foratvise
Get Started
Mixing and formulation

Homogenization

High-pressure valve homogenization of pasteurized milk

Validated modelOEE bound 0.982 [0.975–0.986]20,000 Monte Carlo samples · seed 20261115 · computed November 16, 2026

In short

Homogenization can use at most 98 of every 100 planned hours. The rest goes to limits no plant can avoid.

  1. 98usable at most
  2. 2are a physical limit

The three limits that set it

Running time: up to 98 %

The rest are stops imposed by physics or by regulation: cleaning, changeovers, defrosting.

Speed: up to 100 %

How close to its maximum speed it can run without damaging the product or the equipment.

Good product: up to 100 %

Some product falls outside specification even with the process well under control.

See the technical detail

Technical detail

Uncertainty intervals, physical models per station and sources, for anyone who wants to check each figure.

1Scenario

Standardized whole milk is pumped by a piston pump through a two-stage homogenizing valve in line with the pasteurizer. The fat globules are broken from about 3.5 µm to below 1 µm so that the packaged milk, stored cold, passes the creaming test (48 h quiescent storage, the fat of the top 100 mL of 1 000 mL may not exceed the fat of the rest by more than 1/0.9). The homogenizer is cleaned in place with the pasteurizer once per day of use. Homogenization of other emulsions (cream, ice-cream mix, juices with pulp) follows the same structure but is outside the numbers of this model.

Provenance of inputsComputed2Declared assumption8Verified quote32

2Results

Point value from the nominal parameters and 95% interval from 20,000 Monte Carlo draws of the uncertain parameters.
  • Serum viscosity at the storage temperature2.82 mPa·s
  • Log standard deviation of the homogenized globule-size distribution0.559 [0.520–0.601]
  • Largest volume-weighted mean rise distance that passes the creaming test1.98 [1.51–2.45] mm
  • Largest volume-weighted median globule diameter that passes the creaming test0.636 [0.519–0.778] um
  • Surface-weighted (Sauter) mean diameter at the specification0.544 [0.441–0.669] um
  • Increase of the fat/serum interfacial area from native to specification globules6.59 [5.36–8.13]
  • Minimum homogenizing pressure of the reference two-stage valve that passes the creaming test20.4 [15.9–26.5] MPa
  • Pressure set point of the capable process (minimum plus z standard deviations)21.9 [17.5–28.4] MPa
  • Tension of the casein-covered milk-fat/serum interface10.3 [8.1–12.6] mN/m
  • TheoreticalThermodynamic minimum: free energy of the new interface0.00364 [0.00248–0.00518] kJ/kg
  • Hydraulic work at the minimum pressure of the reference valve (p/rho)19.9 [15.5–25.7] kJ/kg
  • Electricity at the set-point pressure of the reference valve (drive efficiency of the documented conventional machine)23.2 [18.5–30.0] kJ/kg
  • Best available techniqueElectricity of the advanced-design homogenizer (97 bar, BAT reference document)10.5 kJ/kg
  • Best documentedElectricity of the efficient-design homogenizer (135 bar, BAT reference document)14.6 kJ/kg
  • TypicalElectricity of the conventional homogenizer (200 bar, BAT reference document)21.2 kJ/kg
  • Hydraulic power over electric power of the documented conventional homogenizer0.920
  • Thermodynamic minimum divided by the typical electricity1.72 × 10⁻⁴ [1.17 × 10⁻⁴–2.45 × 10⁻⁴]
  • Temperature rise of the milk per MPa of pressure drop (all hydraulic work dissipated)0.248 [0.244–0.254] K/MPa
  • Temperature rise at the set-point pressure5.45 [4.35–7.07] K
  • Bernoulli velocity in the valve gap at the set-point pressure207 [185–235] m/s
  • Milk homogenized per kW of drive at the set-point pressure of the reference valve151 [117–189] L/h/kW
  • TheoreticalMaximum yield (no stream split)1.00
  • Share of milk homogenized below the minimum pressure (fails the creaming test)0.00135 [0.00025–0.00576]
  • TheoreticalMaximum availability (daily cleaning and wear-part exchange)0.9834 [0.9778–0.9876]
  • TheoreticalMaximum performance factor (positive-displacement pump at its rated flow)1.00
  • TheoreticalMaximum OEE quality factor (creaming test)0.9987 [0.9942–0.9997]
  • TheoreticalPhysics-bounded OEE0.982 [0.975–0.986]

3Ceilings per dimension

Each dimension shows the ceilings the model computes and, where units are comparable, documented benchmarks. Colour and marker shape identify the ceiling type throughout the Oracle.
  • Theoretical
  • Best available technique
  • Best documented
  • Typical
Energy per unit of productkJ/kg · lower is better
  • TheoreticalThermodynamic minimum: free energy of the new interface0.00364 [0.00248–0.00518] kJ/kg
  • Best available techniqueElectricity of the advanced-design homogenizer (97 bar, BAT reference document)10.5 kJ/kg
  • Best documentedElectricity of the efficient-design homogenizer (135 bar, BAT reference document)14.6 kJ/kg
  • TypicalElectricity of the conventional homogenizer (200 bar, BAT reference document)21.2 kJ/kg
Yieldhigher is better
  • TheoreticalMaximum yield (no stream split)1.00
Availabilityhigher is better
  • TheoreticalMaximum availability (daily cleaning and wear-part exchange)0.9834 [0.9778–0.9876]
Performancehigher is better
  • TheoreticalMaximum performance factor (positive-displacement pump at its rated flow)1.00
Qualityhigher is better
  • TheoreticalMaximum OEE quality factor (creaming test)0.9987 [0.9942–0.9997]
OEEhigher is better
  • TheoreticalPhysics-bounded OEE0.982 [0.975–0.986]

4Loss cascade

What separates each ceiling from the next one, from the physical limit to typical operation.
  1. TheoreticalBest available techniqueEnergy per unit of product
    0.00364 [0.00248–0.00518] kJ/kg → 10.5 kJ/kg (gap 10.5 kJ/kg)

    The thermodynamic minimum is the free energy of the new fat/serum interface, gamma times the new area: a few joules per kilogram of milk. A valve homogenizer dissipates the whole hydraulic work p/rho in the turbulent jet behind the gap (one kelvin of heating per 4 MPa); only the eddies of the globule size break globules, so more than 99.9 % of the work ends as heat. The advanced valve design of the BAT reference document reaches the same homogenization effect at 97 bar by shaping the gap so that more of the turbulence acts at the globule scale.

  2. Best available techniqueBest documentedEnergy per unit of product
    10.5 kJ/kg → 14.6 kJ/kg (gap 4.06 kJ/kg)

    The efficient design needs 135 bar for the same effect; the difference to the advanced design is valve geometry, applicable only above about 15 000 L/h.

  3. Best documentedTypicalEnergy per unit of product
    14.6 kJ/kg → 21.2 kJ/kg (gap 6.58 kJ/kg)

    A conventional valve needs 200 bar. The creaming test in this model asks for a median globule diameter close to the one a conventional two-stage valve gives at 20 MPa, which is why the conventional pressure and the computed minimum pressure of the reference valve nearly coincide; partial homogenization (cream plus a small share of skim milk) cuts the pumped volume and the power by about two thirds.

5Physics-bounded OEE

The highest OEE the operation can reach: availability, performance and quality limited only by unavoidable stops and physical capability.
Fraction of loading time that remains after each unavoidable loss. Axis starts at 0.90.
  • Loading time (ideal)
    1.000
  • Availability
    factor 0.9834 [0.9778–0.9876]
    0.983 (−0.017)
  • Performance
    factor 1.00
    0.983 (−0.000)
  • Quality
    factor 0.9987 [0.9942–0.9997]
    0.982 (−0.001)
  • Physics-bounded OEE
    0.982 [0.975–0.986]

6Full derivation

Mass and energy balances, rate-limiting step, process capability and unavoidable stops, with every term explained.

Scope. Standardized whole milk is pumped by a piston pump through a two-stage homogenizing valve in line with the pasteurizer, at 55–80 °C. The fat globules are broken from about 3.5 µm to below 1 µm so that the packaged milk, stored at or below 7 °C, passes the creaming test: after 48 h of quiescent storage of 1 000 mL, 0.9 times the fat content of the top 100 mL must be below the fat content of the rest. Symbols and their values, units, uncertainty distributions and sources are listed in the parameter table generated from config/physics/homogenization.yaml.

1. Mass balance — maximum yield

Homogenization has no phase change, no reaction and no stream split: every kilogram that enters the pump leaves the valve. The steady-state yield is exactly 1. The product–water mixing zones at start-up and shut-down belong to the pasteurization line in which the homogenizer sits and are counted in that model, so

Ymax=1.Y_{max} = 1 .

2. The quality law — what the creaming test demands

Creaming. Homogenized globules rise individually (they no longer cluster by cold agglutination), at the Stokes velocity

v=Δρeff g d218 η,Δρeff=fcoat (ρm−ρf),η=ηref μw(Ts)μw(Tref).v = \frac{\Delta\rho_{eff}\, g\, d^2}{18\,\eta}, \qquad \Delta\rho_{eff} = f_{coat}\,(\rho_m - \rho_f), \qquad \eta = \eta_{ref}\,\frac{\mu_w(T_s)}{\mu_w(T_{ref})} .
  • ρm\rho_m, ρf\rho_f — densities of milk (1.028 g/cm³, lower end of the documented 1.028–1.038) and of milk fat (0.93 g/cm³).
  • fcoatf_{coat} — declared reduction (0.6–1.0) of the density difference by the casein layer that covers the new globule surface; casein is denser than fat.
  • η\eta — serum viscosity at the storage temperature TsT_s = 7 °C (legal maximum for pasteurized milk): the handbook value at about 35 °C (1.42 mPa·s) scaled with the viscosity of water μw\mu_w (IAPWS through CoolProp), about 2.8 mPa·s.

The test as a bound. In a column of height HH the top layer has height htop=H Vtop/Vtesth_{top} = H\,V_{top}/V_{test}. A globule class that rises a distance ss in the test time moves fat into the top layer (relative concentration 1+s/htop1 + s/h_{top}) and leaves a cream-free band at the bottom of the lower layer (relative concentration 1−s/(H−htop)1 - s/(H-h_{top})). Summed over the volume distribution, with m=∑iwisim = \sum_i w_i s_i the volume-weighted mean rise, the test f ctop<cbottomf\,c_{top} < c_{bottom} with ff = 0.9 passes while

m≤mmax=1−ff/htop+1/(H−htop).m \le m_{max} = \frac{1-f}{f/h_{top} + 1/(H - h_{top})} .

For HH = 0.2 m, mmaxm_{max} = 1.98 mm. Because s=v ttests = v\,t_{test} and v∝d2v \propto d^2,

m=Δρeff g ttest18 η ⟨d2⟩v,⟨d2⟩v=dv502 e2σ2m = \frac{\Delta\rho_{eff}\, g\, t_{test}}{18\,\eta}\,\langle d^2\rangle_v, \qquad \langle d^2\rangle_v = d_{v50}^2\,e^{2\sigma^2}

for a log-normal volume distribution with median dv50d_{v50} and log standard deviation σ\sigma. The test therefore fixes the largest allowed median diameter dv50,specd_{v50,spec} — about 0.64 µm with the point values.

Width of the homogenized distribution. Milk homogenized in two stages at 20 MPa has a volume median of 0.648 µm (the largest of the medians reported for 20, 25 and 30 MPa, all size parameters decreasing with pressure) and 21.87 % of its fat volume in globules larger than 1 µm. For a log-normal distribution the share above 1 µm is Φ(ln⁡(dv50/1 μm)/σ)\Phi(\ln(d_{v50}/1\,\mu m)/\sigma), which gives σ≈0.56\sigma \approx 0.56.

3. Size law and minimum pressure of the reference valve

In the turbulent jet behind the valve gap, the largest stable globule is set by the balance between the dynamic pressure of eddies of its own size and the Laplace pressure 4γ/d4\gamma/d (Kolmogorov–Hinze). In the inertial subrange this gives dmax∝(γ/ρ)3/5 ϵ−2/5d_{max} \propto (\gamma/\rho)^{3/5}\,\epsilon^{-2/5}, where ϵ\epsilon is the power dissipated per unit mass. The jet velocity is the Bernoulli velocity vgap=2p/ρv_{gap} = \sqrt{2p/\rho} and ϵ∝vgap3/ℓ∝p3/2\epsilon \propto v_{gap}^3/\ell \propto p^{3/2} for a fixed gap geometry ℓ\ell, hence

d∝p−3/5.d \propto p^{-3/5} .

Measured series in the industrial range fall more steeply: the coarse-tail shares of the 20, 25 and 30 MPa columns of the same table correspond, at constant width, to exponents of 0.87 and 1.08. The model writes

dv50(p)=dref(ppref)−b,pmin=pref(drefdv50,spec)1/b,d_{v50}(p) = d_{ref}\left(\frac{p}{p_{ref}}\right)^{-b}, \qquad p_{min} = p_{ref}\left(\frac{d_{ref}}{d_{v50,spec}}\right)^{1/b},

with bb declared uniform between 0.6 and 1.1. Because the specification diameter lies close to the reference diameter, pminp_{min} is insensitive to bb: about 20 MPa for the reference two-stage valve. This minimum belongs to that valve geometry. The BAT reference document shows that other geometries reach the same homogenization effect at 135 bar (efficient design) and 97 bar (advanced design) instead of 200 bar (conventional), so the pressure is a technology variable, not a physical constant; that is why the energy ceilings below the theoretical one are documented values (Section 4).

4. Energy balance — minimum energy per unit of product

Thermodynamic minimum. Homogenization changes no chemistry; it creates interface. The reversible work is the free energy of the new fat/serum interface,

wmin=γ ΔA,ΔA=6 xfρf(1d32,spec−1d32,0),w_{min} = \gamma\,\Delta A, \qquad \Delta A = \frac{6\,x_f}{\rho_f}\left(\frac{1}{d_{32,spec}} - \frac{1}{d_{32,0}}\right),
  • γ\gamma — tension of the casein-covered interface, the tension of purified milk fat against water (30.50 mN/m) minus the surface pressure of micellar casein at a milk-like concentration of 2.5 % (20.16 mN/m): 10.34 mN/m.
  • xfx_f — fat content of the milk (declared 3.0–4.0 %).
  • d32,0=6/(ρfSfat)d_{32,0} = 6/(\rho_f S_{fat}) — Sauter diameter of native globules from their specific surface (1.8 m²/g fat): 3.58 µm.
  • d32,spec=dv50,spec e−σ2/2d_{32,spec} = d_{v50,spec}\,e^{-\sigma^2/2} — Sauter diameter at the specification (about 0.54 µm).

The new area is about 350 m² per kg of milk — a 6.6-fold increase of the interface, slightly above the four- to six-fold increase the handbook states for typical homogenization, consistent with the test requiring fine globules — and wminw_{min} is a few joules per kilogram.

Hydraulic work and heating. A positive-displacement pump delivers the work p/ρp/\rho per kilogram, and the valve dissipates all of it in the jet; the milk warms by

ΔT=pρ cp,\Delta T = \frac{p}{\rho\, c_p},

0.25 K per MPa with cpc_p = 3.915 kJ/(kg K) — the handbook's rule of 1 °C per 4 MPa. At the minimum pressure the hydraulic work is about 20 kJ/kg; with the drive efficiency of the documented conventional machine (pV˙/Pelp\dot V/P_{el} = 200 bar × 25 000 L/h / 151 kW = 0.92), the electricity at the set-point pressure is about 23 kJ/kg. These two values carry no ceiling label (they belong to the reference valve).

Ratio to practice. wminw_{min} divided by the typical electricity is about 2×10−42\times10^{-4}: the handbook's statement that less than 1 % of the energy is used for homogenization is an understatement. The rest is turbulent dissipation at scales larger than the globules, heat in the valve and drive losses.

The four energy ceilings. Theoretical: γ ΔA\gamma\,\Delta A. Best available technique: advanced valve design, 75 kW at 25 000 L/h (97 bar), 10.5 kJ/kg. Documented: efficient design, 104 kW (135 bar), 14.6 kJ/kg. Typical: conventional design, 151 kW (200 bar), 21.2 kJ/kg. All three electric values are converted with the milk density and are for the same homogenization effect (75 % NIZO).

5. Rate-limiting step — time and throughput

Globule break-up in the gap takes microseconds (the handbook gives 10–15 µs; the gap velocity at the set point is about 210 m/s, within the documented 100–400 m/s), so kinetics are not limiting. The homogenizer is a positive-displacement pump: its throughput is the pump displacement, and the physics enters through the power it needs. The throughput per kW of drive at the set point of the reference valve,

V˙Pel=ηdrivepset,\frac{\dot V}{P_{el}} = \frac{\eta_{drive}}{p_{set}},

about 150 L/h per kW, is reported without a ceiling label because it depends on the valve. A homogenizer at its rated flow has an OEE performance factor of 1.

6. Process capability — quality ceiling

The homogenizing pressure under control is modelled as normal with standard deviation σp\sigma_p (valve wear, pump pulsation) around a set point zz standard deviations above the minimum:

pset=pmin+z σp,d=Φ(−z),Qmax=1−d.p_{set} = p_{min} + z\,\sigma_p, \qquad d = \Phi(-z), \qquad Q_{max} = 1 - d .

Milk homogenized below pminp_{min} fails the creaming test and is reworked. The margin costs z σp/ρz\,\sigma_p/\rho of hydraulic work per kilogram, a small amount.

7. Unavoidable stops — maximum availability

Two stops are forced. Hygiene regulation requires milk equipment to be cleaned at least once each day used; the homogenizer is cleaned in place with the pasteurizer, and the duration tCIPt_{CIP} is taken through a parameter link from the cleaning model (output cip_time_min). That model is valid for recirculated pipe and plate circuits; the pump block and valve are outside its validity, so the linked duration is a lower bound. The homogenizing valve and plunger seals wear under the pressure load and must be exchanged (interval tweart_{wear} and duration tservt_{serv} declared; the BAT reference document notes that lower pressures lengthen the interval):

Amax=1−tCIPtcyc−tservtwear.A_{max} = 1 - \frac{t_{CIP}}{t_{cyc}} - \frac{t_{serv}}{t_{wear}} .

8. Physics-bounded OEE

Scope of the bound. The theoretical label of OEEmaxOEE_{max} means physics-bounded under the declared scenario: the ceiling follows from the physics of the operation for the declared product, equipment and schedule, and it is a bound only conditional on those declarations. It follows the common definition of all models: AmaxA_{max} counts only stops required by regulation or forced by a physical mechanism, at their physical minimum duration, plus the start-up they force; PmaxP_{max} is the ratio of the physics-limited ideal cycle to the cycle of the capable process; QmaxQ_{max} is the share of conforming units, and mass losses of conforming product are reported only as yield.

OEEmax=Amax×Pmax×Qmax,Pmax=1.OEE_{max} = A_{max} \times P_{max} \times Q_{max}, \qquad P_{max} = 1 .

The result, with the Monte Carlo interval over the declared uncertainty of every input, is reported in the output table. The homogenizer sits in the pasteurization line; when both are modelled as one line, the daily cleaning is the same stop and must be counted once.

9. Assumptions and validity

See the assumption and validity lists generated from the configuration file. The specification diameter is most sensitive to the declared density reduction by the casein coat and to the height of the test column; the minimum pressure inherits that sensitivity, the energy ceilings below the theoretical one do not (they are documented). Cream, ice-cream mix and ultra-high-pressure homogenization follow other size laws and are outside the numbers of this model.

References

  • Tetra Pak Processing Systems. Dairy Processing Handbook, chapters 7.3 Homogenizers, 7.2 Centrifugal separators and milk standardization, and 2 The chemistry of milk (online edition).
  • Kim, S.-C., Yun, S.-Y., Ahn, N.-H., Kim, S.-M., Imm, J.-Y. (2020). Effect of homogenization pressure on plasmin activity and mechanical stress-induced fat aggregation of commercially sterilized ultra high temperature milk during storage. Food Science of Animal Resources, 40(5), 734–745. doi:10.5851/kosfa.2020.e48
  • Yan, G., Li, Y., Wang, S., Li, Y., Zhang, L., Yan, J., Sun, Y. (2024). Oil–water interfacial behaviour of different caseins and stability of emulsions: effect of micelle content and caseins concentrations. Food Chemistry: X, 23, 101784. doi:10.1016/j.fochx.2024.101784
  • Lopez, C., Cauty, C., Guyomarc'h, F. (2015). Organization of lipids in milks, infant milk formulas and various dairy products: role of technological processes and potential impacts. Dairy Science & Technology, 95, 863–893. doi:10.1007/s13594-015-0263-0
  • Hinze, J. O. (1955). Fundamentals of the hydrodynamic mechanism of splitting in dispersion processes. AIChE Journal, 1(3), 289–295. doi:10.1002/aic.690010303
  • Kolmogorov, A. N. (1991). The local structure of turbulence in incompressible viscous fluid for very large Reynolds numbers. Proceedings of the Royal Society A, 434, 9–13 (translation of the 1941 paper). doi:10.1098/rspa.1991.0075
  • U.S. Food and Drug Administration. Grade "A" Pasteurized Milk Ordinance, 2023 Revision.
  • European Commission, Joint Research Centre. Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries, 2019.
  • Bell, I. H., Wronski, J., Quoilin, S., Lemort, V. (2014). Pure and pseudo-pure fluid thermophysical property evaluation and the open-source thermophysical property library CoolProp. Industrial & Engineering Chemistry Research, 53, 2498–2508. doi:10.1021/ie4033999

7Parameters and provenance

Every input with its value, unit, role, uncertainty distribution and provenance. Open a quote to read the passage of the source that states the value.

8Benchmarks

Documented values from open sources used to check the ceilings; no documented value may beat the theoretical limit.
BenchmarkCeilingValueProvenance
Conventional homogenizer, 200 bar, 151 kW at 25 000 L/h (BAT reference document; electricity per kg computed with the milk density)
Show verbatim quote
Homogenisation pressure (bar) 200 135 97 Electric power needed (kW) 151 104 75Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 405. EUR 29978 EN; JRC118627; doi:10.2760/243911
Typical151 kWVerified quote
Advanced-design homogenizer, 97 bar, 75 kW at 25 000 L/h for the same homogenization effect (per-kilogram value: output energy_bat)
Show verbatim quote
Homogenisation pressure (bar) 200 135 97 Electric power needed (kW) 151 104 75Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 405. EUR 29978 EN; JRC118627; doi:10.2760/243911
Best available technique75 kWVerified quote
Documented statement that less than 1 % of the energy is used for the homogenization itself (compare with the output surface_energy_efficiency)
Show verbatim quote
Less than 1% of the energy is utilised for homogenization.Dairy Processing Handbook, chapter 7.3: Homogenizers (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.3 (read 2026-11-15)
—1 %Verified quote
Documented increase of the fat/plasma interfacial area by homogenization, lower end of 'four- to six-fold' (compare with the output area_increase_ratio)
Show verbatim quote
This is accompanied by a four- to six-fold increase in the fat/plasma interfacial surface area.Dairy Processing Handbook, chapter 7.3: Homogenizers (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.3 (read 2026-11-15)
—4 Verified quote
Documented temperature rise: 1 °C per 4 MPa of pressure drop (reproduced by the output temperature_rise_per_MPa)
Show verbatim quote
every 4 MPa (40 bar) in pressure drop over the device gives a temperature rise of 1 °CDairy Processing Handbook, chapter 7.3: Homogenizers (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.3 (read 2026-11-15)
—4 MPaVerified quote
Documented liquid velocity in the homogenizing gap, lower end of 100–400 m/s (compare with the output gap_velocity)
Show verbatim quote
The velocity of the liquid is normally 100 – 400 m/s in the narrow annular gap.Dairy Processing Handbook, chapter 7.3: Homogenizers (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.3 (read 2026-11-15)
—100 m/sVerified quote

9Unavoidable stops

Stops that physics or regulation impose; they set the maximum availability.
  • Cleaning at least once per day of useVerified quote

    Hygiene regulation: milk equipment must be cleaned at least once each day used. The homogenizer is cleaned in place with the pasteurizer circuit; minimum programme of the cleaning model (a lower bound for the pump block, assumption a5).

    Show verbatim quote
    all equipment is thoroughly cleaned at least once each day usedGrade "A" Pasteurized Milk Ordinance, 2023 Revision (opens in a new tab). U.S. Department of Health and Human Services, Public Health Service, Food and Drug Administration, 2023. p. 98. PMO 2023 Revision
  • Exchange of worn valve parts and plunger sealsVerified quote

    The homogenizing valve and the plunger seals wear under the pressure load and must be exchanged; interval and duration are declared scenarios. The BAT reference document notes that a lower pressure makes these intervals longer.

    Show verbatim quote
    A lower pressure also means less load on the homogeniser, so maintenance and worn parts’ replacement intervals will be less frequentBest Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 405. EUR 29978 EN; JRC118627; doi:10.2760/243911

10Assumptions

  1. Creaming of homogenized milk follows Stokes' law for independent globules (homogenized globules do not cluster by cold agglutination); a globule class rising a distance s enriches the top layer by s/h_top and depletes the bottom by s/(H - h_top) (valid while s is much smaller than the top layer).
  2. The homogenized size distribution is log-normal by volume, with the documented median at 20 MPa and the width fixed by the documented share of globules larger than 1 µm; its median scales as p^-b with an unchanged width.
  3. The serum viscosity at the storage temperature is the handbook reference value scaled with the viscosity of water (constant relative viscosity, IAPWS through CoolProp).
  4. The minimum pressure belongs to the reference two-stage valve of the documented size data; other valve geometries reach the same size at other pressures, so the energy ceilings below the theoretical one are documented values, not computed ones. Homogenizing pressure under control is normally distributed; milk homogenized below the minimum pressure fails the creaming test and is reworked.
  5. Availability: the daily cleaning of milk equipment at the minimum duration of the cleaning model (valid for recirculated pipe and plate circuits; the pump block and valve are outside that validity, so the linked duration is a lower bound) and the wear-forced exchange of valve parts and seals at declared intervals. The homogenizer is a positive-displacement pump: its rate is fixed by the pump, P = 1, and no product is lost in it (yield 1; the start-up mixing zones belong to the pasteurization line).

11Range of validity

  • Whole or semi-skimmed bovine milk homogenized warm (55–80 °C) in a two-stage valve homogenizer and stored at or below 7 °C.
  • Not for high-fat cream, ice-cream mix or ultra-high-pressure homogenization (above about 50 MPa), whose size laws differ.

12Sensitivity

Elasticity: relative change of the output for a relative change of each parameter. Spearman rank correlation from the Monte Carlo draws is shown alongside.
  • tcyct_{cyc}Maximum interval between two cleanings of milk equipment
    elasticity +0.016Spearman ρ —
  • tCIPt_{CIP}Minimum duration of a cleaning-in-place programme (output cip_time_min of the cleaning model, with its Monte Carlo interval)
    elasticity -0.016Spearman ρ -0.82
  • zzMargin of the pressure set point above the minimum pressure, in standard deviations
    elasticity +0.014Spearman ρ +0.48
  • tweart_{wear}Operating hours between exchanges of worn homogenizing valve parts and plunger seals
    elasticity +0.001Spearman ρ +0.15
  • tservt_{serv}Duration of the exchange of valve parts and seals
    elasticity -0.001Spearman ρ -0.13
  • Output rises with the parameter
  • Output falls with the parameter

13Validation checks

A model is published only when every check passes.
  • Dimensional analysis27 outputs converted to their declared unitsPassed
  • Every cited value verified against its source32 verified, 0 failedPassed
  • No documented value beats the theoretical limitPassed
  • Ceilings in physical orderPassed
  • Fractions within 0 and 1Passed

14Causes of loss in this operation

Probability that each cause is documented as a relevant loss of a line, pooled over all segments.

15References

  1. Dairy Processing Handbook, chapter 7.3: Homogenizers (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.3 (read 2026-11-15)
  2. Dairy Processing Handbook, chapter 2: The chemistry of milk (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 2 (read 2026-11-15)
  3. Dairy Processing Handbook, chapter 7.2: Centrifugal separators and milk standardization (opens in a new tab). Tetra Pak Processing Systems AB (online edition, openly readable), 2026. Tetra Pak Dairy Processing Handbook, online edition, chapter 7.2 (read 2026-11-15)
  4. Effect of Homogenization Pressure on Plasmin Activity and Mechanical Stress-Induced Fat Aggregation of Commercially Sterilized Ultra High Temperature Milk during Storage (Kim, Yun, Ahn, Kim, Imm; Food Science of Animal Resources 40(5): 734-745) (opens in a new tab). Korean Society for Food Science of Animal Resources (open access, CC BY-NC 3.0); full text served by Europe PMC, 2020. doi:10.5851/kosfa.2020.e48; PMC7492180
  5. Oil-water interfacial behaviour of different caseins and stability of emulsions: Effect of micelle content and caseins concentrations (Yan, Li, Wang, Li, Zhang, Yan et al.; Food Chemistry: X 23: 101784) (opens in a new tab). Elsevier (open access, CC BY-NC-ND 4.0); full text served by Europe PMC, 2024. doi:10.1016/j.fochx.2024.101784; PMC11403417
  6. Organization of lipids in milks, infant milk formulas and various dairy products: role of technological processes and potential impacts (Lopez, Cauty, Guyomarc'h; Dairy Science & Technology 95: 863-893) (opens in a new tab). Springer / INRA (open access); full text served by Europe PMC, 2015. doi:10.1007/s13594-015-0263-0; PMC4641158
  7. Grade "A" Pasteurized Milk Ordinance, 2023 Revision (opens in a new tab). U.S. Department of Health and Human Services, Public Health Service, Food and Drug Administration, 2023. p. 57. PMO 2023 Revision
  8. Best Available Techniques (BAT) Reference Document for the Food, Drink and Milk Industries (opens in a new tab). European Commission, Joint Research Centre; Publications Office of the European Union, 2019. p. 405. EUR 29978 EN; JRC118627; doi:10.2760/243911
  9. doi:10.1002/aic.690010303 — Maximum stable drop size in turbulent dispersion (Hinze).
  10. doi:10.1098/rspa.1991.0075 — Local structure of turbulence (Kolmogorov, English translation of the 1941 paper).
  11. doi:10.5851/kosfa.2020.e48 — Fat globule size distributions of milk homogenized at 20, 25 and 30 MPa.
  12. doi:10.1016/j.fochx.2024.101784 — Interfacial tension of milk fat against water and casein dispersions.
  13. doi:10.1021/ie4033999 — Thermophysical property library (viscosity of water).

Processes that use this operation