeling of a common goal needed to maintain it.

Compliance. Our current globalised market is full of rules and regulations that directly inflicts upon the creation of a printed circuit board. As long as this is not possible to communicate together with the language, new channels will arise.

Why extending or creating a new format is not the way to go

As long as the design of a printed circuit board itself only represents a minor part of all that needs to be communicated, it does not make sense to build everything into the format. The designer neither should or can input all needed information for the production. All parts of a supply chain holds bits and pieces of the total amount of information needed. The design, the fabrication data, the capabilities of all parties in the chain, the standards and defaults to follow, the “do’s and dont’s” that overrules the input. The compliance data – they all need to be communicated over the same interface and in the same language.

Security is a major issue. As long as we keep building everything into the original file, all parts of the supply chain needs to possess all the data, not just the parts needed for giving a price or checking capabilities.

To give way to new technologies. The landscape is changing – 3D printing is a good example. Actual printed circuits will need to know most of the same data as described above, but can not utilise the current formats.

Too many formats! Nobody is served right by the current battle.

Let’s address this together. Elmatica is currently initiating and funding an open source alternative – circuitdata.org – built specifically to address all of the current problems.

To read more visit Elmatica.

Comments

Please note the following: Critical comments are allowed and even encouraged. Discussions are welcome. Verbal abuse, insults and racist / homophobic remarks are not. Such comments will be removed.
Further details can be found here.
Load more news
August 13 2017 4:10 PM V8.5.9-2