Ice Cream Overrun, Fat Globule Clustering and Texture

Industrial dairy science formalised overrun measurement in the early twentieth century as a quality-control metric for commercial ice cream production. The underlying fat-globule physics were mapped in detail by food physicists in the 1980s and 1990s, later synthesised for kitchen application in Modernist Cuisine (Myhrvold, Young, Bilet, 2011) and in McGee's On Food and Cooking (2004).

Overrun is the percentage increase in volume that air incorporation gives your mix during churning. A base that starts at one litre and finishes at one and a half litres has 50% overrun. That number is not arbitrary decoration — it is the single most direct dial you have over final texture. Commercial soft-serve runs 80–100% overrun. Premium gelato holds 20–35%. Most restaurant ice creams sit between 30–60%, depending on fat content and intent. The mechanism driving texture is not just air bubbles. It is what happens to fat globules during churning. Raw cream holds fat inside intact globule membranes. As the dasher works the mix at sub-zero temperatures, mechanical shear destabilises those membranes and causes partial coalescence — fat globules cluster into a loose network without fully merging. That network does two things: it stabilises the air bubbles that give overrun, and it creates a semi-solid fat matrix that registers as body and richness on the palate. Modernist Cuisine Volume 3 (pages 234–241) spends considerable time on this fat-globule clustering mechanism, noting that ageing your base 4–24 hours at refrigerator temperature before churning allows the fat to crystallise partially inside the globules, which makes partial coalescence more efficient during the churn. Skip that ageing step and your globules skid past each other rather than clustering; you lose structural integrity and end up with either a greasy texture or a weak, icier product. Temperature during churning matters just as much. If your batch freezer canister is not cold enough at the start, the mix warms the barrel instead of the barrel cooling the mix, you get foam rather than a structured frozen emulsion, and overrun becomes meaningless because the air cells collapse the moment you pull the product. Fat percentage and emulsifier load set the ceiling on how much overrun your base can sustain before the structure breaks down. Egg-yolk lecithin from a full-yolk custard base gives you more emulsifier headroom than a Philadelphia-style (no-egg) base, which is why high-overrun gelato-style products usually lean heavier on yolk or add a secondary emulsifier such as mono- and diglycerides.

Fat-soluble aroma compounds — lactones (coconut, peach notes in cream), diacetyl (butter character), and vanillin if present — are carried primarily in the fat phase. When fat globules cluster correctly and coat the tongue during meltdown, they release those compounds in a wave that corresponds with the cool melt sensation. Overrun affects this: high overrun dilutes aroma concentration per unit volume; low overrun concentrates it. A 25% overrun gelato will deliver more intense dairy and flavour-base character than the same formula at 60% overrun. Sucrose and other sugars suppress freezing point and also interact with aroma perception — McGee (2004, Chapter 2) notes that sweetness enhances perception of fat-carried volatile esters. Egg-yolk phospholipids contribute their own sulphur-adjacent notes at high concentrations, which is why heavy-yolk custard bases have that cooked-custard signature that some styles want and others avoid.