Circuits Assembly Online Magazine - ‘I Just Want Good Boards’

Who pays for those failures in the corner that are going nowhere?

It is a disconcerting and time-consuming fact of life in our industry that the level of bluff and bluster on the part of OEM supply chain personnel often is inversely proportional to their technical grasp of the subject matter.

This leads to pearls of wisdom like these:

“I don’t care how you do it. I just want good boards.” A common refrain. Those of us graced with the task of sitting on the seller’s side of the desk have all heard it. The more astute among us also have picked up the subtext: “And I don’t want to pay for testing.”
Poof. It’s black magic, of the free variety. Just manage a flawless assembly process, and all will be well, notably without testing.
Yeah, right.
Your customer probably wants motherhood, apple pie, clean air and water, two cars in every garage, and a chicken in every pot, while we’re at it. All for free as well. Kind of like school bond issues: We don’t pay for them, do we? Somebody else does? Out of sight, …
Nonetheless, that’s your problem, esteemed Contract Manufacturer. A cost of doing business, but certainly not one to be passed along. So stop whining and suck it up, especially if you want future business.
The customer has spoken. And as we know, he/she is always right. Case closed.

Now what? Panic at the thought of that test expense burning a hole in your pocket? Give it all up? Clearly we are all going to die! (Just kidding. I’m writing this on April 1.)

Seriously, it’s time for an adult conversation with your customer about testing. It is a legitimate cost of assembling a printed circuit board. As legitimate as solder paste, stencils, reflow ovens, chip shooters, payroll and advertising. And somebody must pay those costs.

Why is testing necessary? Because stuff happens. And somebody must catch it. Or else hordes with clipboards materialize at the back door, wanting to know why and how and where about everything. In quintuplicate. Suddenly, Somebody is justifying rather than manufacturing. And Somebody is undisputedly not a charity.

It doesn’t take a Six-Sigma Black Belt with a minor in Lean manufacturing to recognize that machines break, work instructions are misinterpreted, chemicals fail to live up to their shelf lives and humans screw up. In short, stuff happens.

Still not convinced? Want to see some true believers? I bet they’re testing 787 lithium batteries right now at Boeing like their lives depend on it. And in a way, they do.

Now I have your attention.

But do I have your concern, and, most important, your buy-in? Still rationalizing that your process is robust enough to avoid testing? Still insist that it’s best not to raise neuralgic issues like testing with the customer?

Remember the iron law of the Power of One. As in 1000 attaboys cannot erase the fallout from one Oh…(you know what).

As they say, you can pay me now or pay me later. I can help you now, or I can attempt to salvage your reputation later. A cliché, yes, but occasionally clichés are true. And painful. Noticeably the part called “later.” “Later” is usually associated with hard, negative numbers.

So consider this an intervention. An appeal to face facts and admit what common sense dictates. It’s for your own good, trust me. I know these things. I once was in your shoes.

Testing cannot, and should not, be avoided.

There. With that as our premise, it’s time to talk seriously about broadening horizons and removing past perceptions.

The good news is that nowadays you have many cost-effective options. Your choices are not constrained by ICT – and its reputation for cost-prohibitiveness – alone. We have flying probe and JTAG/boundary scan. We even have flying probe incorporating JTAG/boundary scan. We have varieties of x-ray inspection: 2D, 3D, AXI and even variations on CT scanning, and we can do all of it fast and a lot cheaper than you may think at first glance.

I know what your next objection is going to be: We already have functional testing, and it’s good enough.

Really?

What do you do when boards fail? Who pays for those failures? What do you do when your functional test apparatus – most likely designed to suit a laboratory or an NPI environment – fails under the demands of production. Also, how effective is your functional test setup at fault diagnostics? Do you get anything more substantive than a red light or a green light? Again, who is paying for those failures piling up in the corner and going nowhere? Did you account for this attrition in your original quote?

Uh oh. Forgot something?

Wouldn’t it be preferable to have a test regime in place that directly contributes to higher functional test yields? And if it could be demonstrated that implementation of such a regime generates a payback for the incremental cost of testing?

Still think your process is perfect and you don’t need testing?

We’ve barely touched on the environmental considerations. Remember those boards piling up in the corner? Besides adding to your “conversation starter” repertoire, what are they doing for you? Nothing good, that’s for sure. But wait, there’s a solution for those problems too. Testing enables you to rehabilitate and even recycle many of those “bad” boards and return them to productive service, thereby restoring some lost revenue, while contributing to a greener world. How? Because resources aren’t needed to make fresh boards from scratch. Less e-waste means fewer landfills, with more metals and other materials in use rather than in the ground. It’s economical and environmentally friendly. Statisticians (and CFOs) call that a win-win. And you thought all we Test Guys do is probe for opens and shorts.

Green virtue aside, it’s still cost-prohibitive, you insist. Oh yeah? Ever heard of something known as the Cost of Quality?

It’s that pesky concept in which quality costs are a prime factor in quantifying the total cost of manufacturing, including part defects, process deficiencies, and time. Lots of time. Like high-paid engineers’ time, trying to figure out what went wrong, and how to fix it. Time diverted from shipping fresh product and developing new products until the crisis has been addressed and abated. And management time, fighting fires and spinning perceptions. Stated somewhat differently, the Cost of Quality isn’t the price of creating a quality product or service; it’s the cost of not creating a quality product or service. That cost finds no clearer expression than in the unplanned, unnecessary expenditure of peoples’ time.

It’s unnecessary. There is a solution. Trust me.

Just saying…