Modern Science · Regional Recipes · From the Sea

Two Eggs, Two Temperatures

A firm, free-form poached white needs about 80 °C (176 °F). A custard yolk locks at about 63 °C (145 °F). Those two numbers are seventeen degrees apart, and no single bath sits in two places at once — which is the whole secret of poaching at volume. The question was never “how do I poach a hundred eggs faster?” It is “which texture am I serving, and which line lets me cook it ahead?” Get that right and one cook plates a brunch service single-handed, on a moving boat, without a single egg going to order in the rush.

What you'll get
  • The one fact that runs everything: the white and the yolk set in different temperature bands.
  • Why a sous-vide “poached” egg slides off the toast — and when that's exactly what you want.
  • The make-ahead line for a classic firm white: poach short → ice-shock → hold → reheat to order.
  • The strainer trick that does the vortex's job and actually scales.
  • How to fire a whole perforated tray to order and keep the plate dry.
  • The one safety number people get dangerously wrong — and the folklore to drop.

One bath cannot give a firm white and a runny yolk, so cook ahead. For a classic poached egg: strain off the loose white, par-poach 30 s short at a smiling 82 °C (180 °F) — never a boil — ice-shock, hold submerged at ≤5 °C (41 °F) for up to three days, then fire to order in a 66 °C (150 °F) perforated tray. For a custard yolk in a cup or bowl: hold the eggs shell-on at 63 °C (145 °F) for 45–60 minutes — un-overcookable, near-zero labour. The protein doesn't negotiate; your service plan has to bend.

80°CFirm white (ovalbumin)
63°CCustard yolk locks
70°CYolk sets firm
82°CPoach water — not boiling
66°CReheat-to-order tray
3 daysCold-hold cap

The problem: one bath can't do both

Egg white is not one protein but a crowd of them, and they set at different temperatures. The earliest to go is ovotransferrin, which firms in the low sixties. But more than half the white — about 54% — is ovalbumin, and ovalbumin does not functionally firm until around 80 °C (176 °F). The yolk, by contrast, thickens to a custard by about 65 °C (149 °F) and only sets firm near 70 °C (158 °F). Line those up and you get an inversion that decides everything:

The yolk sets before the bulk of the white. So at a steady 63 °C (145 °F) onsen bath, the yolk goes jammy and custardy while roughly half the white — the ovalbumin — never coagulates at all. That is why a sous-vide “poached” egg is a glorious thing in a bowl and a puddle on toast. The protein that would hold a free-form shape was never taken hot enough. (Note the two numbers people blur: ovalbumin's lab peak is ~84 °C / 183 °F; its functional firm-set is ~80 °C / 176 °F. Either way, far above the onsen bath.)

The Three Temperatures

Custard yolk locks~63–65 °C (145–149 °F)
Yolk sets firm~70 °C (158 °F)
Firm free-form white~80 °C (176 °F) — ovalbumin, 54%
The gapWhite firms ~15 °C above the custard yolk
The consequenceNo steady bath gives both at once
The architectureDecouple the cook from the plate

You are not poaching faster. You are deciding which of two textures the dish wants — then cooking it ahead.

Pick your texture first

Two methods, and the texture chooses

Because the proteins force the split, every workable batch method is one of two lines — plus a hybrid and the live poach you keep only for tiny numbers. Decide the texture, and the method is decided for you.

MethodWhat you getParametersUse when
A · Make-ahead poachFirm ruffled free-form white, liquid yolkPoach 82 °C 2–2.5 min → ice-shock → hold ≤3 days at ≤5 °C → reheat 66 °C 3 minBenedict, eggs on toast — the classic poached egg at volume
B · Sous-vide onsenGlossy custard yolk, loose milky white63 °C (145 °F), 45–60 min, in shellCup, ramen, rice or grain bowl — lowest labour, foolproof
HybridCustard yolk + poached-look whiteHold 63 °C, then crack, drain, finish 30–60 s in 82 °CA luxe plate to order — more kit, a delicate pickup
Live to orderFresh single poachStrain, 82 °C, 3–4 min eachOnly past ~6–8 eggs total, or true à la carte
Method A · the make-ahead line

The make-ahead poach: a firm white, fired to order

This is the only route that gives a true poached-white texture and à-la-minute speed across twenty to a hundred covers. It runs in five moves.

1 · Strain every egg. Crack each into a fine-mesh strainer or perforated cul-de-poule and swirl a few seconds to drain the loose, watery white (you lose ~15–25% of the volume). 2 · Par-poach short. Slide batches of 6–8 into a bare, “smiling” simmer at 82 °C (180 °F) — small bubbles only — for ~2–2.5 min, about 30 s under done. 3 · Ice-shock at once to stop the cook and firm them for handling; trim any wisps. 4 · Hold cold — fully submerged in fresh cold water, ≤5 °C (41 °F), for up to three days. 5 · Fire to order in a perforated GN tray lowered into 66 °C (150 °F) water for 3 min — it holds there ~20 min without overcooking, and you lift the whole tray in one motion to drain.

Strain, don't vortex. The single-egg swirl was always a trick to keep the loose white from feathering — and it cannot scale; you cannot vortex a pot of eight. Straining off the thin white does the same job and works for any number. It also means even a not-quite-fresh egg poaches clean, and it makes vinegar optional. Freshness still helps most — a fresh white is held compact by its ovomucin and pours less — but the strainer is the lever that scales.

Method B · the onsen hold

The onsen: a custard yolk, and almost no labour

When the dish wants a soft set — an egg dropped into dashi, over a rice or grain bowl, into ramen — hold the eggs shell-on at 63 °C (145 °F) for 45–60 minutes. The result is a glossy custard yolk and a deliberately loose, milky white. The beauty for service is that the egg cannot exceed the bath temperature, so it is impossible to overcook: forty minutes and fifty minutes are near-identical, and a held rack just waits for you. Capacity is limited only by the size of your bath. For a plate that should look poached, crack the egg out, drain the loose white, and finish it 30–60 s in 82 °C (180 °F) water to set a thin firm skin over the soft interior.

Two facts make the onsen forgiving. First, yolk texture is a function of time and temperature together, not one magic degree — it follows gelation kinetics, building structure slowly, so the “63.5 °C exactly or it's ruined” line is a myth. Second, a 63 °C hold for 45–60 min comfortably exceeds the food-safety target for eggs and is effectively self-pasteurising — the long gentle hold does double duty as a kill step.

For service

Plating the rush, single-handed

With either line, the egg is no longer the bottleneck — assembly is. That is the entire point of cooking ahead. A few moves keep the pass clean.

Drain hard before it touches the plate. This is the number-one Benedict sin: a film of poaching or reheat water rides onto the plate and pools under the muffin. Rest each egg on a clean towel, or lift the whole batch in the perforated tray and let it drip; blot if you need to. Hold safe. Cold-hold at ≤5 °C (41 °F); hot-hold at ≥57 °C (135 °F) — both the 66 °C reheat and the 63 °C onsen clear that floor with margin — and keep total time in the 5–57 °C danger zone under four hours. As for throughput: with the eggs pre-cooked, a single cook can plate at the rate they can build the dish, not the rate they can poach. Treat any “plates per minute” figure as a planning heuristic for your own line, not a number to quote.

Safety & folklore

The one number to get right

A runny yolk is a Salmonella route, so use clean, well-refrigerated fresh eggs and serve hot; warn vulnerable guests. If you want a guaranteed-safe runny egg, pasteurise it in the shell — and this is where a dangerous error circulates, so be precise.

In-shell pasteurisation is ~57 °C (135 °F) for ~75 minutes. That holds the egg long enough for heat to penetrate the shell and reach the yolk while keeping the white liquid and usable. Do not use the “60 °C (140 °F) for 3.5 minutes” figure for whole shell eggs — that is the standard for liquid egg, and 3.5 minutes cannot pasteurise through an intact shell to the centre. The two get conflated constantly; the shell egg needs the long, low hold.

Folklore to drop. “Drop them into boiling water” — a rolling boil (100 °C / 212 °F) shreds the white; stay at a smiling 82 °C (180 °F). “Vinegar is essential” — the mechanism is real (acid speeds the surface set) but straining is the better lever, and too much vinegar sours and films the white; a splash (15–30 mL / litre) is the most it earns. “The vortex” — a single-egg trick that does not scale. “Older eggs poach better” — false; that is hard-boil peeling. Poaching wants the freshest eggs, because a fresh, lower-pH white holds together and an aged one feathers into wisps.

The takeaway

Stop trying to make one bath do two jobs. Decide which texture the dish wants — a firm free-form white or a soft custard yolk — and let that choose the line: make-ahead poach-shock-hold-reheat for the classic poached egg, sous-vide onsen for the cup-and-bowl custard. Strain the thin white, keep the water at a smiling 82 °C and never a boil, and fire a whole perforated tray to order from 66 °C water. Do that, and you can put up a hundred covers single-handed on a moving boat. The protein doesn't negotiate; your service plan has to do the bending.


Tested at sea.

The receipts
The white sets in two stages Donovan JW, Mapes CJ, Davis JG, Garibaldi JA. “A differential scanning calorimetric study of the stability of egg white to heat denaturation.” Journal of the Science of Food and Agriculture, 1975. doi:10.1002/jsfa.2740260109 Grade A · ovotransferrin 65 °C, ovalbumin peak 84 °C
The yolk sets lower than the bulk white Le Denmat M, Anton M, Gandemer G. “Heat denaturation of egg yolk plasma and granules.” Journal of Food Science, 1999. doi:10.1111/j.1365-2621.1999.tb15863.x Grade A · plasma onset ~69 °C, granules ~76 °C
Yolk texture is time and temperature Vega C, Mercadé-Prieto R. “Culinary biophysics: on the nature of the 6X °C egg.” Food Biophysics, 2011. doi:10.1007/s11483-010-9200-1 Grade A · refutes the “exact degree” myth
In-shell pasteurisation — the corrected figure Baldwin DE. “Sous vide cooking: a review” / A Practical Guide to Sous Vide; with USDA FSIS Annex G shell-egg immersion data. Int. J. Gastronomy & Food Science / FSIS Grade B · ~57 °C/135 °F, ~75 min (NOT the liquid-egg 60 °C/3.5 min)
Cook-and-hold safety US FDA Food Code (3-401.11) & USDA FSIS “Danger Zone.” FDA / FSIS Grade A/B · 63 °C/145 °F 15 s to order; hot-hold ≥57 °C

Also drawn on: McGee, On Food and Cooking 2004 (practical set onsets — white thickens ~63 °C, firm ~80 °C; yolk thickens ~65 °C, sets ~70 °C; the iron-sulfide grey ring on long hot holds); Modernist Cuisine 2011 and Baldwin on equilibrium in-shell cooking (the egg can't exceed bath temperature — why the onsen is un-overcookable); the ovomucin / Haugh freshness literature (fresh albumen pH ~7.6 climbs to ~9.5 as carbon dioxide escapes, thinning the white and causing feathering); J. Kenji López-Alt (Serious Eats) and America's Test Kitchen for the strain-and-make- ahead line and the 66 °C/150 °F reheat-hold. Honesty: the protein denaturation temperatures and yolk kinetics are Grade A; McGee's set onsets and the FDA/FSIS numbers are Grade B; the make-ahead poach line and the 66 °C reheat are Grade C practitioner standard — reliable kitchen practice, not lab-validated. The “holds five days” claim has no food-safety study behind it, so the cap here is three; any single-cook “plates per minute” figure is a planning heuristic, not a citable number. No funding; no conflicts.

Have a technique or a question? Share it.

Join the conversation