A proposed QML will cost OEMs and suppliers more – and won’t solve the problem.

The apparent disconnect between what OEMs want insofar as incoming PWB quality versus what fabricators are shipping was the focus of an article in the September/October 2008 issue of the IPC Review. The article, “The Road to Qualification,” claims OEMs are frustrated with fabricators that oversell capabilities by claiming “IPC-6012 certified” – a designation that does not exist. A blue-ribbon panel commissioned by IPC, the article says, is seriously investigating the necessity of such a program and its implementation.

My first reaction: disbelief. Here’s my response to IPC. This proposed program is unnecessary because IPC already has in place an overabundance of documented educational, manufacturing and qualification programs specific to our industry. Numerous specifications and educational seminars to ensure quality PWBs and assemblies are readily available and more than adequate, provided customers and manufacturers just use them.

Who is more to blame? The supplier who claims to be IPC-6012 certified or the OEM purchasing department so ignorant (or lazy) to accept such a qualification exists? It is not the PCB fabricator, as the article implies, so much as it is the customer willing to live with the costs involved in receiving a fully documented/traceable PCB order. I can’t tell you how many times I have heard, “We want IPC Class 2.5,” or in other words, “Give us the best quality without us spending a nickel more than we have to.” The cost of quality is very real, and most OEMs either don’t know how to document fabrication notes properly, ask us to ignore fabrication notes in favor of less documentation, or simply don’t want to pay for them.

More important, this road to “certification,” as would more be accurate, would not prevent quality problems. Scrap will continue; sales people will continue to “oversell” their organization’s capabilities, and customers will continue to be naïve about the existence of the IPC quality programs already in place, but they don’t have to be.

IPC has a program for what I call a “voluntary qualification” called IPC PCQR2. This is a database of PCB manufacturers that have submitted test panels to demonstrate their capabilities of conductor and space width, via registration and reliability, soldermask registration and controlled impedance. There is a fee to participate, and the manufacturer is obligated to make occasional resubmissions to keep its data current. The con is that the fabricator produces a number of test panels and submits those it feels best represent its capabilities; in other words, no true first-pass yields. As such, this database does not represent a manufacturer’s day-in, day-out capabilities, but rather is a measure from a small set of test vehicles that, in some cases, could be as many as two years old.

Interestingly, the published participating supplier list as of Aug. 31 has at least one manufacturer that is no longer in business, several who outsource product through an overseas supplier (which begs the question, which of the orders are being built by the tested facility?) and a couple that I personally know have lost business because of quality issues. However, those facts do not negate IPC PCQR2 or the capabilities of those suppliers that have participated in this voluntary program, but rather underscore that ongoing supplier quality cannot be guaranteed by an external third-party program.

IPC needs to promote to fabricators and OEMs the existing documentation for proper board fabrication and the associated test vehicles for acceptance. They are “hit-or-miss,” as the article notes, but old-fashioned education of both supplier and customer is all that’s needed to ensure more of the hits and fewer of the misses. OEMs and EMS companies need to track vendor performance on product received and, when those data are not available, call a customer of a similar caliber as the manufacturer (aka a reference) to help define true capabilities.

Not to sound political, but my position is, enforce the laws already on the books; don’t create new ones. Because IPC or this new proposed entity would not be able to guarantee a manufacturer’s quality or its product, implicit or otherwise, what value has IPC brought to it members and, more important, to the end-customers we serve?

Speaking of serving, has IPC, or anyone for that matter, studied with as much passion the numerous incidents of delamination associated with the use of higher Tg materials? If they have, great! But I haven’t heard much of what I feel is a “real road” to a quality review that needs to be taken, as our industry has been “railroaded” into Pb-free assembly.

Sure, books have been written on Pb-free assembly, and most IPC specs reflect the transition. But rare is the literature on the rash of inconsistency concerning delamination. Other than Werner Engelmaier’s white paper “Recommendations for PCB FAB Notes and Specifications in Printed Circuit Board Drawings for SnPb and Lead-Free Soldering Assemblies, the Qualification of PCB Shops and Activities to Assure Continued Quality” (available at engelmaier.com), not much directs assembler or manufacturer on what is proper. Why aren’t more horns being blown to investigate this problem?

Yes, moisture does play a great part, as it is documented the problem manifests within higher Tg materials and at higher soldering temps more than with conventional FR-4, but why are just a few assemblies here and there affected and not the entire lot? PCB suppliers have an exceedingly difficult time getting consistent evidence from contract manufacturers concerning the true cause of delamination because assembly profiles differ from company to company, and everyone has their own opinion about what is necessary for proper assembly.

As a PCB supplier, it is frustrating to hear such statements as:

“We are learning to live with it; one or two-piece fall-out is acceptable.”

“We ran some of the new date code with the old date code, which ran fine several months ago. But, now the old date code has the problem.”

“All [boards] went through the wave fine, but had a 25% fallout at selective soldering.”

“First pass through wave went well. Several fell out during selective soldering.”

“We had boards delam in the oven.”

“It is just this part number. All others shipped that same week are fine.”

Interestingly – and I don’t know whether to laugh or cry – we have customers that insist boards must be vacuumed-sealed with desiccant, contain humidity strips, shipped in moisture barrier bags, and prebaked prior to assembly, yet they still experience fallout. At the same time, other customers leave the product unwrapped sitting in an uncontrolled environment and never seem to have a problem.

All this underscores that determining whether a fabricator’s capabilities meet a customer’s needs cannot be reduced to picking through an industry database.

As a responsible PCB distributor, we are willing to participate in an IPC commissioned committee to investigate and help resolve this issue. IPC needs to take this road and we, as we are sure there are others, want to help drive.

Greg Papandrew is founder and president of Bare Board Group (bareboard.com), a distributor of offshore-manufactured circuit boards; greg@bareboard.com. His column appears occasionally.

• Prev
• Next