Shrinkage Itself Is Not the Enemy
The previous guide left us holding two ideas worth keeping close. First, as a wet green body dries, the capillary stress of water curving inside its tiny pores draws the particles together and the whole body physically contracts — this is drying shrinkage, and it is large and unavoidable. Second, all of that contraction is crammed into the first, constant-rate act of drying and stops the moment the particles touch. Now we face the sharper question those two facts raise. If a body simply shrank as one obedient whole, drying would hold no terror at all. The danger is born entirely from unevenness: from one region shrinking before its neighbour.
Just how much does a body shrink? Track the one thing that never changes — the solid particles — and let the water leave. Say the fresh body is 55 percent solid by volume (45 percent water), and the particles first touch, at the critical moisture content, when packing reaches 64 percent solid. The total volume falls in step with the solid becoming a larger share of it: the volume shrinkage is 1 - 55/64, about 0.14, or 14 percent. Since a part shrinks in all three directions at once, the linear shrinkage you actually measure is roughly the cube root of that, 1 - (55/64)^(1/3), which is about 0.05 — 5 percent. That is why a plastic clay body famously shrinks 5 to 8 percent on drying, while a barely-damp pressed technical part, starting with far less water, shrinks well under 1 percent.
How Shrinkage Turns into Stress: the Skin-Core Tug-of-War
Picture a thick wall drying in open air. Its outer skin meets the dry air first, races through the constant-rate act, and shrinks — but its core, buried a centimetre deep, is still soaking wet and full-sized. Now the shrunken skin is a coat one size too small, stretched over a body that has not slimmed down. That mismatch is differential shrinkage, and it is exactly the physics behind the polygon cracks in the bed of a dried-up pond: the mud's surface dries and shrinks while the mud below stays plump, and the surface, unable to shrink freely, tears itself into tiles. Your ceramic is a well-behaved version of that same mud.
DRYING A THICK WALL (cross-section, air drying both faces)
face face
air | DRY SKIN | WET CORE | DRY SKIN | air
| shrunk, | still full-size, | shrunk, |
| wants | has NOT shrunk | wants |
| to be | yet | to be |
| smaller | | smaller |
| <-- held | --> pushed <-- | held --> |
| TENSION | COMPRESSION | TENSION |
water: low | high | low
stress: + (pulled apart) | - (squeezed) | +
A brittle skin stretched over a plump core:
the SURFACE is in tension -> it finds its worst
flaw -> a CRACK opens there and drives inward.Follow the forces the mismatch sets up. The skin, wanting to be smaller than the core will allow, is stretched — held in tension. The core, forced to stay larger than the skin wants, is squeezed — held in compression. And here the brittle nature of ceramics bites. A ceramic is strong in compression but weak in tension, and it is only as strong as its worst flaw — a nick, a pore, a scratch from forming — because tension prises that flaw open like a nick at the edge of a sheet of paper. So the tensile skin is the vulnerable face, its ugliest flaw is the trigger, and the crack, when it comes, opens at the surface and drives inward. Note the subtlety worth carrying: what cracks a part is not how wet or how dry it is on average, but the gradient — the difference between skin and core. A uniformly wet part and a uniformly dry part are both perfectly safe.
Three Ways a Part Fails: Warp, Crack, Laminate
The same differential shrinkage can express itself in three different failures, depending on how free the part is to move. The gentlest is warping. If a flat part dries faster on its top face than its bottom — because the top is exposed while the bottom sits on a shelf — the top shrinks more, and the whole piece curls up at the edges like a slice of bread going stale. Nothing tears; the part simply bows to relieve the imbalance. Warping also has a sneakier source that has nothing to do with drying speed: if the green body was already denser in some regions than others, those regions shrink by different amounts no matter how perfectly you dry, and the part warps anyway. That flaw was locked in during forming.
When the part cannot bow freely to relieve the stress, the tension has nowhere to go but into a tear — a drying crack. Cracks announce themselves at the places where stress concentrates: sharp inside corners, the junction where a thick section meets a thin one (they dry at different rates and fight each other), and any rim or feature held rigid while the rest still wants to shrink — a handle fixed to a shrinking cup, or a flange gripping a mould. This is why designers learn to hate sudden changes in wall thickness and sharp reentrant corners: they are drying cracks waiting to happen. Unlike a warp, a crack is a clean structural failure, and it does not heal — the firing to come will only widen it.
The third failure belongs to parts built up in layers. A pressed disc carries faint horizontal planes from the way powder filled the die; a tape-cast sheet or a stack of laminated tapes is layered by design. When drying (or the later burnout) stresses those weak planes, they can peel apart into sheets — a defect called lamination or delamination. It is the same tension at work, but now it finds a ready-made plane of weakness to open rather than having to carve a fresh crack. The lesson braided through all three modes is one: uneven shrinkage is the villain, and it finds whichever exit — bow, tear, or peel — the part's shape and history make easiest.
Why Thick and Nonuniform Parts Suffer Most
Everything so far points a finger at one culprit: the moisture gradient between skin and core. So what makes that gradient steep? Thickness, above all. Water can only crawl out of a body so fast, and the deeper the core, the longer its water must travel and the further behind the skin it falls. Double the wall thickness and, very roughly, the time to dry safely quadruples, because diffusion time scales with distance squared. A thin tile has almost no gradient and dries in minutes with little risk; a thick block dried in the same air develops a savage skin-to-core difference and cracks. It is a size effect with the same flavour as the rule that bigger ceramic parts are weaker — scale up, and the hidden stresses grow faster than your ability to manage them.
The second aggravator is nonuniformity anywhere in the body. If the green part is denser in one region than another — the classic pressing density gradient, where die-wall friction leaves the powder more compacted near the punch than in the middle — then even a body dried with perfect, gradient-free care will still shrink by different amounts in different places, and pull itself crooked. This is why so much of the earlier ladder harped on green-body uniformity: a part that is uneven in density, moisture, or wall thickness before drying begins is already carrying the seed of a warp or crack. Drying does not so much create these flaws as reveal and act on the unevenness that forming baked in.
The Cure: Dry Slow, Dry Even, Control the Humidity
The cure follows straight from the diagnosis: keep the skin and core close in moisture, so no gradient, no tug-of-war, no crack. The master trick is to control the humidity of the surrounding air. Start with the air kept deliberately damp, so the surface is only allowed to give up water as fast as the core can resupply it from within; the whole part then shrinks together, gradient-free, through the whole dangerous constant-rate act. Only once shrinkage is safely finished — past the critical moisture content — do you gradually lower the humidity and raise the temperature to chase out the last, harmless pore water. That staged recipe of humidity and heat over time is the drying schedule, and for a thick, awkward part it can run for days.
- Slow the surface, not just the clock. Tent or wrap the part and keep the surrounding air humid, so the skin can only lose water as fast as the core resupplies it — this flattens the gradient, which is what actually cracks parts.
- Dry every face alike. Raise the part on an open rack and rotate or flip it so no one face races ahead; shield thin edges and corners that would otherwise dry first and pull on the slower body.
- Support against warping. Bed the part flat, or on a contoured setter that matches its shape, and never leave a rim or feature clamped rigid while the rest still wants to shrink.
- Spend your patience before the critical moisture content. This is the only window in which the body shrinks and can crack; lavish the slow, humid care here, because once the particles touch, shrinkage stops and the rest of drying is far safer to speed up.
- Then lower humidity and warm gently. With shrinkage over, drop the humidity and raise the temperature to drive off the last pore water, and only then hand the dry part on toward binder burnout.