REMOVING lead from electronics may be a good idea, but the transition to lead-free solder does dramatically alter the electronics assembly procedure, particularly for high-reliability products. Tin/lead solder joints have been comprehensively characterised over many decades, and manufacturers of products for safety-critical or life-dependent applications are able to reassure their reliability.
This is not the case for lead-free alternatives. Although staff at standards bodies such as the Association Connecting Electronics Industries (IPC) and the International Electrotechnical Commission (IEC) have devoted many man-months to characterising solder joints formed using their recommended lead-free alloys, there is much more to do.
In addition, high-reliability products such as avionics systems or utility meters can have a life expectancy of up to 25 years.
Rework procedures using tin/lead solder are proven and reliable, while questions inevitably remain about those formed from lead-free alloys.
The task of fully characterising lead-free solder joints becomes doubly important when one considers that consumers might not feel so comfortable about moving quickly to substitutes if they realised there could be lingering doubts about lead-free reliability; especially if their lives depend on lead-free electronics such as being a passenger on a commercial aircraft or in a vehicle equipped with an airbag.
Reliability is underwritten by a number of attributes, including good design, high quality components, and PCBs that are able to withstand shock, vibration and thermal excursions. Nonetheless, solder joints have always been considered a weak link and the introduction of lead-free solder has done nothing to mitigate the risk.
To lower the probability of failure, manufacturers routinely test component and PCB pad solderability. Without good solderability–a measure of how well molten solder wets component joints and PCB pads indicating the robustness of the surface finish–the likelihood of forming a reliable solder joint is dramatically diminished. Joints exhibiting poor solderability may still form a solder joint and even pass final electrical test; but the likelihood of failure due to stress, heat or vibration is high.
But the change to lead-free soldering has exposed deficiencies in traditional test methods used to establish solderability.
The importance of solderability testing
Component manufacturers and PCB makers prepare their devices for soldering by applying one of several surface finishes to the device terminations and pads on the board. Examples of surface finishes are organic solderability preservatives (OSPs), immersion tin and immersion silver, electrolytic and electroless nickel/gold, and fused or hot air–levelled solder.
These surface finishes are designed to both protect the terminations and pads prior to assembly, and then provide a surface that encourages wetting by the solder during the reflow or wave solder operation. Although each surface finish has its merits, it is generally agreed that solder is the best finish when fusing to solder paste or itself.
Unfortunately, all surface finishes are subject to a degree of degradation.
The mechanism that causes solderability degradation is down to thickening of the oxide and intermetallic layer. The rate of growth is influenced by storage factors such as humidity, temperature and time. Figure 1 demonstrates the effect of oxide and intermetallic growth over time on solderability. It can be seen that solderability decreases as a function of ageing. Stages I and II represent areas where soldering is possible, stage III represent soldering failure.
Measure solderability, not soldering ability
Experienced manufacturing engineers are well aware of this failure mechanism, and routinely take steps to determine if their components and PCBs are still satisfactory prior to assembly. However, the methods they use to test whether solder joints will last the life of the product have been called into question by the migration to lead-free alloys and the IPC’s and IEC’s comprehensive testing. To make matters worse, there is an industry-wide lack of understanding as to what attribute is actually being tested.
The problem is that manufacturers often fail to draw a clear distinction between solderability– how well molten solder wets– and soldering ability –a term used to describe how well a specific combination of flux and solder work together to ensure a component is soldered to a PCB.
It’s important to note that a device with “unacceptable” solderability (according to the standards) can often still be soldered with the “right” combination of flux and solder. But that’s no longer an option when using lead-free alloys for two reasons: The much narrower process window for lead-free soldering is less forgiving of process changes during production and because additional flux has a detrimental effect on board cleanliness.
The IPC’s and IEC’s analyses established that the most important criterion for ensuring that lead-free soldering produces high integrity interconnections–while maintaining tight process control and limiting potentially failure-inducing contamination –is “acceptable” solderability.
“Solderability evaluations are made to verify that the solderability of component leads and terminations meets the established requirements and to determine that storage has had no adverse effect on the ability to solder the component to the board”, notes the IPC J-STD-002C & 003B documents.
And, as Dave Hillman, Chairman of IPC-J-STD 002 and 003 puts it: “The JSTD-002/003 solderability test standard’s purpose is to test the robustness of a finish for wettability. Too many people attempt to use and/or believe that the standards are designed to mimic production conditions, they are not. This would be ‘soldering ability’ not ‘solderability’”.
Testing a component using a manufacturers regular component/flux/solder combination to ascertain the solderability of a surface finish doesn’t adhere to any standard and isn’t likely to provide the kind of repeatable or reproducible result demanded by an unforgiving lead-free soldering process.
Choosing the correct test method
The IPC committee agreed to undertake a “round-robin” test program to study the characteristics of lead-free alloys and fluxes. The round-robin tests looked at alloy, surface finish, temperature, flux and gauge R&R.The standards, which are due to be published imminently, will also formally clarify the difference between solderability and soldering ability. The standards will define a process for solderability testing that precisely specifies the test equipment, how the test should be conducted, and the materials to be used to ensure good Gauge R&R.
Furthermore, the solder in the solder bath used for solderability testing should also be chemically or spectrographically analyzed or replaced each 30 operating days according to strict standard-defined contamination limits. This includes the composition of the lead-free solder (including maximum contamination levels) being maintained during testing with the silver and copper element levels adjusted for alloy requirements.
But what of the test methodology itself? Is the decades-old dip-and-look test method formalised in standards such as J-STD-002 and 003 an acceptable solution?
In this test (which is well defined), technicians dip representative component samples into a molten solder bath and observe how far the meniscus climbs. With experience, this provides a qualitative measure of solder wetting and hence solderability. The method is also quick, easy and cheap to implement.
The major problem with dip-and-look, however, is that as a test method it exhibits poor Gauge R&R. In other words, even if the materials used for the test are strictly controlled (as required by the new standards), two people conducting the test at different times are likely to interpret the results differently.
“Users who believe that the dip-and-look methodology has a respectable Gauge R&R would be in for an extreme shock” said the IPC. Backing up these comments the IPC also added: “The IPC committees have voted that no new solderability test methods will be introduced into the standards without a demonstrated, industry acceptable Gauge R&R value.”
While this recommendation is not presently included in the IEC Standards, the committee have agreed to commence a further revision within 2 years, instead of the normal 5 years, and solderability testing with acceptable Gauge R&R will be included at that time.
Force measurement recommended
According to the IEC: “[We recommend] wetting balance force measurement and globule testing and we are attempting to harmonise standards documents, to provide acceptable Gauge R&R to its defined methodology.”
Solderability measurements using a wetting balance measures force to an extremely high level of accuracy (to a resolution of milli-Newtons). Some electronics-based instruments even measure to levels of better than 1µN/Bit. Although the type of wetting balance used for plated through-hole (PTH) and surface mount (SM) components differs, both are based on the same physical principles.
Namely: if a metallic body is dipped into molten solder, the weight and speed with which the solder meniscus climbs upwards on the body’s immersed surface indicates how well the solder wets it and thus its solderability. The greater the solderability, the higher the meniscus will climb and that can be measured as a change in the vertical force action on the suspended specimen. (Figure 2 demonstrates the test procedure graphically.)
For certain TH components and circuit board coupons, the specimen device is immersed in a bath of molten solder and the forces of buoyancy and surface tension action upon it are measured. For smaller SMDs, a higher resolution method is required: the microwetting balance procedure that employs a solder globule. Here the solder bath is replaced by a globule block of 4, 3.2, 2 or 1-mm size employing 200, 100, 25 or 5-mg pellets of solder alloy (depending on specimen size) allowing the individual leads of a multi-leaded component to be tested.
Time for change
The introduction of lead-free has disrupted a process–assembly using tin/lead solder that has been characterised over many decades. The knock-on effect is that the IPC and IEC have determined–after extensive examination involving over 30,000 tests–that traditional test methods and process corrections are no longer sufficient to determine the reliability of lead-free joints.
New methods and clarifications of the difference between solderability and soldering ability are due from the organizations before the end of this year.