Last Updated:08/17/2026 Author: Zeon
I still bear in mind the very first board that showed me to dislike a surface area finish. A customer's fine-pitch BGA layout-- a 0.5 mm pitch networking ASIC surrounded by a ring of 0402 passives-- came back from SMT setting up with 14% of the panels falling short connection. The failures were all in the very same place: solder bridges in between BGA pads that were meant to be 0.3 mm apart. The X-ray photos resembled a person had actually poured solder over the whole BGA location.
The board home's initial instinct was to blame the stencil. Then the solder paste. After that the reflow account. We examined every one of them. They were gerat.
Then a person ultimately asked the concern nobody had asked throughout ordering: what's the surface area coating? The solution was HASL-- hot air solder progressing. Low-cost, fast, and absolutely wrong for this board. The customer had saved $0.11 per board on the finish. The rework and scrap costs for that manufacturing run involved simply under $19,000.
To recognize why HASL stops working on fine-pitch boards, you have to understand the process. A bare board obtains dipped into a bathroom of liquified solder-- lead-free SAC305 in a lot of modern-day fabs-- and and then pulled through a set of hot air knives that blow the excess solder off the surface. The goal is a thin, also covering over eveyr revealed copper pad.
That's the concept. The physics are much less participating.
HASL thickness varies extremely accross a single board. Depending on pad geometry, board orientation, and exactly how the air blades hit, the solder layer can range from 1 micron to 40 microns. Some pads wind up virtually level. Others end up with a visible dome of solder. This isn't an issue in a details manufacturing facility-- it's a fundamental quality of the process. Every HASL board in the world has it.
On a crude board with 1mm pads and broad spacing, a 20-micron height difference in between adjacent pads is meaningless. The solder mask does its job, the parts are big, and every little thing jobs.
On a fine-pitch BGA impact, the numbers inform a various story. A 0.5 mm pitch BGA has pads around 0.25 mm in size. A 0.4 mm pitch BGA-- significantly typical in customer and industrial items-- has pads closer to 0.2 mm. Now stack the math: a 25-40 micron solder dome on a 200-micron pad is a 12-20% variant in pad elevation. Throughout the hundreds of pads under a single BGA, you get a surfce that resembles a small mountain range as opposed to a level touchdown area.
BGA solder spheres uncommitted about hills. They need level.
The irregular HASL surface develops the two traditionnal BGA flaws: solder connecting and open circuits. They seem like revers, yet they share the same beginning-- the balls no longer land where the design presumed they would certainly.
Solder bridging takes place at the extremes. Where a HASL dome is highest, the solder layer extends laterally past the pad's nominal limit. On a 0.3 mm space, that's enough to substantially minimize the clearance in between surrounding pads. When the BGA sphere reflows and the solder paste melts, the mix of extra solder from the HASL finish and reduced spacing is a recipe for bridges. The solder does not require to be severely misaligned-- it simply needs to discover a course across a gap that's already been narrowed by the coating.
Open up circuits hapen at the contrary extreme. A BGA package's solder spheres are co-planar to within a couple of microns when they leave the plan supplier. The board's pads require to be co-planar too-- that's the entire point of the "land" in "land pattern." When some pads rest 30 microns greater than their next-door neighbors, the BGA rounds touch the high pads first during positioning and reflow. The reduced pads can be entrusted to a weak joint, a partial joint, or no joint at all. These are the failings that pass a fast visual check and afterwards turn up as periodic mistakes at functional test-- or worse, months later on in the field.
There's a 3rd, quieter issue too. Solder paste printing presumes a level surface area. The stencil aperture is designed to deposit an exact quantity of paste onto each pad. On a HASL dome, the paste spreads unevenly, thins at the top, and pools at the sides. Even boards that do not bridge can end up with inconsistent solder joint volumes throughout the BGA-- which shows up as nullifying and integrity concerns down the line.
Nobody develops a board and claims "I 'd such as some solder bridges, please." HASL obtains defined for three factors, and all three are easy to understand.
Cost. HASL is the most affordable surface area coating in manufacturing, and at quantity the per-board delta against ENIG-- electroless nickel immersion gold-- builds up. For a board without fine-pitch components, that conserving is reputable. The problem is when the same price reasoning obtains put on a board which contians a single BGA. One 0.4 mm pitch part is enough to invalidate HASL for the hwole board, but the acquiring decision does not always make that link.
Familiarity. HASL has actually been around for years. Older engineers and procurement groups have used it on every board they've ever before constructed, and it's never ever attacked them. What they're not always aware of is that their older styles had 0.8 mm or 1.0 mm pitch parts. The sector's ordinary pitch has diminished significantly in the lastr 5 years, and the finish that worked with the old boards doesn't transfer.
Factory defaults. This is the one that frustrates me most. An unexpected number of board residences default to HASL when a customer does not specify a coating, due to the fact that it's their cheapest, highest-throughput line. The order goes in, the surface area finish field is left blank, and HASL obtains used by default. Nobody asks whether the board contains fine-pitch parts. The customer uncovers the issue at the initial setting up run.
The advice right here isn't controversial-- every significant board residence and standards paper states the same thing. For fine-pitch BGA setting up (0.5 mm pitch and below), ENIG is the conventional selection. The nickel layer (typically 3-6 microns) is deposited by chain reaction, which means it conforms to the pad geometry atom by atom instead of being blown on by hot air. The immersion gold flash (0.05-0.1 microns) secures the nickel and supplies a flat, solderable surface area. The outcome: pad coplanarity gauged in microns instead of 10s of microns.
OSP-- natural solderability chemical-- is the flat, low-cost alternative. The organic film is just 0.2-0.5 microns thick and doesn't modify pad elevation at all. The tradeoffs are real: OSP has a shorter shelf life (generally 6-12 months), demands mindful handling, and the film burn throughout the initial reflow, revealing bare copper.
There are situations where HASL remains the right telephone call: through-hole-dominant boards, coarse-pitch designs, and products where numerous reflow cycles and lengthy storage space are more crucial than fine-feature monotony. The key is making that selection deliberately, based on the actual component set-- not relenting by default.
Let's put the numbers side-by-side, because this is where the choice really obtains made.
On a run of 5,000 boards, that's someplace in between $500 and $2,500. A meaningful number, yet small contrasted to what adheres to.
A solitary BGA solder bridge that gets away examination costs an end product. A 14% panel failure rate-- the one I started this short article with-- prices remodel labor, replacement components, expedited shipping, and lost manufacturing capability. In that customer's instance, the savings from choosing HASL had to do with $550 on that order. The costs were $19,000. The ratio does not enhance with volume; it becomes worse, due to the fact that field failings arrive after the item ships.
If you're specifying a PCB surface area coating and the design consists of a fine-pitch BGA, ask the board home one concern: what does the pad coplanarity spec appear like with this finish? If they can't respond to with a number, or if the response entails "it depends upon where the air blades struck," you have your response.
The flat surface area isn't a high-end on a BGA board. It's the entire factor.

Article written by Zeon
Hi, I'm Zeon — 20 years in PCB and electronics manufacturing. Front-end design and R&D, components sourcing, precision SMT, DIP through-hole assembly, and complete unit assembly. That's the full path from concept to finished product, and it's the path I've walked for two decades.
Hot News2026-09-17
2026-09-09
2026-09-07
2026-09-04
2026-09-03
2026-09-01
2026-08-31
2026-08-27