Kelly Dack and Mark Thompson Unite in the War on Failure


Reading time ( words)

There’s been a lot of talk about fighting the war on failure in the PCB industry. But what strategies should our generals follow to prosecute this war? What exactly constitutes a failure in the first place? Is this war even winnable? I recently spoke with longtime designer Kelly Dack and CAM support veteran Mark Thompson of Prototron Circuits about the best battle plans for beating failure, and why designers and manufacturers must team up against this common enemy.

Andy Shaughnessy:  Kelly, you recently said, "Failure wins when it leverages the guerrilla warfare tactic of divide and conquer to confuse and disorient the product development community. Designers and manufacturers must unite to fight this war.”

Why don't you start off by saying what failure means to each of you at the design stage and at the board shop stage?

Kelly Dack:  Andy, failure is defined as not meeting the intended objective. But I think the concept is easily confused if coming from isolated design and manufacturing perspectives. Meeting the intended objective successfully can only come from allied efforts.  Designers need to describe or define objectives for their projects in terms of concise orders in order to make successful products, but this will only come about if the folks giving orders are also able to follow orders.

Mark Thompson: Absolutely, I agree with Kelly entirely. I'd say from the fabrication side, following orders is easy if the orders are given from time-proven, industry standard specification. We open the door for failure when we have to question orders. But we manufacturers have to occasionally, when the orders are outside of the guidelines of conventional process or machine capability. Designer and manufacturer communication and respect is key here, and I think this is why Kelly mentions that it is important that those giving orders must be willing to take them if the order would put the part at risk of failure.

Shaughnessy: Is failure always someone's fault or is it the fault of a bad process, or is it a combination of both?

Dack: I’m not comfortable pointing fingers at individuals. The causes for success and failure can be attributed to people, but it can also be caused by unforeseen conditions or even acts of God. A good strategy is to identify performance requirements and operating conditions. The “battle plan” needs to identify constraints: design constraints, manufacturing constraints, performance constraints, even cost constraints. Knowing the enemy and anticipating every possible way a design could possibly fail is crucial, but it can’t happen in a vacuum.  It must occur through open communication with the allies – stakeholders of the product who are connected through the process steps.

Thompson: If you're asking if every failure is someone's fault, I would say yes, at the most base level. It will always turn out to be someone's fault either due to negligence or lack of vigilance. But, as Kelly said, there are extenuating circumstances. There can be dynamic situations where environmental, human or mechanical conditions create a failure in the field. Or it could be a poor design type of situation, where something wasn't properly considered in the original design phase.  It’s a good time to draw in the metaphor of a combat unit here. No one person in a combat unit takes credit for the unit’s success or failure.  They are a team!

To read this entire article, which appeared in the August 2015 issue of The PCB Design Magazine, click here.

Share

Print


Suggested Items

Just Ask Heidi Barnes: The Exclusive Compilation

01/15/2021 | I-Connect007 Editorial Team
We asked for your questions for Keysight Technologies' Heidi Barnes, and you took us up on it! We know you all enjoyed reading these questions and answers, so we’ve compiled all of them into one article for easy reference. We hope you enjoy having another bite at the apple.

TTM’s Approach to Stackup Design: Train the Customer

01/12/2021 | I-Connect007 Editorial Team
In this interview with the I-Connect007 Editorial Team, TTM’s Julie Ellis and Richard Dang drill down into stackup design, detailing some of the common stackup challenges that their customers face when designing for both prototype and volume levels, and offering advice to designers or engineers who are struggling with stackup issues. They also discuss why having too many different prepregs in a stackup can be asking for trouble, and how proper stackup design can optimize both the fabrication and assembly processes.

Cutting Respins: Journey to the Single-spin PCB

01/07/2021 | Chris Young, The Goebel Company
PCB design is more than a short sprint to the finish line; it is a journey best suited for the prepared adventurer. According to a study by Lifecycle Insights, the average PCB design project requires 2.9 respins. These respins can cost anywhere from tens of thousands to millions of dollars—each! As an engineer/business owner, I find respins frustrating because I would rather spend my time and money applying scientific principles inventing, improving technology, and solving problems. I am not an advocate for perfectionism, but rather I focus on becoming a better adventurer. Sometimes I get to taste the sweet wine that is a single spin PCB. As fellow adventurers, let’s discuss some topics that influence unnecessary return trips on our PCB design journey: simulation, technical reviews, and interest in PCB design.



Copyright © 2021 I-Connect007. All rights reserved.