HomeBrew Methods › How to froth milk

☕ How to Brew · 2026

How to Froth Milk: Stretch, Swirl, and Stop at 140-149°F

The two phases of texturing milk from Breville's own instruction book: stretch to add air, swirl to make it silky, stop between 140 and 149°F. Plus jug size, plant milks, and how to fix big bubbles.

Written and reviewed by the CoffeeLabAdvisor editorial team · Published 9 September 2026

8 facts, each with a source below 7 questions answered 4 verified products

Texturing milk is two phases that blend into one movement. First stretch: hold the steam tip just at the surface so it draws air in and the milk grows in volume. Then swirl: lower the tip below the surface and keep a vortex spinning so the air folds in and the texture turns glossy. Stop when the jug is hot to touch, which Breville puts at 140-149 °F (60-65 °C). Then tap the jug to collapse big bubbles, swirl to polish, and pour immediately — textured milk separates within a minute of standing still.

On this page
  1. The two phases, and why they are one movement
  2. The method, step by step
  3. Why 140 to 149°F, and what happens past it
  4. The jug, which matters more than the machine
  5. Plant milks, which texture better than their reputation
  6. Fixing the four things that go wrong
  7. Verified products
  8. FAQ
Phase 1Stretch — tip at the surface, adds air
Phase 2Swirl — tip below surface, folds it in
Stop at140-149°F / 60-65°C, per Breville
MilkFresh and cold, straight from the fridge
Jug sizeJust below the spout's V — not full
After steamingTap to collapse bubbles, swirl to polish
ThenPour immediately — it separates as it stands
Plant milksOat and soy texture well; barista editions best
Two phases of texturing milk. Stretching: the steam tip sits just at the surface so air is drawn in and the milk grows in volume. Swirling: the tip is lowered beneath the surface and the vortex folds the air in until the milk is smooth, finishing between 140 and 149 degrees Fahrenheit. Phase 1 · Stretch Tip AT the surface. Volume grows. This is the only air you get. Phase 2 · Swirl Tip BELOW the surface. The vortex folds the air in until it is glossy. 149°F 140°F Stop here 60–65°C Both phases blend into one continuous movement; the temperature is Breville's own figure.

Stretch then swirl. The tip's position relative to the milk surface is the whole technique; the temperature band is Breville's own figure.

Steamed milk poured into a glass of espresso
Correctly textured milk pours as one glossy body, not as foam sitting on liquid. Photo: Lex Sirikiat / Unsplash
A cappuccino with a thick foam cap
A cappuccino wants more air than a latte — the same technique, stretched for longer. Photo: Lex Sirikiat / Unsplash

The two phases, and why they are one movement

Breville's instruction book states the structure more clearly than most explanations manage: “There are two phases to texturing milk. The first is stretching the milk to aerate it and the second is swirling the milk to make it silky smooth. These two operations should blend into one.”

Stretching is the only point at which air enters the milk. The tip sits at or just below the surface, and you will hear a light tearing or hissing sound as it pulls air in. Volume grows. This is the phase that decides whether you are making a latte or a cappuccino: a couple of seconds for a latte, longer for a cappuccino's thicker cap.

Swirling adds no air. The tip goes deeper and the jug rotates the milk into a vortex, which folds the air already there into smaller and smaller bubbles until the surface looks like wet paint rather than bubble bath. Skip this phase and you get stiff foam sitting on hot milk — two layers instead of one drink.

The reason they must blend is temperature: you only have until the milk hits about 149 °F to finish both, so stretching too long leaves no time to smooth what you made.

The method, step by step

These steps follow Breville's own sequence for a machine with a steam wand. Details differ between machines, but the physics do not.

  1. Start cold and fresh. “Always start with fresh cold milk.” Cold milk gives you a longer working window before the temperature ceiling; milk that has already been heated once does not texture again.
  2. Fill the jug correctly. Breville's instruction is to “Fill the jug just below the ‘V’ at the bottom of the spout.” The milk will grow, so a full jug overflows and a nearly empty one cannot form a vortex. De'Longhi gives volumes for its own machines: 3.5 oz for a cappuccino, 6 oz for a latte.
  3. Purge the wand first. Point it into the drip tray and run steam briefly. Condensed water collects in the wand between uses and you do not want it in the milk.
  4. Insert and start the spin. Breville's positioning is specific: the tip “¼-¾” (1-2cm) below the surface of the milk, close to the right hand side of the jug at the 3 o'clock position”. Off-centre is what creates rotation. Keep it there “until the milk is spinning clockwise, producing a vortex (whirlpool effect)”.
  5. Stretch. “With the milk spinning, slowly lower the jug. This will bring the steam tip to the surface of the milk & start to introduce air into the milk.” Keep the vortex going while you do it.
  6. Swirl. “Lift the jug to lower the tip beneath the surface, but keep the vortex of milk spinning.” The sound changes from tearing to a low rumble.
  7. Stop at temperature. “The milk is at the correct temperature (140-149°F or 60-65°C) when the jug is hot to touch.” Turn the steam off before lifting the tip out, or you will spray milk across the machine.
  8. Wipe the wand immediately, then purge it again. Milk bakes onto a hot wand in seconds and is unpleasant to remove later.
  9. Tap and swirl. “Tap the jug on the bench to collapse any bubbles”, then “Swirl the jug to ‘polish’ and re-integrate the texture.”
  10. Pour immediately. Breville's closing note is that “The key is to work quickly, before the milk” separates. De'Longhi says the same about waiting: pour straight away or the milk and foam come apart.

Why 140 to 149°F, and what happens past it

Breville's band of 140-149 °F (60-65 °C) is narrow for a reason, and the useful part is knowing what sits on either side of it.

Below it, the drink is warm rather than hot, and the sweetness has not developed — milk tastes noticeably sweeter warmed into this band than it does cold or lukewarm.

Above it, texture and taste both fall away. The foam you built starts to collapse, and the milk takes on a distinctly cooked, thin flavour that no amount of coffee hides. This is the single most common mistake in home milk frothing, and it happens because people wait for the milk to feel very hot through the jug rather than merely hot.

The practical answer is your hand: Breville's own cue is that the jug is “hot to touch” — the point where you want to let go but can still hold it. If you would rather not learn that by feel, a clip-on thermometer costs under $10 and teaches it in about three drinks.

CDN ProAccurate Frothing Thermometer (IRT220-F)
ThermometerCDN ProAccurate Frothing Thermometer (IRT220-F)

The cheapest way to stop guessing: it clips to the jug so you learn what the right temperature feels like.

$8.50Live Amazon price · as of 09/12/2026 12:03 AMCheck price on Amazon →

The jug, which matters more than the machine

A frothing jug is not a decorative accessory. Three things about it change the outcome.

Size relative to the drink. The milk needs room to grow, and the vortex needs depth to form. A 12 oz jug is right for one drink; a 20 oz jug is right for two. Using a 32 oz jug for a single latte is the most common reason a home barista cannot get a vortex started — there is not enough milk for the tip to reach and spin.

A narrow spout. This is what lets you pour a controlled stream, and it is the difference between pouring latte art and pouring a puddle. It matters even if you have no interest in art: a wide spout dumps foam first and milk second.

Metal, and thin. Stainless steel lets you feel the temperature through the wall, which is the cue Breville tells you to use. A double-walled or plastic jug insulates your hand from the one measurement you have.

De'Longhi DLSC060 Milk Frothing Jug, 12 oz
Milk pitcherDe'Longhi DLSC060 Milk Frothing Jug, 12 oz

12 oz is the size for one or two drinks - a jug much bigger than the drink will not texture properly.

$19.95Live Amazon price · as of 09/12/2026 12:03 AMCheck price on Amazon →

Plant milks, which texture better than their reputation

Oat milk textures genuinely well — it holds a stable microfoam and its body carries through espresso in a way that suits flat whites and lattes particularly. Soy textures well too and has done for decades in cafés. Almond is thinner and gives a lighter foam that collapses faster, which suits a cappuccino's airier cap more than a latte's glossy body.

Barista editions are worth buying, and not as a marketing exercise: they are formulated with slightly more fat and added stabilisers precisely so the foam holds, and the difference from a standard carton is obvious in one drink.

Two adjustments help across all of them. Plant milks tend to reach their best texture at the lower end of the temperature band, so aim nearer 140 °F than 149 °F. And stretch a little less: plant milks aerate faster than dairy, so the same few seconds gives you more volume than you wanted.

One thing worth knowing about shelf-stable cartons, and De'Longhi says it about milk generally in its own guide: “some milks that do not require refrigeration will not produce an acceptable foam”. That is a fair caution for UHT dairy and for some ambient plant milks — but it is a property of the individual product, not of plant milk as a category, and the barista editions exist precisely to solve it.

Every machine review here carries a plant-milk rating, and our frother guide covers the standalone gear.

Fixing the four things that go wrong

What you gotCauseFix
Big soapy bubblesStretched too long, or the tip stayed at the surfaceGet the tip under sooner and keep the vortex going; tap and swirl afterwards
Hot milk, no foamThe tip never reached the surface, so no air went inLower the jug at the start until you hear the tearing sound
Foam and milk in two layersNo swirl phase, or the milk stood before pouringSwirl until glossy, then pour immediately
Thin, cooked tasteWent past about 149 °FStop when the jug is hot to touch, not when it burns

If the wand loses power partway through, that may be the machine rather than you: Breville notes that “After 5 minutes of continuous steaming, the machine will automatically disable the steam function.” And on a single-boiler machine, brew the espresso before steaming — one heater cannot do both at once, which is why our De'Longhi operating guide puts the order first.

Verified products for this page

ℹ️ Affiliate disclosure: CoffeeLabAdvisor is an Amazon Associate and may earn a commission from these links, at no extra cost to you. Commissions play no part in what appears here — every product traces to a verified catalog entry or a published review.

A jug sized to the drink, something to read the temperature with, and a wand for anyone without a steam-equipped machine. None carries a score; the claims are the makers' own.

De'Longhi DLSC060 Milk Frothing Jug, 12 oz
Milk pitcher
De'Longhi

DLSC060 Milk Frothing Jug, 12 oz

$19.95Live Amazon price · as of 09/12/2026 12:03 AM

12 oz is the size for one or two drinks - a jug much bigger than the drink will not texture properly. Listing states a 350 ml (12 oz) stainless steel frothing jug with a pour spout, compatible with all De'Longhi pump espresso makers.

Check price on Amazon →
ZOWIE KING Milk Frothing Pitcher, 20 oz
Milk pitcher
ZOWIE KING

Milk Frothing Pitcher, 20 oz

$12.99Live Amazon price · as of 09/12/2026 12:03 AM

The two-drinks-at-once size, with a narrow spout for pouring rather than dumping. Listing states a 304 stainless steel 20 oz pitcher with liquid measurements engraved on both sides and a narrow spout, described as suited to two or three cappuccinos.

Check price on Amazon →
CDN ProAccurate Frothing Thermometer (IRT220-F)
Thermometer
CDN

ProAccurate Frothing Thermometer (IRT220-F)

$8.50Live Amazon price · as of 09/12/2026 12:03 AM

The cheapest way to stop guessing: it clips to the jug so you learn what the right temperature feels like. Listing states a 0 to 220 F instant-read frothing thermometer with a 1 inch magnified dial, a 5 inch stem described as suited to small frothing pitchers, and an adjustable clip.

Check price on Amazon →
Zulay Kitchen Milk Frother Wand
Milk frother
Zulay Kitchen

Milk Frother Wand

$13.99Live Amazon price · as of 09/11/2026 11:57 PM

The $13 entry point - froths heated milk in seconds, and the classic coffee-lover stocking stuffer. Listing states a battery-powered handheld electric whisk foam maker for coffee, lattes, cappuccino and matcha, with Duracell batteries included.

Check price on Amazon →

Frequently asked questions

What temperature should milk be frothed to?

Breville's own instruction book gives 140 to 149 °F (60 to 65 °C), and says the milk is at the correct temperature when the jug is hot to touch. Past that band the foam starts to collapse and the milk takes on a cooked, thin taste.

What are the two phases of frothing milk?

Stretching and swirling. In Breville's words, the first is stretching the milk to aerate it and the second is swirling it to make it silky smooth, and the two should blend into one movement. Air only enters during stretching; swirling folds that air into smaller bubbles.

Where should the steam wand tip go?

Breville positions it a quarter to three quarters of an inch (1 to 2 cm) below the surface, close to the right-hand side of the jug at the 3 o'clock position. Off-centre is what starts the milk spinning into a vortex, which is what makes the texture smooth.

Why does my milk have big bubbles?

Too much stretching, or the tip stayed at the surface too long, so large bubbles formed and were never folded in. Get the tip under the surface sooner while keeping the vortex spinning, then tap the jug on the bench to collapse the bubbles and swirl to polish it.

How full should the milk jug be?

Breville says to fill it just below the V at the bottom of the spout. The milk grows as you texture it, so a full jug overflows — and a jug that is too large for the drink prevents a vortex forming at all, which is a common reason home frothing fails.

Can you froth oat milk and soy milk?

Yes, and both texture well — oat particularly, holding a stable microfoam with body that suits lattes and flat whites. Barista editions are formulated with stabilisers so the foam holds, and the difference is obvious. Aim for the lower end of the temperature band and stretch a little less, because plant milks aerate faster than dairy.

Do I need a thermometer to froth milk?

No — Breville's own cue is that the jug is hot to touch, and that is learnable. A clip-on thermometer costs under $10 and teaches the feel in about three drinks, which is a reasonable shortcut rather than a requirement.

Sources for this page: The two-phase description, 'always start with fresh cold milk', the fill level just below the V, the tip position a quarter to three quarters of an inch below the surface at the 3 o'clock position, the vortex instruction, the stretch and swirl steps, the 140-149 °F / 60-65 °C band with the 'hot to touch' cue, turning steam off before removing the tip, tapping the jug to collapse bubbles, swirling to polish, working quickly before the milk separates, and the five-minute continuous-steaming cutout: Breville's own instruction book for the Barista Express (BES870), read 2026-09-09 · Milk volumes of 3.5 oz for a cappuccino and 6 oz for a latte, the fresh-cold-milk guidance, the caution that some milks not requiring refrigeration will not produce an acceptable foam, and pouring immediately so milk and foam do not separate: De'Longhi's own EC685 Get Started guide, read 2026-09-09 · Jug and thermometer specifications: the makers' own copy on the verified Amazon listings B07JMNYK7X, B07RP52VCC and B003M5SPNO via the Creators API, read 2026-09-09 · Plant-milk texturing behaviour is described qualitatively and is not attributed to a manufacturer: we have not run a controlled comparison, and the page says which adjustments help rather than claiming measured results We have not lab-tested brewers; where a figure is a manufacturer's, the page names the manufacturer, and machine capabilities come from our own published review records.

Keep going