Adhesive Cure Control in PCB Electronics Assembly Guide
Adhesives hold together more of an assembled board than most people realise: they attach components before wave soldering, they stake heavy parts against vibration, they underfill area array packages, and they seal connectors and housings. In each case the adhesive only does its job if it reaches the degree of cure it was designed for, and that depends on the cure schedule being matched to the material, the dispense volume and the thermal mass of the assembly. Adhesive cure is therefore a process variable rather than a material property, and it has to be verified on the product rather than assumed from the supplier data sheet.
Where Adhesives Are Used
On a mixed technology board an adhesive dot holds the surface mount parts in place through the wave soldering process, and a poor cure means parts that shift, tombstone or fall off into the solder pot. On a high reliability board the same chemistry may be used to stake a heavy inductor or to add mechanical support to a connector.
Underfill is a different application again, where the material is drawn by capillary action beneath a package and then cured to distribute the thermal expansion mismatch between the die, the package and the board. The cure requirement there is set by the need to reach the far corners of a small gap, not simply to reach a temperature. A partly filled package still looks acceptable from the outside, and the weakness only appears after thermal cycling has cracked a solder ball at the corner.
Types of Adhesive Chemistry
Epoxy adhesives dominate, in one part and two part forms, because they bond well to the usual board and component surfaces and they can be formulated for fast cure. Ultraviolet cured acrylics are used where a fast tack is needed and the geometry allows light to reach the joint. Silicones are chosen where flexibility and a wide temperature range matter more than strength.
Each family has its own cure mechanism, and the mechanism dictates the schedule. An epoxy needs heat and time, a ultraviolet material needs the right wavelength and intensity at the bond line, and a moisture curing silicone needs humidity. Applying the wrong energy to any of them leaves the material partly reacted. Storing a two part epoxy below its recommended temperature range, or exposing a light cured acrylic to a wavelength outside its absorption band, has the same effect as shortening the schedule.
The Cure Schedule
The schedule is a combination of temperature and time, and it is normally written as a ramp, a hold and a cool down. The hold is the part that determines the degree of cure, and the supplier usually specifies a minimum condition that applies to the adhesive itself rather than to the board it sits on.
That distinction matters because the board has to reach the temperature before the clock starts. A large copper plane, a thick ceramic substrate or a metal housing all act as a heat sink, so the profile has to be measured on the product rather than taken from the oven display. A travelling thermocouple taped to a real assembly shows the delay, and the hold time should be counted from the point where the bond line reaches temperature. Our profiling guide covers this for heavy copper boards.
Pot Life and Mixing
Two part adhesives begin to react as soon as they are mixed, so the pot life becomes a real constraint on a production line. A batch that is used beyond its pot life will have a higher viscosity, a shorter working time and a cure that starts from an already advanced state, which changes both the dispense volume and the final properties. A simple log of mix time and batch number is enough to catch the problem before a whole shift of boards is bonded with material that was already reacting in the syringe.
Mixing itself is a source of variation. An incomplete mix leaves soft streaks in the cured material that behave as stress raisers, and an over mixed batch can be full of entrained air. Metering and mixing equipment removes most of this variation, but the ratio and the dispense pressure still have to be checked. Weighing a dispensed dot on a scale is a quick check that the volumetric ratio is still where it was when the equipment was set up.
Curing Underfill and Encapsulants
Underfill has to flow into a gap measured in tens of micrometres before it cures, so the cure schedule is normally a two stage affair: a short period at a moderate temperature to let the material flow and wet, followed by the full cure at a higher temperature. Starting the cure too early stops the flow and leaves voids. The flow stage is judged by watching the fillet creep along the package edge, and the transition to the cure stage should be set with a margin beyond that point.
Encapsulants and potting compounds share the same problem on a larger scale. A large mass of resin generates heat as it cures, and in a thick section that exotherm can raise the internal temperature well above the oven setting, which can crack the material or damage the components it surrounds. Thin sections and a slow cure at a lower temperature are the usual answers, and the exotherm should be measured rather than estimated for any large potted assembly.
Bond Strength and Its Measurement
The strength of a bonded joint is measured by a lap shear or a die shear test on a prepared specimen, and the result depends on the cure as much as on the adhesive. An under cured joint typically fails cohesively in a soft, rubbery way rather than with a clean fracture. A cohesive failure inside the adhesive is the sign of a good joint, while an interfacial failure points at surface preparation or at contamination rather than at the cure.
For production control a simpler test is often used, such as a push off test on a dispensed dot or a torque test on a staked component. Our defects guide describes how mechanical failures are classified at gopcb when they appear after assembly. Recording the failure mode alongside the strength value turns a routine test into a diagnostic tool for the assembly line.
Outgassing and Contamination
Uncured or partly cured adhesive releases volatiles when it is heated, and those volatiles can condense on contacts, on optical surfaces or on the inside of a sealed enclosure. In a sealed module the effect can be severe enough to cause leakage currents or intermittent contact resistance.
A full cure removes most of this risk, and a post cure bake is often specified for exactly that reason. The same bake also relieves stress, which reduces the chance of a crack forming at the edge of a large bonded area. Venting a sealed enclosure, or specifying a low outgassing material, is the alternative where a bake would damage the components already fitted.
Defects and Their Causes
The common defects are components that move during wave soldering, adhesive that has spread onto a pad, voids under a package and a joint that fails mechanically after thermal cycling. Each of these has a different root cause, but several of them come back to the cure rather than to the dispense step.
Adhesive on a pad is the one defect that is immediately visible, and it also points at the dispense program. Component movement usually points at a cure that was too slow or a dot that was too small, while voids under a package point at a flow stage that was cut short. A fourth case is a soft joint that passes at room temperature and fails in the thermal cycle, which almost always indicates an incomplete adhesive cure.
Process Control Points
The controls are the material lot and expiry date, the mix ratio for two part systems, the dispense volume and pattern, the measured profile on a production assembly, and the cure oven temperature verified with a travelling thermocouple. Our production flow guide places these checks in the wider assembly sequence. Every one of those items is a measurement rather than an opinion, which is what makes the process repeatable across shifts and operators.
A cure that is verified by measurement rather than by trusting a dial is what separates a stable process from one that produces intermittent failures. Our component guide explains the related storage and shelf life controls that keep the materials within specification. With those controls in place the adhesive stops being a source of intermittent faults and becomes a predictable part of the assembly process.
FAQ
How do I know the adhesive is fully cured? Measure the temperature at the bond line on a real assembly rather than relying on the oven setting, and confirm the result with a shear test or a glass transition measurement on a sample.
Why do components move during wave soldering? The adhesive dot either did not reach full cure or was too small for the component mass. Both show up the same way, so the dot size and the profile should be checked together.
What causes voids under an underfilled package? The cure started before the material finished flowing. A two stage schedule with a flow period at moderate temperature gives the resin time to reach the corners.
