Estimated reading time: 4 minutes

Nolan’s Notes: The Cost of Rework

I’ve been repairing an electric guitar—a no-name copy of the iconic Fender Stratocaster. The instrument came to me by way of my guitar teacher, full-time musician, and good friend, Ken Brewer, in a condition described as “functional, but unplayable.” It had potential, so we decided to see if we could bring it back to life.

I’m not sure what the previous owner had in mind, but their small maladjustments in one place had to be compensated for by medium-sized maladjustments elsewhere, leading to big maladjustments that rendered the instrument mostly unusable. Nothing worked as it should.

In this issue of SMT007 Magazine, we visit the rework processes, arguably one of the highest-skilled, highest-pressure jobs on the manufacturing floor. What are the technical challenges? How is the rework job function being changed by new packaging technologies and increasing board densities? When faced with a staffing shortage in rework, how can we remedy the situation? How will the skill sets for rework need to evolve?

There aren’t many functional pieces to an electric guitar; the bill of materials is pretty short. It’s not like, say, the PCB of a cellphone. But when it comes to rework, the process is quite similar. For example, when first handed to me, it was as if that Stratocaster copy had been rejected by test. The next steps were to diagnose the problems and determine a course of action. Was it repairable? Was it—like so many of our PCBs in production—even worth repairing? Our triage suggested there was hope, and it was worth our while to try.

So, we carefully removed the essential parts and tested them. Electrically, everything worked. The problems were, it seemed, entirely mechanical. Some of the fasteners used in the previous work, were not as specified, leading to much of the guitar’s maladjustment. Compare this to assessing a faulty assembled board. What parts need to be tested, removed, or replaced? How will reworking those parts of the board affect the performance? For my guitar, because of failures and bad repairs by the previous owner, all it needed was a second rework to put it right. Similarly, what does this look like in a test and inspection shop? Should an OEM or fabricator try to set up an in-house rework department, or is the time and money better spent just sending it straight to dedicated experts?

My guitar project, just as with blemished board assemblies, required time-intensive rework. All that triage, disassembly, error resolution, and re-assembly takes patience and skill; understanding the subtle interactions requires true expertise.

It took my own experience and Ken’s many years of expertise to fix my guitar. Similarly, a rework department requires skilled technicians, especially given the high degree of troubleshooting required to tackle cutting-edge repairs. My guitars aren’t quite as technical as circuit boards, but there are a lot of interrelated mechanical parts to deliver a good tone.

Ken has been working on them since the 1960s; it’s his lifelong profession. His body of knowledge is immense and soaking that up quickly helped me learn my way around. I got the reworked guitar into playable condition after a couple of hours, but it took an entire evening of Ken’s wisdom to go from good to great. That human expertise can only come from direct experience.

That is where my guitar story and this issue’s topic start to converge. Rework, regardless of the discipline, requires a whole lot of human expertise and dexterity. Unlike classic electric guitars, however, printed circuit board technology advances quickly. That evolution likewise constantly shifts the balance between staffing for rework, hiring it out, or not doing it at all. Doing the build right the first time is the best strategy, of course, but we can rarely be fully 100%.

In this issue, we highlight key areas across the rework discipline. We have a feature interview with Intel’s Maria Mejias, who not only oversees production rework at Intel, but also contributes to rework recommendations passed down to Intel customers. The team at BEST, the rework specialists, discuss the ROI dynamics in pursuing rework for your products. BEST also contributes an article on their latest part removal method, cold milling. We have included a technical paper from Essemtec, published at IPC APEX EXPO 2023, on novel methods for automatic repair.

Of course, the obvious approach to rework efficiency is to have no rework at all. That’s where Delvitech’s AI AOI, and Hanwha’s portfolio of assembly equipment come into play.

Our band of columnists pick up the rework beat as well. Michael Ford examines the current state of rework and sees a business opportunity; Mike Konrad delivers a comprehensive overview that, while valuable for all readers, should be especially helpful to readers less familiar with rework. Rounding out the band this month, we have Jennie Hwang’s ongoing series on critical materials, IPC CEO John Mitchell, and Ron Lasky’s latest Maggie Benson installment.

In the end, refurbishing cost me very little in parts, but it did require that I make mechanical changes to the placement of the bridge on the guitar body. Comparable, I guess, to removing a part and resoldering it with correct placement—almost all the cost of the rework was in the labor.

This column originally appears in the January 2024 issue of SMT007 Magazine.