SMD Preconditioning Chamber Per JESD22 A113

What preconditioning is for

A semiconductor package fresh from the line is not the part that will sit on a board for years. Between the two lies shipping, storage in humid air, and the violence of being soldered down, and all of it changes the part before its working life begins. JESD22-A113 puts that change in before the reliability test rather than pretend it never happened, so the test judges a part in the state it will truly be in. It is the step that makes every test after it honest.

The story it reproduces

The sequence retells a small disaster that plastic packages are prone to. A nonhermetic package, sealed in moulding compound rather than metal or ceramic, slowly drinks moisture from the air it sits in during storage, the water creeping into the plastic and the interfaces inside. Then the part is soldered to a board, and reflow soldering means heating the whole package to well above two hundred degrees in seconds. The moisture trapped inside flashes to steam, its pressure spiking faster than it can escape, and in a weak package that steam delaminates the layers or cracks the body outright, the failure the trade calls popcorning for the sound and the bulge it leaves.

A package rated to a low moisture sensitivity level can drink enough water in a few days of ordinary storage to crack on the first reflow, and that risk is the reason the moisture rating and a counted floor life outside the dry bag exist at all. Preconditioning stages exactly that small disaster on purpose, a controlled dose of moisture followed by the heat of reflow, to see whether the package survives the commonest cause of its own early failure.

The moisture goes in cold and turns to steam in the oven.

A sequence, not a single chamber

The word chamber undersells what preconditioning needs, since the flow runs across several machines in turn. It begins with electrical and visual checks to record the part as it starts, then commonly an optional temperature cycle to shake the part loose of any assembly stress, then a high-temperature bake to dry it to a clean baseline, then the moisture soak that loads it to a defined level, then the reflow passes that apply the heat, and finally the inspection that reads what the heat did.

Three distinct pieces of equipment carry the load: a bake oven, a moisture-soak chamber, and a reflow oven, with an acoustic microscope to judge the result. The soak chamber is the one people usually mean by the preconditioning chamber, and it is the one whose precision decides whether the whole sequence is repeatable, but it does its work as one link in a chain.

The bake that empties the part

The sequence starts by drying the part to nothing so the soak can fill it to a known amount. A package arriving for preconditioning carries whatever moisture its history left in it, an unknown and uneven charge, and a test cannot load a defined dose onto an unknown starting point. The bake, commonly a stretch at a hundred and twenty-five degrees for a day or so, drives that residual moisture out and leaves every part at the same dry baseline.

Only from that clean zero can the moisture soak that follows put in a repeatable amount, the same from sample to sample and from lab to lab. The bake temperature is kept below anything that would damage the part, well under the reflow peak, so it removes water without adding stress of its own. The bake is unglamorous and easy to skip in a hurry, and skipping it is how a preconditioning result stops meaning anything.

The moisture soak, the heart of it

The soak is where preconditioning earns its name, and where the chamber matters most. After the bake leaves the part bone dry, it is held in warm, damp air for a fixed time so that moisture diffuses into the plastic and settles at the level the part would reach sitting on a shelf or a factory floor for its rated life. The conditions are tied to the moisture sensitivity level being proved: a milder level is soaked at a gentle thirty degrees and sixty percent humidity for many days, a tougher qualification at a warmer, wetter, accelerated setting that packs the same uptake into a shorter run. What the part absorbs is the whole point, so the chamber has to hold its temperature and humidity dead steady for the entire soak, since a few degrees of drift or a humidity that sags overnight changes how much water goes in and quietly turns a Level 3 soak into something easier or harsher than the standard names. It has to keep that climate even across a full tray of parts so one in the corner takes up as much as one in the middle, and the soak is timed to the hour, because the moisture trapped in the plastic is exactly what the reflow that follows will try to turn against the part.

The reflow that springs the trap

Then the heat comes, and the soak turns against the part.

With the moisture in, the reflow passes apply the heat that turns it dangerous. The parts are run through a reflow oven that follows the soldering profile the standard names, a controlled ramp up to a peak that for lead-free solder sits near two hundred and sixty degrees, a brief time held above the roughly two hundred and seventeen degrees where lead-free solder melts, perhaps sixty to ninety seconds above that liquidus, and a cool down, the same thermal trip a board makes through a production reflow line.

The sequence runs the parts through that profile three times rather than once, standing in for a board soldered on both sides and perhaps reworked, so the package meets the heat as many times as a real assembly might give it. Each pass flashes the absorbed moisture toward steam and works the interfaces, and a package that was going to popcorn does it here, in the precondition, rather than later in the field. The reflow is the moment the whole sequence has been building toward.

What the heat leaves behind

The damage preconditioning provokes is mostly internal and quiet. The steam pressure can lift the moulding compound away from the die or the lead frame, opening a delamination that no outside look would catch, or it can crack the package body where the bulge of trapped vapour finds a weak corner. A delaminated interface is the seed of later failure even when the part still works, since the gap it leaves is a path for moisture and a place for cracks to grow once the reliability test or the field gets to work on it. The visible popcorn crack is the obvious case; the hidden delamination is the one preconditioning is truly there to find, the flaw that would otherwise pass a quick electrical check and fail months later.

Reading the result

Seeing inside a sealed package without opening it falls to sound. Scanning acoustic microscopy sends ultrasound into the part and listens for the echo a delamination throws back, mapping the bonded and unbonded areas of every interface before the soak and after the reflow. The difference between the two scans is the measurement that matters, showing exactly what the moisture and the heat did to the interfaces inside.

Why it comes first

Preconditioning sits at the front of a qualification for a reason of logic. The reliability tests that follow, the damp-heat bias soak, the thermal cycle, the autoclave, all assume they are stressing a part in its real working state, and a part that has not been through assembly is not in that state. Running the precondition first puts the part into the worst realistic condition a customer could receive it in, and only then does the reliability test begin, so its result speaks for parts in the field rather than for pristine samples.

A qualification that skipped preconditioning would flatter its parts, passing packages that would crack the first time a customer reflowed them onto a board. The order is the whole point: the abuse of assembly comes before the test of endurance, exactly as it does in the world the part is built for. Skip it, and a damp-heat or thermal-cycle result describes a sample no customer will ever hold, flattering the package by testing it in a condition kinder than its first day on a board.

Run it, and every number that follows speaks for the part as it will truly arrive, already aged by the journey and the soldering iron. The sequence is the gate every other test stands behind, and the order of the gate is the whole reason the qualification means anything.

What the soak chamber has to hold

For the chamber at the centre, the demand is steadiness over flair. It needs to hold its set temperature and humidity within tight bounds across a soak that can run for days, recover quickly when the door opens to load or unload, and keep the condition uniform so a tray of parts in the corner takes up the same moisture as a tray in the middle. It carries no bias, no pressure, and no fast transitions, the simple opposite of the harsher chambers in the catalogue, and its whole virtue is constancy.

A soak chamber that holds thirty degrees and sixty percent humidity dead level for a week, corner to corner, is doing precisely the unglamorous job the standard asks, and doing it is what lets a preconditioning result mean the same thing everywhere.

Pulling it together

Preconditioning per A113 ages a surface-mount part through its own assembly before the real testing starts: a bake to dry it to a clean baseline, a precise moisture soak to load it to the level its rating allows, and three passes of reflow heat to flash that moisture toward steam and spring the popcorn trap. The flow runs across a bake oven, a soak chamber, and a reflow oven, with an acoustic scan to read the hidden delamination the heat leaves behind, and the soak chamber's steady temperature and humidity are what make the dose repeatable. Run first, before every other reliability test, it puts the part into the state a customer will truly meet, so that everything measured afterward speaks for the field rather than for the factory.