Milk Foam Simulator: Understanding Silky Microfoam
Silky microfoam is no accident — and above all, it isn’t a question of the stopwatch. Microfoam is foam whose bubbles are so fine that you can no longer make out a single one. The surface then doesn’t look foamy at all; it shines like wet paint. It comes from the interplay of milk protein, fat, temperature and steam. Try it out interactively here: choose the milk type, how much air you work in, the final temperature and the steam power — and see how microfoam, volume, perceived sweetness and creaminess change.
The sliders: Milk type has four buttons — whole milk (~3.6 % fat), skimmed milk (~0.1 % fat), and barista-grade oat and soy drinks. Work in air (stretching) is a scale from 0 to 100 without a unit: 0 means no air at all, 100 means as much as you possibly can. These are neither millilitres nor seconds. Final temperature is the only slider with a real measured quantity: 45 to 75 °C in steps of 1 °C. Steam power has three levels (low / medium / strong) and stands for the power of your steam wand — again without a unit, because steam power can’t meaningfully be expressed as a single number.
The results: foam volume, microfoam, creaminess and perceived sweetness are scores from 0 to 100, not measurements and not lab results. They show you the direction your milk foam is moving in, not how it will taste. The “Goes with” line isn’t a verdict but a pointer: it names the drink that this amount of foam usually belongs to. The thermometer in the animation also shows a value calculated by the model, not a measurement taken on your jug.
The basis: that more protein and a higher casein share make foam more stable comes from dairy research (Borcherding et al. 2009); that fat reduces the amount of foam, from Kamath et al. (2008); the temperature threshold, from measurements on heated milk (Yang et al. 2021). The figures the tool calculates with, on the other hand, are reasoned assumptions, not measured curves — for example, that quality declines evenly from 65 °C up to 77 °C. It’s a model that shows the direction. And it shows it reliably.
This is a simulator. It calculates with a simplified model and shows you relationships — not measurements and not a promise. What ends up in your cup depends on your machine, your water, your beans and the way you work.
Our recommendations are based on experience from consultations, courses and our showrooms. They are non-binding and given without guarantee. We accept no liability for any damage resulting from their use. If in doubt, follow what your manufacturer says in the operating manual.
The milk foam simulator is loading … If nothing appears, please reload the page — or have a look at milk pitchers.
Straight to the results: foam volume, microfoam, creaminess and perceived sweetness are scores from 0 to 100 from our model — not measurements. And the degrees on the thermometer in the animation are a value calculated by the model, not a measurement taken on your jug or your machine.
The science behind the foam
Milk foam is air trapped in a network of milk protein. Most of this protein consists of tiny clusters floating freely in the milk — so small that several hundred of them side by side only add up to the thickness of a hair. The technical term for them is casein micelles. They wrap themselves around every air bubble like an elastic skin and hold it in place. The second milk protein, whey protein, is much less good at this. Hence the rule: the more protein overall, and the larger the casein share of it, the more stable the foam.
Fat gets in the way of foam formation — but not always to the same degree. It holds foaming back most between about 15 and 45 °C: in that range the milk fat is partly solid, and the semi-solid fat globules tear freshly formed bubbles open again. Above roughly 45 °C the fat is completely liquid, and whole milk foams noticeably better. That’s why the common saying “skimmed milk foams with more volume and more stability” is only half true. It applies to cold and lukewarm milk. Right in the window where you actually steam — 55 to 65 °C — whole milk catches up, and skimmed milk foam loses stability: in the same series of measurements, skimmed milk foam was most stable at 45 °C and increasingly less so above that. What remains: skimmed milk gives more volume and a drier, stiffer foam; whole milk gives less volume and a fuller, creamier texture in the mouth. The tool calculates with the simple picture (skimmed milk foams a little more readily and is a little more stable); the temperature dependence isn’t built into it — worth knowing when you compare the bars. There’s no right or wrong here: it depends on which drink and which texture you want.
Temperature is the trickiest lever — and until recently this page listed three different thresholds (60–70 °C, 67–72 °C and 68 °C). We have corrected that. There is one single threshold, the same one as in the Latte Art Trainer: aim for 55–65 °C. From 65 °C it starts to tip. From 70 °C it’s over.
Behind this is the most common whey protein in milk; its technical name is beta-lactoglobulin (written β-lactoglobulin). As long as it stays tightly folded, all is well. In heated milk, nothing measurable has happened yet at 55 °C or at 65 °C; only from about 70 to 75 °C does the protein unfold irreversibly. In the process, a sulphur-containing docking site is exposed that was previously hidden inside — the technical term is thiol group. It’s exactly this site that produces the cooked, slightly eggy smell, and because the protein network loses its shape at the same time, the foam collapses. Between 65 and 70 °C lies the transition: not yet spoilt, but already noticeably less sweet and less stable. That’s why the tool calculates with increasing loss from 65 °C, and the Latte Art Trainer gives an explicit warning from 67 °C.
Sources (scientific): Borcherding et al. (2009), International Journal of Dairy Technology — protein content & casein:whey ratio determine foam stability; Kamath et al., “The influence of temperature on the foaming of milk” (International Dairy Journal) — fat reduces foam formation/stability; Halabi et al. (2020), Foods 9(7):874 — heat denaturation of β-lactoglobulin; Yang et al. (2021), Frontiers in Nutrition 8:714869 — milk at 55 and 65 °C shows no measurable change in free sulphur groups; irreversible unfolding only above that. The weightings in the simulator’s calculation engine are assumptions, not measurements.
How to get silky microfoam — by feel, not by the clock
- Start cold & fresh: cold milk (~4 °C) gives you more time to texture. Fill the pitcher only ⅓ to ½.
- Work in air (“stretching”): steam tip just below the surface — a soft, even hiss. Only briefly and only at the start, while the milk is still cool.
- Texture (“rolling”): steam tip a little deeper, create a whirlpool. No more air now — this is how large bubbles break down into fine microfoam.
- Temperature by feel: hand on the jug. At ~55–65 °C it’s warm, without cooked notes. No timer — sound, swirl and warmth guide you.
- Tap & swirl: tap the jug briefly, then swirl until the surface shines like wet paint.
- Pour straight away: ready for a flat white, a cappuccino or your first latte art.
What you need
Three things make the biggest difference: a well-shaped milk pitcher (spout & volume shape the whirlpool), a machine with powerful, dry steam — this is where dual boilers play to their strengths, because the steam boiler works independently — and a bit of practice. Without an espresso machine, you can also make creamy foam with the Subminimal NanoFoamer. How the brew and steam boilers work together is shown in the interactive boiler animation.
Frequently asked questions about steaming milk
What is the ideal temperature?
55 to 65 °C. In this window the milk comes across as warm and slightly sweet. From 65 °C the quality declines; from around 70 °C the whey protein (β-lactoglobulin) unfolds irreversibly: cooked and sulphurous notes develop, and the foam collapses. We use the same threshold in the Latte Art Trainer. Listen closely: you don’t need a timer for this, just your feel for the jug.
Why does my foam have large bubbles?
Usually too much air is worked in, or it’s worked in too late. Air belongs only briefly at the start (“stretching”); after that it’s all about rolling/texturing — the whirlpool breaks large bubbles down into microfoam. A suitable pitcher and enough steam pressure help.
Whole milk or oat milk?
Whole milk gives a creamier, rounder texture thanks to its fat; skimmed milk gives more volume and a drier foam. The often-heard addition “and more stable” only applies to cold and lukewarm milk — at 55 to 65 °C the difference is small. Barista plant milk (oat, soy) contains added protein and stabilisers and therefore foams much better than the standard version. Try out in the simulator how volume and creaminess differ — and go with what you enjoy.
Why does milk that’s too hot taste “cooked”?
When milk is overheated, the whey protein β-lactoglobulin unfolds and exposes a sulphur-containing docking site that was previously on the inside (technical term: thiol group). These sites bond with one another, and that’s what creates the typical cooked, slightly eggy aromas. In milk this happens from about 70 °C. So it’s better to stop earlier.
You’ll find all our interactive tools together in our Coffee Tools.
