Making Whey & Second-Run Cheese
Real ricotta, anari, mizithra, mysost, brunost. The second harvest — a cheese made from the liquid you were about to pour away.
The family page covers the category. This page is the make, and it has an unusual property: whether it works at all was decided by the cheese you made first.
The one thing: the whey has to still have something in it
Ricotta means recooked, and that is a literal description. When rennet coagulates milk, it takes the casein and leaves the whey proteins — lactalbumin and lactoglobulin — dissolved in the liquid. Those proteins don’t respond to rennet or to mild acid. But heat them hard enough and they denature and precipitate, and you can skim them off.
So a second-run cheese is a second protein, not a second helping of the first one. And it depends entirely on two things being left in the whey:
Enough protein. Which is a question of what you did.
A high enough pH. ⭐ This is the one that decides it. Whey drained at pH 6.0 or above — sweet whey — will give you ricotta. Whey from a cheese that acidified hard before draining is acid whey, and it will give you almost nothing, because the proteins have already been partly precipitated and dragged off with the curd.
| Whey from | pH at draining | Second run |
|---|---|---|
| Halloumi (little or no culture) | ~6.5 | Excellent — this is anari |
| Alpine, gouda, mozzarella | 6.1–6.4 | Good |
| Cheddar drained late | 5.8–6.0 | Poor |
| Any acid-set cheese | below 5.0 | Essentially none |
Which produces a nice piece of kitchen logic: the drier and sweeter the first cheese, the better the second one. Make halloumi and you get anari for free. Make cottage cheese and there is no second cheese in that pot, and no technique will conjure one.
The make, step by step
1. Use it the same day. Sweet whey acidifies on its own within hours at room temperature, so this is a make you start when the first one finishes, not tomorrow.
2. Strain out the fines through a fine cloth. Curd dust in the pot scorches and makes the ricotta grainy.
3. ⭐ Add whole milk — 5 to 10% of the whey volume. This is standard practice, including commercially, and it isn’t cheating: the added casein gives the denatured whey protein something to nucleate on and roughly doubles a very small yield. A gallon of whey takes a cup or two of milk.
4. Heat, slowly, stirring, to 175–185 °F / 80–85 °C. Then stop stirring and take it the rest of the way to 195–200 °F / 90–93 °C.
5. Add acid if it needs it. Some sweet whey flocculates on heat alone. Most wants a little help: a tablespoon or two of vinegar or lemon juice per gallon, added at 185 °F and stirred in once, very gently.
6. Let it rise, and leave it alone. A raft of fine white curd will form and float. Give it 10–20 minutes off the heat. Stirring now breaks the raft into particles that go straight through any cloth you own.
7. Skim, don’t pour. Lift the raft off with a fine skimmer into a cloth-lined colander. ⚠ Butter muslin, not cheesecloth — the curd here is far finer than anything else in this section and ordinary cheesecloth will lose most of it.
8. Drain 20 minutes to a few hours, depending on whether you want it spoonable or sliceable. Salt lightly at the end, if at all.
What to expect, honestly
A gallon of good sweet whey gives a few ounces. Three to six, typically, and more like eight with the added milk. That is not a failed batch — that is the yield. The whey proteins are a small fraction of milk protein to begin with, and you are recovering some of them.
I’d rather state that plainly than have somebody conclude they did it wrong. A second-run cheese is a bonus, not a second batch.
The members
True whey ricotta. The make above. Note that most ricotta sold, and most home ricotta recipes including the one in this collection, is made from whole milk with acid — which is a fine cheese and belongs with the fresh ones. It is not this.
Anari. Cypriot, made from halloumi whey, and the best-case version of this page — eaten fresh, or salted and dried hard enough to grate.
Mizithra. Greek, from whey plus milk. Sold fresh, or dry-salted and aged into a hard grating cheese that is a completely different product under the same name.
Ricotta salata. Whey ricotta pressed, salted and aged for a few months. The obvious thing to do with a batch you can’t eat fast enough, and worth planning for.
🔴 Brunost / gjetost / mysost — and these are not cheese.
There is no coagulation at all. The whey is boiled down with cream and milk until the water is gone and it goes brown, and the result is poured into a mould to set. Brown, sweet, fudge-like, and made by a process with more in common with boiling sap into syrup than with anything else on this site.
🔴 A correction to what this page used to say. It said the lactose caramelises. That is the common description and it isn’t quite right: the browning of heated milk products is predominantly a Maillard reaction — reducing sugars reacting with amino groups from the milk proteins to form the brown polymers called melanoidins. True caramelisation of lactose happens too, and it is the smaller contributor.
⭐ That matters at the pan rather than only on paper, because Maillard needs the protein as well as the sugar. It is why a whey with more protein left in it browns deeper, why adding cream or milk intensifies the colour instead of diluting it, and why you cannot imitate this by boiling down a sugar syrup. The protein isn’t a passenger — it’s half the reaction. The full version is in the mysost recipe.
It’s delicious and it’s an ingenious use of a waste stream. It just isn’t a cheese by any definition this section uses, and the Scandinavia page has the fuller argument — including Anne Hov, who added the cream at Gålå in 1863 and turned a peasant staple into an export.
Making it at home: a gallon of whey reduces to a few ounces of brunost over three to five hours, and the last twenty minutes is where it burns. Stir constantly at the end, and take it off when it’s the colour of caramel rather than the colour of coffee.
What goes wrong
Nothing rises. Whey was too acid, or not hot enough. 195 °F is the number and a thermometer is worth more here than judgement.
It came out grainy. Boiled too hard, or stirred after the acid.
Most of it went through the cloth. Cheesecloth instead of butter muslin.
Scorched. Whey has less fat than milk to protect the pot floor. Heavy pan, moderate heat, and stir until you stop stirring on purpose.
And if there’s no second cheese in it
Most home wheys won’t give you much, and it’s worth knowing what else the liquid is good for rather than treating it as a failure: it’s excellent in bread in place of water, it makes a good lacto-ferment brine, it’s a legitimate soup base, and it goes to livestock in every dairying tradition on earth. What it should not go into is a septic system in quantity.
The formulas
Traditional Whey Ricotta is in the collection and it is the make described on this page.
⚠ I list ricotta among the cheeses I’ve made over thirty years, but I genuinely can’t tell you which of the two I made and when — whole-milk or whey — so I’m not going to claim the harder one. The recipe is there and it’s right; the provenance of my own memory isn’t.
And the rest of the family is written too:
- Cheese, Anari — from halloumi whey, and the best case on this page. Not a make you plan; a make you’re handed.
- Cheese, Mizithra — the same cheese under another name, and one word covering at least three genuinely different products.
- Cheese, Ricotta Salata — the answer to a small awkward yield, and the most practical recipe here.
- Cheese, Mysost — the lean original, and no coagulation anywhere in it.
- Cheese, Brunost (Gjetost) — mysost plus one idea, with a date, a place and a name on it.
The two auxiliaries: Cheese Culture, Mesophilic Mother and Cheese Brine, Saturated — the brine one being the more important, because a plain salt brine takes a good cheese apart in about a week.
Next: smoked and flavoured — cold smoke, dried herbs, and the easiest trick in the section.
Comments (0)