Cleaning And No Clean Residue On Assemblies

A no clean process is one in which the flux residue left after reflow is considered harmless and is left on the assembly. The decision is a chemical judgement rather than a cosmetic one, and it depends on the activity of the flux, the amount of residue, the environment the product will see and the way contamination is measured.

This article explains what residue is, why a no clean process sometimes still has to be cleaned, how contamination is measured, and what the measurements mean.

What Residue Actually Is

The residue on a reflowed assembly is a mixture of the flux vehicle, its reaction products, and anything the flux dissolved during soldering. A rosin based flux leaves a hard, electrically benign film that encapsulates its own activators. A water soluble flux leaves an ionic material that must be removed. A no clean flux sits between them: its activators are designed to decompose or to be locked into the residue at reflow temperature so that the film that remains is not conductive and not corrosive.

That design is only valid if the profile actually reaches the temperature at which the activators decompose. A board that sees a short peak, or a region of the board that runs cooler than the rest, leaves an unterminated flux whose activators are still chemically available. The residue then looks identical to a properly processed one but behaves very differently, which is why the profile is part of the no clean specification and not only the flux. The metallurgy that sets those temperatures is discussed under lead free versus leaded solder.

Why A No Clean Process Is Sometimes Cleaned

Three situations force a cleaning step even when the flux is qualified as no clean. The first is a conformal coating, because most coatings do not adhere reliably over a flux film, and the coating then lifts and takes the protection with it. The second is a high voltage or high impedance circuit, where even a benign residue can carry a leakage current that a low impedance digital circuit would never notice. The third is an environment where the residue will be wetted, such as condensation or salt fog exposure.

Where cleaning is required, the flux choice may also have to change. A rosin residue that was designed to be left in place is difficult to remove because it is intended to be insoluble; the process then needs a flux designed for cleaning and a cleaning chemistry matched to it. Choosing the flux and the cleaning process separately, one after the other, usually leads to a combination that removes neither completely. The interaction of a cleaned surface with a coating is described under conformal coating as board protection.

Measuring Ionic Contamination

The standard measurement is a conductivity test. The assembly is rinsed with a defined volume of a solvent mixture, and the change in conductivity of the rinse is converted into an equivalent mass of sodium chloride per unit area. The result is reported as micrograms of sodium chloride equivalent per square centimetre, and it is compared against a limit that comes from the product specification or from the industry standard the product is built to.

The test has known limitations and they are worth understanding before it is used as a control. It measures what the solvent can dissolve, so a residue that is insoluble or encapsulated does not appear. It measures the whole assembly rather than a specific location, so a small contaminated area is diluted by the clean area around it. And it says nothing about what the residue will do in service, only how much ionic material is present. A passing result is therefore necessary but not sufficient.

Surface Insulation Resistance And Other Checks

Where the answer needs to be about behaviour rather than quantity, a surface insulation resistance test is used. A test coupon carrying a comb pattern is processed with the assembly, and the resistance between the teeth of the comb is measured under a bias at elevated temperature and humidity. A clean, well processed coupon holds a high resistance; a residue that is ionic or hygroscopic allows the resistance to fall, and the fall accelerates as moisture is absorbed.

Electrochemical migration is the failure this test is designed to catch. Under bias and humidity, metal dissolves at the anode, migrates through the residue and plates out at the cathode as a dendrite that eventually bridges the gap. The process needs ionic contamination, moisture and a voltage, and the test applies all three deliberately. The spacing of the comb pattern sets the sensitivity: teeth that are close together fail first, which is why the test is run at the finest spacing the product uses. Protecting a board once it has been cleaned is a related question, covered under potting and dispensing adhesives.

Cleaning Process Choices

Cleaning is done with water and a saponifier, with a solvent, or with a semi-aqueous combination. Water based processes are the most common because they handle the ionic residues of a water soluble flux well, but they require rinsing with deionised water and drying that is thorough enough to remove water from under components. Solvent processes handle rosin residues better and evaporate readily, but they have environmental and cost penalties. The choice follows the flux, not the other way around.

Drying is the step that is most often underdone. Water trapped under a ball grid array or inside a connector housing emerges later, either as corrosion or as a violent escape during a subsequent reflow or coating cure. Where the assembly includes a component that cannot be dried quickly, the cleaning process is designed around it, and the cleanliness specification is written in terms of what can actually be achieved rather than what the ideal would require. The design decisions that make cleaning possible are part of the wider set of rules under design guidelines for manufacturability.

Writing The Specification

The specification for a no clean process should state the flux class, the profile window including the minimum peak and the minimum time above liquidus, the maximum time between printing and reflow, and the contamination limit with the method used to measure it. If cleaning is required, it should state the method, the cleanliness limit and the drying criterion. Stating a flux class without the profile is the most common gap, because the same flux can be benign under one profile and active under another.

The specification should also say which areas are exempt. Connector contacts are often specified as clean because a residue there increases contact resistance, and a board that is otherwise no clean may need a local clean at the connector. Deciding those exemptions at the design stage, rather than discovering them in production, is what keeps the process stable and the measurement meaningful.

FAQ

Is a no clean flux residue conductive? A properly processed no clean residue is not conductive and is designed to be harmless. The same flux processed at too low a peak temperature leaves activators that have not decomposed, and that residue can be both ionic and hygroscopic.

Can a board be cleaned after a no clean process? It can, but the residue may be intentionally insoluble, so a normal wash may remove little of it. If cleaning is going to be required, the flux should be chosen for cleanability from the start.

Does conformal coating remove the need for cleaning? No, and the reverse is true for most coatings. A coating applied over a flux film adheres to the film rather than to the board, and when the film lifts the coating follows it, leaving the circuit exposed.