A custard base is cooked for the emulsifier as much as the thickening

What the yolk brings besides thickness

Egg yolk contains proteins that coagulate with heat, which is the property every recipe mentions, and it also contains phospholipids, principally lecithin, which are natural emulsifiers. Those molecules position themselves at the boundary between fat and water and hold the two together. In a frozen dessert that second contribution is arguably the more important one, because a well-emulsified base freezes into a smoother, more homogeneous structure than an unemulsified one ever will.

Fat in a dairy mixture exists as droplets suspended in water, each wrapped in a membrane. Yolk lecithin partially displaces that natural membrane, leaving the droplets more able to interact with one another during churning. That interaction is what builds the fat network that gives churned ice cream its body. So the yolk is not merely thickening the liquid before it goes into the machine — it is preparing the fat for what the machine will do to it.

The window between thickened and scrambled

Yolk proteins begin to set somewhere around sixty-five degrees Celsius on their own, but in a custard they are diluted by milk and shielded by sugar, both of which push the setting temperature higher. In practice a custard base thickens usefully between about seventy-five and eighty-five degrees and curdles above that, and the margin is small enough that a distracted minute over direct heat is genuinely enough to ruin it.

Sugar is doing more of the protecting than most people realise. It competes for water and physically interferes with proteins finding each other, so a sweet custard tolerates more heat than an unsweetened one. This is why an ice cream base is more forgiving than a plain savoury sauce made the same way, and why reducing sugar in a base narrows the safety margin in the cooking as well as changing how hard the finished ice cream freezes.

Reading the thickening rather than trusting a time

The traditional test asks you to coat the back of a spoon and draw a finger through it, looking for a line that holds rather than closing up. It is a viscosity test, and viscosity is what actually tells you how far the proteins have gone, so it is a better instrument than the clock. A thermometer is better still if you have one, and holding the base at a target temperature for a couple of minutes gives a slightly thicker result than touching that temperature briefly.

Tempering the yolks by adding a little hot milk before combining is not ceremonial either. Yolks introduced directly into hot liquid set locally, in lumps, before they have dispersed, and once a protein has coagulated no amount of whisking will unset it. Raising their temperature gradually keeps the setting distributed evenly through the mixture. Straining afterwards is a sensible last line of defence rather than an admission of failure.

Ageing the base changes it further

Chilling a cooked base for several hours before churning is standard in commercial practice and worth the wait at home. Two things happen during that rest. The fat crystallises within its droplets, which is a prerequisite for the partial coalescence that builds structure during churning, and the proteins and any stabilisers present hydrate fully, which raises viscosity beyond what the cooking achieved on its own.

The practical effect is a base that churns faster and freezes smaller. A warm base entering the machine spends its first minutes simply losing heat, and every minute of that is time during which large ice crystals can form near the barrel wall. There is a limit to the benefit, and a base left several days will eventually pick up refrigerator odours and lose freshness, so overnight is the conventional compromise for good reason.

Where the eggless versions stand

Philadelphia-style ice creams skip the custard entirely, relying on cream, sugar and flavour alone, and they are not inferior so much as different. Without yolk the flavour of the dairy and the flavour added to it come through more cleanly, which suits fresh fruit and delicate infusions. The trade is a thinner body and a shorter useful life in the freezer, since there is less emulsifier and less dissolved solid to hold the structure together.

Modern eggless formulations close much of that gap using milk powder for solids and small quantities of gums or commercial emulsifiers for the rest. That is a legitimate approach rather than a shortcut, and in some respects it is more controllable, since the maker can adjust each function independently instead of getting all of them bundled into one ingredient. But an egg custard base contributes a flavour of its own, and no substitute reproduces that part.