A lot in Electrical CAD and manufacturing quality hinges on data accuracy and completeness of the library data. From the carrion-crow-like perch of someone supporting design automation software, the immense and patient contribution of the ECAD librarians is often missed.
A couple of co-inciding recent events brought me to a renewed understanding of the significance of the work librarians do. Those 2 events were a visit to a customer, and reading a book. The book, I picked up at a church hall used book sale. Quality is Free – the Art of Making Quality Certain. It was first published in paperback in 1980, ISBN 0451624688 by author Philip B. Crosby.
This book has a few things going for it, one of which is it is a Mentor publication – but no relation to Mentor Graphics. The Mentor Books – New American Library, is, or at least was then, a sub-division of McGraw Hill Publishing. For a chance encounter it was interesting reading and a good slice of luck. I think it is out of print now. The domain the book covers is general industrial production and as well as offering plenty of nuggets of straightforward wisdom like “act now for reward later” Crosby goes into convincing detail about the patience required to achieve quality improvements.
I like that a librarian themed blog post involves a book. Next time you go into a public library be sure to wish the person behind the counter “Many Happy Returns”
A high quality electrical parts library to support wiring designs is achieved first by putting in place the right resources for the right length of time. But a close second is senior management makes the responsibilities of librarians untouchable and uncompromised. Giving librarians the authority and autonomy to manage their domain is highly beneficial. A division of labor has been popular in for profit organizations since mediaeval times. It is a winning formula. Librarians have better expertise and the experience in component data management than harness design engineers for example, who to give them their due are far better and more valuable as designers than the librarians may be. Better quality data in the library is not a finish line to pass and then forget about and go on to the next thing, it is an ongoing commitment.
Comprehensive and complete data.
Imagine there is a missing electrical connector in the library. Is the quality problem of the part not being present solved by the part reference being added? If you think at first blush that this is the end, then consider whether for electrical CAD purposes a basic definition really will suffice. Depends what you mean by a “basic” set of information might be your reply. What to include in the set of basic information then? A majority of engineers may want an accurate cross-reference to who the supplier of the part is additional their own internal part numbering system. You want to know where the part can be sourced from. If that part reference, any part reference contains separation or position significance coding, well that’s got to be accurate too for consumers of the data down the line.
Yes, accuracy makes a tremendous difference:
What’s the difference in the part number ………….. from supplier
3579246 3589246 35-79246 35-792E6 357924-Blue …….. Johnson or Johnston or Johnstone or Jonsson.
It is knowing and handling these subtleties which librarians excel at. Or in CHS, they Capital Library at. (second bad pun of the post).
Now the surface has been scratched relating to one or two small pieces beginning a reasonable “basic” subset of data, let’s be bold and examine a few more touch points. What about the CAD symbol or symbols to represent the individual connector? Are they part of the minimum data set you are looking to achieve before you would consider the part to be ready in your database?
Terminals taken and the wires that are valid for inclusion in the wires and in which cavities these can be inserted.
Locking mechanisms attachment methods and strain relief configuration choices.
Plugs and sealing parts
Conformity to environmental control and recycling/pollution standards
Extract of or link to official specification documents
Why is this detail important to you?
Because parts engineers and librarians are not just people too, they are pivotal to your success.
Simple five level elevator plan.
The customer I was visiting has guidelines for different levels of completeness of their library data supporting multiple projects in multiple locations. This is a handy way of breaking down the tasks so that the workflow steps are more manageable. Tasks can be split between locations and resources. Librarian task can be split so blocks of data creation and maintenance can be separated out and resourced with temporary help. Per project this perspective helps you to get to a believable % of task complete measurement. The levels provide a simple way of organizing and executing harness component data management and infer the metrics which will be used to track progress.
The levels are:
1) Base data and vendor reference data
2) Design and manufacturing symbol approval
3) Device transmittal layer (CHS footprint)
4) Sub-component interrelationships (for example wire >> terminal >> connector)
5) Models to support electrical simulation/analysis
Being able to measure means you can manage transition from one plateau of accuracy to the next. You can see the dimensions of your data management tasks. You can see it progress to a quality where it supports a high quality design. One of the things I like about this approach is that it acknowledges there is no fixed “minimum” standard. Having the library data good enough to get the job done is not good enough tomorrow when you need different parts information from the library. And then just when you think you have all the information about all the parts you’ll ever need someone will invent some more or want some more data associated with the existing ones or you will get a new product line. The concept of attaining a minimum doesn’t sit easily with the goal of reaching the best quality.