Making Fresh & Acid-Set Cheese
This is the family you can make this afternoon, with nothing you’d have to order. It’s also where I started, thirty years ago, and where I still spend a lot of time.
The family page says what these cheeses are. This one is how to make them — and it opens with the thing that took me twenty years to see, which is that “acid-set” covers two different makes that produce two different materials.
The one decision: hot and fast, or cold and slow
Both routes are chasing the same number. Casein’s isoelectric point is around pH 4.6 — the pH at which it loses its charge, stops repelling itself, and falls out of solution. Every cheese on this page is a way of getting the milk to 4.6.
The hot route gets there in ten minutes with a bottle of acid. The cultured route takes twelve to eighteen hours and lets bacteria do it. And the results are not interchangeable:
| Hot acid | Cultured | |
|---|---|---|
| Temperature at set | 185–195 °F / 85–90 °C | 70–75 °F / 21–24 °C |
| Time | 10–20 minutes | 12–18 hours |
| Texture | Coarse, springy, squeaky | Fine, smooth, spreadable |
| Yield | Higher | Lower |
| Flavour | Of the milk and the acid | Of the culture — tangy, complex |
| Examples | Ricotta, paneer, queso fresco, quick mozzarella | Chèvre, quark, cream cheese, fromage blanc |
⭐ The yield difference has a real cause and it’s worth knowing. At 185 °F and above you are hot enough to denature the whey proteins, and denatured whey protein co-precipitates with the casein instead of running off down the drain. So the hot route captures protein the cold route leaves in the whey. That is also why the hot curds are coarse: you’re catching a second material with different properties, not just more of the first.
🔴 The one place the no-UHT rule does not apply
Everywhere else in this section I’ll tell you that ultra-pasteurised milk will not make cheese. On the hot acid route, that rule is off.
UHT milk fails a rennet set because the heat treatment has already denatured the whey proteins and tangled them onto the casein micelles, so chymosin can’t build a proper gel. But an acid set isn’t building a rennet gel — it’s crashing the casein out by charge. The denatured whey protein that ruins a rennet cheese is, here, free yield.
So: ricotta, paneer and queso fresco will work with the milk in the shop cooler. Cultured chèvre and cream cheese, which take a whisper of rennet to firm the set, will not. Sort the recipe by which route it is before you decide the milk matters.
The hot-acid make, step by step
1. Heat the milk, slowly, stirring. A gallon, in a heavy pot, over medium-low. You are going to 185–195 °F / 85–90 °C. Stir the bottom the whole way — milk scorches at the pot floor before it does anything else, and scorch is a flavour you can’t remove.
Do not boil it. A hard boil makes the curd tough and drives off the delicate flavours that are most of the point.
2. Take it off the heat before the acid goes in. Then add the acid gradually, stirring gently, and stop when the whey turns from milky to translucent yellow-green.
- Vinegar (5% white or cider): start with ¼ cup per gallon
- Lemon juice: about ⅓ cup per gallon
- Citric acid: about 1 teaspoon per gallon, dissolved in ¼ cup cool water first
⚠ Add less than the recipe says and check. Excess acid does not give you more cheese — it gives you a tougher, squeakier curd that tastes of vinegar. The whey colour is a better instrument than the measuring spoon.
3. Stop stirring, and wait 10–20 minutes. The curds knit and rise. Stirring now shatters them into fines that go straight through the cloth.
4. Ladle — don’t pour. Lift the curds out gently into a cloth-lined colander. Pouring the whole pot drops the curd through a column of hot whey and packs it.
5. Drain to the texture you want, which is the only real dial left. Twenty minutes for a moist ricotta; two hours hanging for something you can slice; overnight under a weight for paneer.
6. Salt at the end, not before — salt in the pot interferes with the set. 1–2% of the drained curd weight, or to taste, which for fresh cheese is the honest instruction.
The cultured make, step by step
1. Warm a gallon of milk to 72 °F / 22 °C. That’s it — this make happens at room temperature, and in most kitchens most of the year you can hit it by leaving the pot out.
2. Sprinkle a mesophilic culture on the surface, let it rehydrate a minute, stir it through. A DVI packet is measured in fractions of a teaspoon; a mother culture is measured in ounces. They are not the same dose.
3. Add a whisper of rennet. One or two drops of single-strength liquid rennet per gallon, diluted in ¼ cup of cool unchlorinated water. This is not a rennet cheese — the dose is there to firm the set enough that it drains cleanly rather than turning to soup in the bag.
4. Cover it and leave it alone for 12–18 hours, at room temperature. You are waiting for a set that breaks cleanly and a layer of whey on top.
5. Drain in a bag, 6–12 hours, at room temperature or in the fridge. Room temperature drains faster and acidifies further; the fridge stops the clock. That choice is most of the difference between a mild fromage blanc and a sharp one.
6. Salt, then beat or fold. Beating a drained cultured curd smooths it; folding keeps it in grains.
The members, and what changes
Ricotta (from whole milk). Hot route. Highest heat, gentlest handling, shortest drain. Note that the real thing is made from whey — that’s its own page and a different make entirely.
Cottage cheese. Hot route with vinegar or lemon, cut into curds, rinsed, then dressed with cream or milk. The rinse is what stops it tasting of the acid.
Paneer. Hot route, then pressed under weight for an hour or more. Pressing is the whole difference — it’s what makes it hold together in a curry instead of dissolving. It doesn’t melt for the reason on the melting page: high pH, calcium still in the network.
Queso fresco. Hot route, salted more heavily, pressed lightly. Crumbles rather than melts.
Quick mozzarella. Hot route, but the curd is then stretched — which puts it one step from pasta filata. The cultured version of the same cheese is a genuinely different animal and lives on that page.
Cream cheese. Cultured route, on a milk-and-cream blend. Drained hard.
Mascarpone. Cream only, heated to 185 °F, set with tartaric or citric acid, drained. No culture, no rennet — the outlier that proves the hot route is about heat and charge, not bacteria.
Quark, fromage blanc, chèvre, farmer’s cheese. All the cultured route, differing mainly in milk (chèvre is goat), drain time and what you do to the curd afterwards.
Labneh. ⚠ Strained yogurt, and I’d call it a cheese with an asterisk — the coagulation is done by a yogurt culture at yogurt temperatures, and nothing is added to make cheese of it. Delicious, and sitting slightly outside the definition this section uses.
What “right” looks like
The whey should be translucent, yellow-green, and clear enough to read through. Cloudy white whey means fat and fines are leaving with it, which is yield walking out of the pot and is usually a sign you stirred after the acid or boiled it.
The curd should smell sweet, of milk. Cultured curd should smell faintly like good yogurt.
It should taste of milk first and acid second. If the acid is the first thing you notice, you used too much.
What goes wrong
No curd at all, hot route. Either not hot enough (get to 185 °F — this is the one number that isn’t negotiable) or the milk was already too acidic to take more.
A weak, sloppy set, cultured route. Room too cold, culture too old, or chlorinated water in the rennet dilution. See cultures and rennet — the enemy of a culture packet is moisture, not warmth.
Tough, rubbery, squeaky curd. Too much acid, or a hard boil, or both.
It tastes of nothing. Under-salted, almost always. Fresh cheese has had no ripening to build flavour, so salt is doing more of the work here than in any other family.
It went sour in three days. These are fresh cheeses with high moisture and no rind. A week in the fridge is a reasonable expectation; two is optimistic. That’s not a defect, it’s the family.
Where I came in
My first cheese was the no-rennet mozzarella on this page — raw cow’s milk from the farm down the road, vinegar, and flake salt. It was amazing, and it asked almost nothing of me, which I found faintly suspicious at the time.
Cottage cheese next, with vinegar and then with lemon juice. Both amazing, both so easy, and genuinely two different cheeses, which is why they’re two recipes rather than one with a note.
Then I went back to the mozzarella and used smoked flake salt instead of plain, and that turned out to be the moment the hobby got its hooks in — the first time I changed one variable on purpose to see what would happen. The whole story is here.
⭐ And the correction I owe this page: I filed the vinegar version for years as the beginner’s shortcut, the thing you do before you get serious and buy rennet. It isn’t. Acid and rennet are two different chemistries, and on the archaeological evidence the acid branch is plausibly the older one — the oldest dated cheesemaking artefact is a strainer, and a strainer is exactly what this make needs.
The formulas
This page’s recipes are written. The hot-acid route:
- Mozzarella Cheese (Traditional & Quick No-Rennet) — my first one
- Cottage Cheese (Vinegar) and Cottage Cheese (Lemon Juice) — genuinely two different cheeses, which is why they are two recipes
- How to Make Your Own Paneer (Indian Cottage Cheese)
- Homemade Ricotta — the whole-milk version; the whey version is the real one and belongs with the second-run cheeses
- Cheese, Queso Fresco — and the note on it explains why it browns instead of flowing
The cultured route:
- Cheese, Cream (Cultured) — one I actually make, and an honest account of why the homemade one weeps and the block does not
- Cheese, Mascarpone — cream only, no culture, no rennet
- Cheese, Quark · Cheese, Fromage Blanc · Cheese, Farmer’s — ⭐ one make, three vocabularies and a different drain time, which is this section’s whole argument in miniature
- Cheese, Chèvre (Fresh Goat) — where the milk changes the result more than the method does
Labneh is already in the collection as Labneh, so I have linked it rather than shipping a second one. It sits slightly outside this section’s definition anyway, for the reason above.
⚠ Read the honesty flags on these. Cream cheese is one I make. Mascarpone, quark, fromage blanc, chèvre, queso fresco and farmer’s cheese are all marked UNTESTED, because thirty years of “soft cheeses and others” is not a licence to claim a specific one. They are briefings, and each says so at the top.
Two auxiliaries that every make in this wing needs, and that now exist: Cheese Culture, Mesophilic Mother and Cheese Brine, Saturated.
Next: pasta filata — the same curd, stretched, and a pH window you can miss in twenty minutes.
Comments (0)