Why Degreasing Decides Whether Paint Sticks
The film nobody can see
Frying at 350 degrees throws oil into the air as droplets fine enough to stay aloft for minutes. They settle on every surface in the room, including doors on the far wall, and they do not stay liquid. Cooking oils are unsaturated, so oxygen attacks the double bonds and the deposit slowly polymerises into a soft, tacky, faintly yellow layer that behaves far more like a varnish than like a spill. In a kitchen cooked in daily for a decade, that layer is heavy enough to feel above and beside the range and still detectable by touch twelve feet away. None of it looks like dirt. It looks like a slightly dull cabinet.
Why a coating pulls away from it
Wetting is a contest between two numbers. A substrate has a surface energy, a liquid coating has a surface tension, and the coating spreads and grips only when the substrate's number is the higher of the two. Clean sanded wood or a sound aged finish typically sits around 40 to 45 dynes per centimetre. A cured layer of cooking oil sits nearer 30. Most waterborne cabinet coatings need better than 35 to wet out. Below that threshold the coating stops behaving like a coating: it crawls into islands, opens fisheyes around single specks, or, worst of all, looks flawless and bonds to the grease instead of to the door. That last case passes every visual inspection and lets go within a year, because the weakest layer in the stack was never the paint.
Detergent and degreaser do different jobs
A detergent works by surfactancy. Its molecules carry one end that likes oil and one that likes water, so they surround a droplet and hold it in suspension until it rinses away. That is enough for fresh splatter. It is useless against a polymerised deposit, because there is no droplet left to surround. An alkaline degreaser works chemically instead, running pH up to somewhere between 10 and 12 so the fats saponify, becoming soap, which then rinses. The step people skip is that rinse. Alkaline residue left on a door is a salt layer sitting precisely where the primer needs to bond, and on oak or cherry it will raise grain and darken tannin as well. Two rinses with clean water, changed often, then a final pass with filtered water, since much of the Yakima Valley runs hard enough to leave a mineral haze as it dries.
A solvent wipe has both a place and a trap. Using one rag redissolves the oil and redistributes it as a thinner, more even, entirely present layer. Done properly it takes two rags, one wet to dissolve and one dry immediately behind it to lift, both changed more often than feels reasonable.
Sanding first is the expensive version of the mistake
The instinct is to scuff and start. What abrasive does to a contaminated surface is drive the contamination into it. A 180 grit scratch pattern runs roughly 25 to 40 microns deep, and the paper presses soft polymerised oil into the bottom of every one of those valleys, which is exactly the geometry the primer relies on for mechanical grip. The abrasive also loads and stops cutting within a few strokes, so the door ends up burnished rather than abraded, and a burnished surface is smoother and less receptive than what was there to begin with. Clean, then sand, then clean again to take off the dust the sanding made. That order is not a preference.
How to know a door is clean
The water break test is free and decisive. Flood a horizontal section with clean water and watch it for thirty seconds. A continuous unbroken sheet means surface energy is high and the door is ready. Any beading, any spot where the film retracts and bare surface appears, is contamination, and the test names the location. Two supporting checks earn their time. Drag a white cotton rag wetted with solvent across a suspect area and read it in daylight, where a decade of frying shows as unmistakable yellow brown. Then, before committing a whole kitchen, coat one door, let it cure the full stated time and run a crosshatch adhesion test: eleven blade cuts each way at one millimetre spacing, tape pressed hard across the lattice and pulled back at a sharp angle. Squares that stay put are the only real evidence the preparation worked.