I'm surprised that most of the comments have missed the point. The assemblers did not have a problem. Putting part A from a bin marked A into every spot marked A on a board is a very reasonable assembly process. If every assembly works and never needs to be repaired, it is a way to possibly save money. But 100% yield and no repairs ever needed is an unlikely situation. Steve discovered the problem when he tried to repair a board with several 10K resistors all marked R1. If you want to measure the voltage on the 10K resistor at the output of a circuit, where do you put your probe? You have several different R1 choices and no way to tell the difference between them. You are then forced to trace the copper connections by hand to figure out which resistor is which, because the labels don't differentiate between them. This is an insanely expesive way to troubleshoot a board which is why component designators on a board are always (except at this crazy company) unique.
As a "greybeard" I've come across many poor design decisions that have much in common with this example. Any design decision is an attempt to solve a problem and is always a compromise between many competing tradeoffs. Poor decisions usually result from a "small-picture" view of the problem due to ignorance of the impact the decision might have somewhere on the "puchasing-manufacturing-end user" chain. In this case a slight improvement in the assembly process made troubleshooting almost impossible. A fact which the designer seemed blissfully unaware of.
There are two purposes for silkscreening component numbers on a board - to populate the board with the correct components, and to be able to troubleshoot the board at testing. The second is where the system described here failed miserably.
Let's say you have a board with 3 capacitors labeled "C1" and 10 resistors labled "R1". Your schematic shows two different "C1"s connected to five different "R1"s. If one of your "C1" capacitors has popped, how do you find it on the schematic to troubleshoot it?
I agree that replacing the name "R1" to "10K" would remove some ambiguity during assembly, but isn't that what a BOM is for? Plus they have 1 letter of silk screening or printing of whatever sort. Not sure if they are paying by the letter. Plus schematics, if drawn properly, have all the info needed per component, right?
When the PCB shrinks and populates the boards to the point where silk screening will not help at all, what do we do? What will human hands do then? I worked at a place where they have one person "who was good at the little parts." They would handle surface mount components. After seeing their not-fantastic job are rebuilding, I just printed and placed another board.
Oh yeah... R1 may change in the future... all the "10K" printed boards would be an issue.
The other downside is that the assemblly line workers learn nothing! So when a decision has to be made it will be the same one over and over again since that is the way it is always done. No thinking. No initiative.
It would be nice to allow the assembly line workers to grow in their craft. Reading resistors and capacitors is a very good start. That way when a mistake on the drawing is made, someone can catch it before 1000 boards have to be reworked or thrown away.
Sounds like a good example of a non-technical person participating in the design. We would get crazy stuff like that from sales all of the time...what seems logical to others can really play havoc with electronics design. That is why a lot of distributors have an engineering staff - to keep their sales department from making promises they can't keep.
Truchard will be presented the award at the 2014 Golden Mousetrap Awards ceremony during the co-located events Pacific Design & Manufacturing, MD&M West, WestPack, PLASTEC West, Electronics West, ATX West, and AeroCon.
In a bid to boost the viability of lithium-based electric car batteries, a team at Lawrence Berkeley National Laboratory has developed a chemistry that could possibly double an EV’s driving range while cutting its battery cost in half.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.