Solder Sphere Quality and Ball Diameter Control for BGA
The solder sphere is the smallest component in a ball grid array package and the one that decides whether the package can be assembled at all. Diameter variation, poor sphericity and surface oxide all surface later as bridging, opens or voids, long after the spheres were accepted into stock. Controlling them starts with the purchase specification and continues through inspection, handling and attachment.
What Solder Spheres Do in the Package
In a ball grid array or chip scale package, the spheres form both the electrical and the mechanical connection between the package substrate and the printed circuit board. Their diameter sets the standoff between the two, and that standoff governs how underfill flows, how the joint relieves stress and how easily the assembly can be cleaned or reworked. A ball that is out of tolerance changes all of these at once.
Each sphere also carries the entire current of its net and the entire mechanical load of its corner of the package. In thermal cycling the joint is the place where the expansion mismatch between the package and the board is absorbed, which is why alloy, diameter and attachment quality are chosen together rather than independently by three different people.
Ball Diameter and Tolerance Windows
The nominal ball diameter is fixed by the package design and by the pitch of the land pattern. It is the tolerance around that nominal value that decides whether assembly succeeds. A typical specification allows a few percent of variation, and the distribution must sit inside the limits, with the supplier reporting its own process capability rather than leaving the customer to assume it.
Diameter errors propagate through the process. If the spheres on one package are larger than nominal, the collapse distance during reflow changes and the standoff falls. If they are smaller, the paste volume may not be sufficient to form a sound joint. Both directions produce defects, so the measurement has to resolve a few micrometres rather than merely confirm that a ball is present.

Sphericity, Waviness and Surface Finish
Sphericity describes how closely the ball approximates a true sphere, and it matters because the sphere is placed into paste or flux and must register on a land. A flattened or teardrop shaped ball sits at an angle, can roll or turn during placement, and may fail to contact the paste at all. At fine pitch the same defect also raises the chance of a bridge to the neighbouring ball.
Surface waviness and roughness are separate characteristics from sphericity. A rough surface carries more oxide and consumes more flux, and it also scatters light in a way that confuses vision based inspection. Suppliers measure sphericity with optical or laser methods and report it as a maximum deviation, and that figure belongs in the purchase specification rather than in a marketing brochure.
Oxide, Contamination and Surface Chemistry
Every sphere arrives with an oxide layer, and the real question is how thick it is. A thin, uniform oxide is normal and is removed by flux during reflow. A thick or uneven oxide prevents wetting and produces a joint that looks sound but is not, which is the most expensive kind of defect because it escapes electrical test and fails in the field.
Contamination is a different problem with the same result. Organic residues from packaging, sulfur from the environment and chlorine from handling all degrade wetting, and none of them is visible in a diameter check. Solderability testing on a sample of the received lot is the practical way to detect them before the lot is released to production.
Incoming Inspection and Sampling
Incoming inspection should combine an attribute check with a variable check. The attribute check confirms that the lot is what was ordered: alloy, nominal diameter, packaging and quantity. The variable check measures a sample for diameter and sphericity and, where the application is demanding, for oxide condition or actual wetting performance.
Sample size and acceptance criteria should be agreed with the supplier and recorded, so that a later dispute is settled by data rather than by opinion. Measuring equipment must be calibrated, because a micrometer used carelessly on soft solder flattens the ball slightly and reports a diameter that is larger in one direction and smaller in the other.
<img src="https://www.gopcba.com/wp-content/uploads/2026/08/steps-in-pcb-assembly.webp" alt="Ball shear test performed on an attached solder sphere” />
Ball Shear and Joint Strength
Ball shear testing measures the force required to push a sphere off its pad, and it is the standard way to judge attachment quality. The value depends on the pad finish, the flux, the reflow profile and the sphere itself, so the limit is set from a baseline built with known good material rather than taken from a generic table.
The failure mode matters as much as the number. A ductile failure through the solder indicates a sound joint, while a brittle failure at the interface points to an intermetallic or contamination problem even when the recorded force is high. Recording the mode alongside the value is what makes the test useful, and it is also what makes the trend visible over time.
Ball Attachment and Placement Accuracy
Placement accuracy determines whether the ball is centred on its land, and small offsets are tolerated because the surface tension of the molten solder pulls the sphere into place. Where the offset is large, however, the ball can bridge to a neighbour at fine pitch, and the risk rises as pitch falls and the process window narrows.
Attachment is usually done with flux rather than paste for fine pitch packages, and the flux must be applied in a controlled volume. Too little flux leaves an unwetted pad; too much produces spatter and, on some packages, movement of the ball during reflow. The attachment process deserves the same qualification effort as any other soldering step in the line.
Storage, Handling and Shelf Life
Solder spheres are supplied in sealed containers, often under nitrogen, and the shelf life printed on the label applies only while the seal is intact. Once the container is opened the material begins to oxidise and to absorb moisture from the air, and the rate depends on the alloy and on the storage environment.
Handling rules are simple but easy to ignore. Containers should be resealed promptly, opened containers should be consumed within the period the supplier states, and spheres exposed for an unknown time should be tested rather than used blindly or scrapped. Inventory rotation keeps the oldest lot moving first, and that single habit prevents most wetting problems in this area. The same discipline applied to component tolerance is described in the guide to component tolerance and reliability.
Rework and Ball Replacement
When a ball has to be replaced, the replacement should match the alloy and the diameter of the original, and the site should be dressed so that the new joint has the same geometry as its neighbours. Mixing alloys by accident changes the melting behaviour and the mechanical properties of that one joint, which then behaves differently from everything around it.
After rework the package should be inspected by the same method used for production, and the reworked joint treated as a risk until it is confirmed good. Reballing procedures are covered in the guide to BGA reballing, and the choice of inspection technique is compared in the guide to X-ray versus AOI inspection.
FAQ
How tight should the ball diameter tolerance be? The package design fixes the nominal diameter and the assembly process fixes the tolerance around it. A window of a few percent of nominal is common, but the correct value comes from the paste volume, the collapse behaviour and the standoff the design needs. Ask the supplier for the measured distribution, not only the specification limits.
Does sphericity really affect assembly yield? Yes, mainly at fine pitch. A distorted ball registers poorly on its land, sits at an angle in the paste and is more likely to roll during placement. The result is a higher rate of bridging and of joints with reduced contact area, which may pass test and still fail in thermal cycling.
Why do spheres stop wetting after storage? Oxide growth and moisture pickup on the surface are the usual causes, and both accelerate when a container is left open in a warm, humid room. A sample solderability check will confirm the condition of the lot before it is committed to a build, and resealing containers promptly prevents the problem from starting.



