Recipes

Each milk-based sheet uses 10 L of ordinary milk. Where a recipe uses different animal milks, their volumes total 10 L. Lancashire uses two separate 5 L makes. Aged Şürk starts with 1 kg of prepared milk curd; its source gives no reliable conversion to a milk volume.

Scale factor F = target base volume ÷ source base volume. Each proportional ingredient becomes source amount × F. A teaspoon is taken as 5 mL, a tablespoon as 15 mL, a US gallon as 3.7854 L and an explicitly Imperial gallon as 4.5461 L. Powder volume, mass, packet size and culture units are not interchangeable.

Standard liquid rennet is assumed to be 200 IMCU/mL. New dose (mL) = original dose × original strength × F ÷ new strength. For a stated activity rate: dose (mL) = IMCU/L × stated dosing volume (L) ÷ bottle strength (IMCU/mL). Unknown historical or powder strengths cannot be converted exactly; any replacement starting dose is marked as an adaptation and must be tuned to the specified flocculation or clean-break endpoint. Crescenza is the specific 150 IMCU/mL conversion described on its sheet.

Milk-based culture and additive rates are calculated on the actual 10 L milk batch. DVI cultures should follow their labelled milk volume and activity; fractional source packets do not establish the potency of a different product. Bulk culture percentages refer to active prepared starter, not dry powder. All recipe salt percentages use actual drained curd weight.

Dry salt (g) = actual drained curd weight (kg) × salt percentage × 10. Brine percentages on these sheets mean salt mass ÷ total solution mass unless specified otherwise: 20% w/w = 200 g salt + 800 g water. Grams per litre of water use a different denominator; 300 g/L water is about 23.1% w/w. Salometer degrees are a brine-density reading, not NaCl percent. Brine time depends on wheel thickness, mass, moisture and temperature; do not scale hours only by milk volume.

Press loads depend on mould area and curd condition. Where given, retain source pressure or wheel-weight multiples. For a round mould, area = π × diameter² ÷ 4. Load (kg-force) = pressure (psi) × area (in²) × 0.453592. Fixed source weights are identified with their original mould/batch context.

The standard curve form is the modified Gompertz acidification model: pH(τ) = pH₀ − A × exp{−exp[(μe/A)(λ − τ) + 1]}, where A = pH₀ − pH∞, e = 2.71828, μ is maximum pH decline per reference-temperature hour, λ is lag time and τ is accumulated equivalent fermentation time. Its use in yoghurt research supports the curve form; it does not validate these cheese-specific assumptions.

For the warm make, τ = Σ Δt × f(T). The illustrative temperature factor is f(T) = 2^((T − Tref)/10) up to Tref + 5 °C (Q10 = 2). Above that, the rate is capped and tapered linearly to zero at 45 °C for mesophiles, 50 °C for mixed cultures and 55 °C for thermophiles. Below 10 °C it is set to zero. These are modelling assumptions, not universal culture cardinal temperatures. A hot bath is not a curd-core temperature. The model is stopped at hot stretching and is not restarted as though starter survival were known.

Default illustrative values are pH₀ 6.65; μ = 0.30 for mesophilic, 0.40 for thermophilic, 0.32 for mixed and 0.15 for slow lactic cultures; lag = 0.65 h (1.0 h for lactic). Recipe reference temperatures and pH floors appear below. The floor is an assumed make-stage plateau, usually 0.10 below the style’s stated terminal pH; it is not the biochemical acid limit of the culture.

Culture identity determines the provisional family and temperature response. The displayed dosage is scaled to retain the source concentration. Bulk percentage, live yoghurt grams and DVI powder mass do not determine equivalent viable cell counts. Consequently no invented universal dose-to-rate factor is applied. Where available, curd source targets at their scheduled times fit μ and λ by least squares; they are design anchors, not new observations. A fit using source targets is an interpolation aid, not independent validation. Milk/whey targets are never converted into curd targets by a fixed offset.

Where the source gives a time range, its midpoint is used for prediction. Where it omits a short handling time or room temperature, the sheet identifies the estimate. Rennet time still follows set strength. Source pH targets are labelled and retained; recorded pH is labelled as historical reference; ≈ marks newly inferred values, rounded to one decimal to avoid false precision. A source range may be represented by its midpoint in the pH column, with the operative range retained in the action or notes.

Washing changes buffering, lactose and serum composition; wine or citric acid changes acidity directly; mixed-age curds cannot be assigned an arithmetic average pH. A single culture curve cannot model those changes reliably. Relevant sheets retain measured source checks or explicitly require measurement. No culture-kinetic pH is fabricated for direct-acid Scamorza or supplied Şürk curd.

After brining, pressing and cooling, predictions are provisional make-end baselines. They do not forecast weeks of rind development, surface deacidification, proteolysis or core pH. Temperature cooling, salt inhibition, milk buffering, strain composition and starter freshness can shift real values by several tenths of a pH unit or more. No confidence interval is claimed. Calibrate an appropriate meter, identify milk/curd/whey correctly and measure consistently. Titratable acidity and pH describe different properties; historical TA values do not have a universal conversion to pH.

Pasteurisation is marked “Skip for this cheese” throughout this library because already-pasteurised dairy is specified; this does not make raw milk equivalent to pasteurised milk. Heat steps that are part of the cheese process, such as cooking or stretching, remain active. Follow hygiene and temperature controls appropriate to the cheese. Fresh-cheese storage uses 4 °C, within the Food Standards Agency recommendation of a refrigerator at 5 °C or below. A pH prediction, salt addition or cold smoke does not establish a safe shelf life.