Drybench Bioreactor: methods and checks

How Drybench Bioreactor reaches its results: every calculator, analysis or view with its methods, the references behind them, how far each has been checked, and the app's limits.

Open Drybench Bioreactor · What the levels mean · Every app's methods and checks

Run

The culture over time from the preset's values and your changes: biomass, substrate, product, volume and dissolved oxygen, by the growth model and the oxygen balance, with the vessel's kLa from the correlation or your own value.

Its result goes through 7 methods. The lowest level among them: Tested, not reference-checked.

Below the top level: Yield feasibility: the true yield Yxs (g/g) is refused when it exceeds min(MXc/MSc, γS·MXc/(γX·MSc)), the carbon and electron bounds, where MXc is 12.011 g over the biomass carbon fraction and MSc and γS come from the substrate's formula per C-mol; above them growth would need negative CO₂ production or negative O₂ uptake. The product, counted as biomass, falls under the same bound. (Tested, not reference-checked).

Equation

Growth (Monod, or Haldane and Andrews with Ki): f(S) = S/(Ks + S + S²/Ki); f(C) = C/(Ko + C), multiplied (a model choice)
Uptake and maintenance (Pirt): qS = (μmax/Y + ms)·f(S)·f(C); μ = (μmax + Y·ms)·f(S)·f(C) − Y·ms where positive
Product (Luedeking and Piret): qP = α·μ + β, from the same budget as growth: μ + qP = Y·max(0, qS − ms)
Balances with volume change: dX/dt = μX − (F/V)X; dS/dt = −qS·X + (F/V)(Sf − S); dP/dt = qP·X − (F/V)P; dV/dt = F (a chemostat draws off F = D·V)
Dissolved oxygen: dC/dt = kLa(C* − C) − OUR + (Fin/V)(Cfeed − C), C* from the exhaust gas; OUR from the degree-of-reduction balance, CER from the carbon balance, RQ = CER/OUR
kLa: kLa = a·(Pg/V)^b·vs^c (Van 't Riet), Pg from Hughmark's ratio; or your value, scaled by the same exponents; or your readings, log-log

Limits that apply here

• The yield check refuses only what the carbon and electron balances forbid; a feasible yield is not thereby a likely one.
• A complex medium is context; the model simulates one declared substrate; delivery is not verified.
• Matched by exact compound identity against a short ruled list; names are not read; glucose monohydrate is converted to anhydrous glucose by formula mass, assuming the exact monohydrate.

Methods

Unstructured growth model: Monod (Haldane and Andrews for methanol), Pirt's maintenance, Luedeking and Piret's product, balances with volume change

• Level: Checked against a reference
• Reference: Monod J. Annu Rev Microbiol 1949;3:371-394; Andrews JF. Biotechnol Bioeng 1968;10:707-723; Pirt SJ. Proc R Soc B 1965;163:224-231; Luedeking R, Piret EL. J Biochem Microbiol Technol Eng 1959;1:393-412. Oxygen and CO₂ from carbon and degree-of-reduction balances, ammonium the nitrogen source. doi:10.1146/annurev.mi.03.100149.002103
• How it was checked: State variables checked against SciPy 1.17.0 Radau on 33 scenarios (SciPy's Radau at a relative tolerance of 1e-9, an absolute tolerance a hundredth of the engine's for every state): each point within the larger of 1e-5 of its value, a time shift of max(1e-4 h, 1e-5 of the elapsed time) times its slope, and its floor; inside the window after a change of aeration or feed, the series computed from dissolved oxygen on value alone; events within 1e-3 h; the carbon and electron balances close within 0.5 percent; the right-hand sides agree within 1e-12 relative at 13 states (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Modified Rosenbrock (W-) formula with an analytic Jacobian, error control, its continuous extension and located events

• Level: Checked against a reference
• Reference: Shampine LF, Reichelt MW. The MATLAB ODE Suite. SIAM J Sci Comput 1997;18(1):1-22, §3.1 (implemented from the paper only). doi:10.1137/S1064827594276424
• How it was checked: Checked against SciPy 1.17.0 Radau on the same 33 scenarios, with one step and the linear solve compared with NumPy 2.3.5 and LAPACK, and fixed-step RK4's failure at kLa 1,000 per hour kept as a regression test (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Dissolved oxygen by its own balance, dC/dt = kLa(C* − C) − OUR + (Fin/V)(Cfeed − C), with C* from the exhaust gas and Benson and Krause's solubility

• Level: Checked against a reference
• Reference: Benson and Krause, as the USGS documents them: Office of Water Quality Technical Memorandum 2011.03, attachment eqs. 7 and 9 to 11 (read 6 October 2026).
• How it was checked: Air saturation checked against the USGS DOTABLES program's own output at 9 temperatures and pressures, and the exhaust-gas balance on 3 cases, by SciPy 1.17.0 (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

kLa by Van 't Riet's correlation with Hughmark's gassed power (coalescing constants as Yawalkar et al. 2002 quote them)

• Level: Checked against a reference
• Reference: Van 't Riet K. Ind Eng Chem Process Des Dev 1979;18:357-364; Yawalkar AA et al. Can J Chem Eng 2002;80:840-848 (Tables 2 and 3). The non-coalescing form is 0.002(Pg/V)^0.7 vs^0.2, kLa in s−1, power density W/m³ and velocity m/s. Experimental ranges are approximate. Rushton Np = 5 is a nominal default, not a physical constant: Kaiser SC et al. Eng Life Sci 2017;17(5):500-511, doi:10.1002/elsc.201600096 (§3.3.1, Table 3; read 24 October 2026) compile 3.6 to 5.9 for fully turbulent Rushton turbines in four-baffled flat-bottomed vessels, varying with D/T and blade thickness, and measured 4.17 ± 0.14 themselves (three baffles, torospherical base). Across that range Pg/V moves by −28 to +18 percent and kLa by −12 to +7 percent. Additive multi-impeller power needs calibration. doi:10.1021/i260071a001
• How it was checked: The correlation checked against the same equations in SciPy 1.17.0 at 10 settings, two outside the range (2026-10-07); the coalescing coefficients are read from Yawalkar 2002 (Table 2); the non-coalescing coefficients and the approximate domain are cited to Van 't Riet 1979 and remain Unverified until the paper is read; Hughmark geometry limits are sourced through Yawalkar 2002. Np=5 remains a nominal geometry assumption. These checks establish equation consistency, not predictive accuracy in a particular broth. Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

A kLa value at the setup's agitation and gas flow, scaled to other settings by Van 't Riet's exponents with Hughmark's gassed-power ratio (the power number cancels); your readings at several rpm interpolated log-log

• Level: Checked against a reference
• Reference: Yawalkar AA et al. Can J Chem Eng 2002;80:840-848, Tables 2 and 3 (the exponents and Hughmark's gassed power). doi:10.1002/cjce.5450800507
• How it was checked: Checked against an independent NumPy float64 reference at 18 settings (value and readings; agitation, gas flow, volume, reference-basis flow, Hughmark's cap) within 1e-12 relative (verification 0.6.0, 24 October 2026). The non-coalescing exponents are checked as arithmetic only and remain Unverified. Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Feed laws (constant, or exponential at a set growth rate with a pump limit) and the pump sheet's interval averages

• Level: Checked against a reference
• How it was checked: Checked against an independent Python decimal reference at 40 digits: 12 feeds (constant, exponential, pump limit, the rescue's cap, the Vmax and end-time stops), every pump-sheet row and total, and 4 design rates, within 1e-12 relative (verification 0.7.0, 24 October 2026). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Yield feasibility: the true yield Yxs (g/g) is refused when it exceeds min(MXc/MSc, γS·MXc/(γX·MSc)), the carbon and electron bounds, where MXc is 12.011 g over the biomass carbon fraction and MSc and γS come from the substrate's formula per C-mol; above them growth would need negative CO₂ production or negative O₂ uptake. The product, counted as biomass, falls under the same bound.

• Level: Tested, not reference-checked
• How it was checked: Unit-tested: a yield above the bound is refused with infeasible-yield. The bound is not checked against an independent reference.

Rescue

When dissolved oxygen would fall below the setpoint, the cascade raises the agitation, then the gas flow, then the inlet's oxygen, each within the limits you enter, and last caps the feed. Each step is the smallest on its grid that holds the setpoint to the end of the run.

Its result goes through 4 methods. The lowest level among them: Tested, not reference-checked.

Below the top level: Oxygen rescue cascade: agitation, gas flow, inlet O₂, then a feed cap, each the smallest step on its grid that holds the setpoint, within the vessel's limits (Tested, not reference-checked).

Limits that apply here

• The rescue cascade stays within the agitation, gas-flow and oxygen-enrichment limits you enter, and ignores cooling capacity, vessel pressure and gas-safety limits. Oxygen enrichment near flammable solvents or vapours is a fire and explosion hazard. Do not run a vessel from these outputs without your own engineering review.

Methods

Oxygen rescue cascade: agitation, gas flow, inlet O₂, then a feed cap, each the smallest step on its grid that holds the setpoint, within the vessel's limits

• Level: Tested, not reference-checked
• How it was checked: Unit-tested: every step stays within the vessel's limits, each is the smallest on its grid that holds the setpoint, and the cascade's order holds (6 October 2026). The rescued runs are oracle scenarios like any other.

Unstructured growth model: Monod (Haldane and Andrews for methanol), Pirt's maintenance, Luedeking and Piret's product, balances with volume change

• Level: Checked against a reference
• Reference: Monod J. Annu Rev Microbiol 1949;3:371-394; Andrews JF. Biotechnol Bioeng 1968;10:707-723; Pirt SJ. Proc R Soc B 1965;163:224-231; Luedeking R, Piret EL. J Biochem Microbiol Technol Eng 1959;1:393-412. Oxygen and CO₂ from carbon and degree-of-reduction balances, ammonium the nitrogen source. doi:10.1146/annurev.mi.03.100149.002103
• How it was checked: State variables checked against SciPy 1.17.0 Radau on 33 scenarios (SciPy's Radau at a relative tolerance of 1e-9, an absolute tolerance a hundredth of the engine's for every state): each point within the larger of 1e-5 of its value, a time shift of max(1e-4 h, 1e-5 of the elapsed time) times its slope, and its floor; inside the window after a change of aeration or feed, the series computed from dissolved oxygen on value alone; events within 1e-3 h; the carbon and electron balances close within 0.5 percent; the right-hand sides agree within 1e-12 relative at 13 states (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Modified Rosenbrock (W-) formula with an analytic Jacobian, error control, its continuous extension and located events

• Level: Checked against a reference
• Reference: Shampine LF, Reichelt MW. The MATLAB ODE Suite. SIAM J Sci Comput 1997;18(1):1-22, §3.1 (implemented from the paper only). doi:10.1137/S1064827594276424
• How it was checked: Checked against SciPy 1.17.0 Radau on the same 33 scenarios, with one step and the linear solve compared with NumPy 2.3.5 and LAPACK, and fixed-step RK4's failure at kLa 1,000 per hour kept as a regression test (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Dissolved oxygen by its own balance, dC/dt = kLa(C* − C) − OUR + (Fin/V)(Cfeed − C), with C* from the exhaust gas and Benson and Krause's solubility

• Level: Checked against a reference
• Reference: Benson and Krause, as the USGS documents them: Office of Water Quality Technical Memorandum 2011.03, attachment eqs. 7 and 9 to 11 (read 6 October 2026).
• How it was checked: Air saturation checked against the USGS DOTABLES program's own output at 9 temperatures and pressures, and the exhaust-gas balance on 3 cases, by SciPy 1.17.0 (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Pump sheet

The rate to set for each interval, from the time in its row, and the volume fed so far. Each rate is the feed law's average over its interval, so the volumes add up; holding the rate at the start of an hour would underfeed an exponential feed.

Its result goes through 2 methods. The lowest level among them: Checked against a reference.

Limits that apply here

• The model's feed law; pump calibration, line dead volume and feed density at temperature are yours to check.
• A complex medium is context; the model simulates one declared substrate; delivery is not verified.

Methods

Feed laws (constant, or exponential at a set growth rate with a pump limit) and the pump sheet's interval averages

• Level: Checked against a reference
• How it was checked: Checked against an independent Python decimal reference at 40 digits: 12 feeds (constant, exponential, pump limit, the rescue's cap, the Vmax and end-time stops), every pump-sheet row and total, and 4 design rates, within 1e-12 relative (verification 0.7.0, 24 October 2026). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Feed hand-off: the planned schedule with units, its mass flow and substrate mass, and the substrate matched by exact identity

• Level: Checked against a reference
• How it was checked: Checked against an independent Python fractions reference: 14 schedules over the feed's published pump sheets (rates in L/h and mL/min, mass flow, volume and substrate mass fed), 15 feed programmes with their stops, totals fed and refusals, and 57 substrate matches on the shipped identity table, its keys recomputed with RDKit from a PubChem snapshot, with exact conversions to g/L and their refusals, within 1e-12 relative (verification 0.12.0, 24 October 2026). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Measured data beside this run

Review a growth time course beside this simulation. The selected-window slope is not the model’s μmax. Optical density is not converted to biomass, and no process-model fit or cross-dataset residual is calculated here.

Its result goes through 1 method. The lowest level among them: Checked against a reference.

Limits that apply here

• An apparent ln(response) slope over your window; not the model's μmax; no OD-to-biomass conversion; no interval.

Methods

Specific growth rate in a selected window

• Level: Checked against a reference
• How it was checked: Assay's method, checked by Assay's growth-window set (9 cases): Checked against exact rational least squares (Python fractions) on 5 hand cases with whole doublings and SciPy linregress on 3 noisy cases, with the withheld cases exact: μ and the doubling time within 1e-9 relative, hours against minutes, and blank-then-log distinguished from log-then-blank (10 October 2026). R not run on the 3 noisy cases. A receipt imported from Assay is kept as recorded; a new window receipt here is computed by the same method.

New comparison

Subtracts a user-declared numeric event anchor in source time units and divides by 1 (hours) or 60 (minutes) without altering raw responses, row order or duplicate timestamps.

Its result goes through 1 method. The lowest level among them: Checked against a reference.

Limits that apply here

• Elapsed time from a declared event is not physiological equivalence; responses are not converted or overlaid on one axis; a simulation reference is context only.

Methods

Measured-run elapsed-time anchor alignment

• Level: Checked against a reference
• How it was checked: Elapsed hours checked against exact rational arithmetic (Python fractions): Bioreactor verification 0.3.0, 8 cases (19 October 2026; its 12 refusal entries carry no inputs), and 0.4.0, 3 cases with a failed simulation reference, the snapshot SHA-256 recomputed by hashlib (20 October 2026). Verification 0.5.0 adds 25 cases with complete inputs: every refusal code and the simulation-reference alignment (Python 3.12.14 fractions and hashlib, 24 October 2026). Elapsed hours exact where the reduced fraction's denominator is a power of two and within 1e-15 h otherwise; row identity and order, raw responses, roles, exclusions and qualification codes exact. Draft until the independent bioprocess reviewer signs (Q-19).

kLa calculator

Calculate kLa from power per volume and superficial gas velocity, with the correlation's valid range stated, or enter your measured kLa.

It has a page of its own: kLa calculator.

Its result goes through 2 methods. The lowest level among them: Checked against a reference.

Methods

kLa by Van 't Riet's correlation with Hughmark's gassed power (coalescing constants as Yawalkar et al. 2002 quote them)

• Level: Checked against a reference
• Reference: Van 't Riet K. Ind Eng Chem Process Des Dev 1979;18:357-364; Yawalkar AA et al. Can J Chem Eng 2002;80:840-848 (Tables 2 and 3). The non-coalescing form is 0.002(Pg/V)^0.7 vs^0.2, kLa in s−1, power density W/m³ and velocity m/s. Experimental ranges are approximate. Rushton Np = 5 is a nominal default, not a physical constant: Kaiser SC et al. Eng Life Sci 2017;17(5):500-511, doi:10.1002/elsc.201600096 (§3.3.1, Table 3; read 24 October 2026) compile 3.6 to 5.9 for fully turbulent Rushton turbines in four-baffled flat-bottomed vessels, varying with D/T and blade thickness, and measured 4.17 ± 0.14 themselves (three baffles, torospherical base). Across that range Pg/V moves by −28 to +18 percent and kLa by −12 to +7 percent. Additive multi-impeller power needs calibration. doi:10.1021/i260071a001
• How it was checked: The correlation checked against the same equations in SciPy 1.17.0 at 10 settings, two outside the range (2026-10-07); the coalescing coefficients are read from Yawalkar 2002 (Table 2); the non-coalescing coefficients and the approximate domain are cited to Van 't Riet 1979 and remain Unverified until the paper is read; Hughmark geometry limits are sourced through Yawalkar 2002. Np=5 remains a nominal geometry assumption. These checks establish equation consistency, not predictive accuracy in a particular broth. Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

A kLa value at the setup's agitation and gas flow, scaled to other settings by Van 't Riet's exponents with Hughmark's gassed-power ratio (the power number cancels); your readings at several rpm interpolated log-log

• Level: Checked against a reference
• Reference: Yawalkar AA et al. Can J Chem Eng 2002;80:840-848, Tables 2 and 3 (the exponents and Hughmark's gassed power). doi:10.1002/cjce.5450800507
• How it was checked: Checked against an independent NumPy float64 reference at 18 settings (value and readings; agitation, gas flow, volume, reference-basis flow, Hughmark's cap) within 1e-12 relative (verification 0.6.0, 24 October 2026). The non-coalescing exponents are checked as arithmetic only and remain Unverified. Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

OTR and OUR calculator

The oxygen transfer rate a vessel can give against the oxygen uptake rate a culture needs, from kLa, C* and the cells' demand.

It has a page of its own: OTR and OUR calculator.

Its result goes through 1 method. The lowest level among them: Checked against a reference.

Methods

Dissolved oxygen by its own balance, dC/dt = kLa(C* − C) − OUR + (Fin/V)(Cfeed − C), with C* from the exhaust gas and Benson and Krause's solubility

• Level: Checked against a reference
• Reference: Benson and Krause, as the USGS documents them: Office of Water Quality Technical Memorandum 2011.03, attachment eqs. 7 and 9 to 11 (read 6 October 2026).
• How it was checked: Air saturation checked against the USGS DOTABLES program's own output at 9 temperatures and pressures, and the exhaust-gas balance on 3 cases, by SciPy 1.17.0 (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Scale-up calculator

Agitation at a larger scale by three rules side by side: constant power per volume, constant tip speed and constant kLa, with their assumptions stated.

It has a page of its own: Scale-up calculator.

Its result goes through 1 method. The lowest level among them: Checked against a reference.

Limits that apply here

• Scale-up assumes geometrically similar vessels with the same impellers, Np and gas flow per volume; constant P/V holds ungassed power, so gassed P/V differs between scales. It ignores cooling, mixing time and shear. Every kLa is the correlation's, flagged outside its fitted range, with no prediction interval.

Methods

Scale-up at constant power per volume, tip speed or kLa

• Level: Checked against a reference
• How it was checked: Checked against the same rules worked in SciPy 1.17.0 on 3 cases (2026-10-07). Draft until the independent bioprocess reviewer signs the equations, the presets and the fixtures (Q-19).

Data sources

• Monod 1949; Andrews 1968 (growth and substrate inhibition): Method (cited)
• Pirt 1965; Luedeking and Piret 1959 (maintenance and product): Method (cited)
• Shampine and Reichelt 1997 (the solver): Method (cited)
• Benson and Krause, as documented by the USGS (oxygen's solubility): Method (cited); a US government work in the public domain
• Van 't Riet 1979; Hughmark 1980; Yawalkar et al. 2002 (kLa and gassed power): Method (cited)

Limits

A simulation, not a forecast. Every output follows from the model equations and the parameters you entered. It is a teaching and estimation model, not a validated process model, and not a prediction for any real strain, vessel or process.

• Teaching and estimation model, not a validated process model.
• The rescue cascade stays within the agitation, gas-flow and oxygen-enrichment limits you enter, and ignores cooling capacity, vessel pressure and gas-safety limits. Oxygen enrichment near flammable solvents or vapours is a fire and explosion hazard. Do not run a vessel from these outputs without your own engineering review.
• Presets combine literature values and operating choices from multiple conditions; they are teaching scenarios, not validated strain/process parameter sets. Quantitative process use requires measured parameters and specialist validation.
• Literature growth-fitted Ks is not interchangeable with the uptake Ks used here. Without product/inhibition and with excess oxygen, growth half-saturation is Ks × (1 + 2Yms/μmax). Presets are not silently transformed without refitting.
• Nominal Rushton Np=5 assumes standard baffled turbulent geometry (Re above 10⁴, Kaiser et al. 2017); the app does not compute Reynolds number or broth rheology. Vessel geometry, multiple impellers and real-broth transfer need measurement.
• An unstructured uptake/Monod model with mixed literature parameters has not been quantitatively validated for a given strain or process. Reported parameter ranges do not establish predictive uncertainty for these composite presets.
• The Crabtree effect is not modelled; the S. cerevisiae preset stays glucose-limited, and a run that leaves glucose limitation or runs short of oxygen is flagged.
• The acetate onset is MG1655's; other strains differ.
• CHO and other sparged mammalian cultures are not modelled for oxygen limitation.
• The DO setpoint is a process margin, not a kinetic threshold.
• The gas phase is taken at steady state (no hold-up). Feed oxygen enters dC/dt through (Fin/V)(Cfeed−C); zero feed DO is the default. The chart shows oxygen in the liquid; a DO probe's reading lags it by the probe's response time, which is not modelled.
• Excursions past a limit that last less than about 3 minutes are not flagged: the model leaves out the gas phase's hold-up and the DO probe's response, which act over about a minute.
• Gas flow has an explicit actual or reference basis, including reference temperature and absolute pressure. The same ideal dry-inlet-gas conversion supplies molar flow, superficial velocity and gassed power. Gas-flow humidification and hydrostatic expansion are not modelled; freshwater solubility includes its water-vapour correction, but broth salinity and composition require measurement. Vvm is per current liquid volume; flow rises during feeding while kLa is held between aeration changes.
• Van 't Riet's empirical correlation supplies no validated prediction interval for your vessel or broth. Yawalkar et al. 2002 report substantial deviations for some conditions; use measured kLa when available.
• Real broths fall anywhere between the coalescing form used by default and the non-coalescing form; enter a measured kLa if you have one.
• The presets' 10 L vessel sits at the lower edge of the tank sizes Hughmark's correlation was fitted on (0.21 to 3.33 m).
• kLa is taken at the volume where the aeration is set, at the start or at a rescue step, and held as the volume grows.
• The product is biomass-equivalent for substrate yield, carbon fraction and degree of reduction. Product-enabled OUR, CER and RQ are approximate for chemically different products.
• When uptake falls below maintenance, nothing grows and nothing decays: death and endogenous decay are not modelled.
• The CO₂ evolution rate is the CO₂ produced: dissolved CO₂ and bicarbonate are not modelled.
• No heat transfer, pH control, parameter fitting or Monte Carlo.
• Scale-up assumes geometrically similar vessels with the same impellers, Np and gas flow per volume; constant P/V holds ungassed power, so gassed P/V differs between scales. It ignores cooling, mixing time and shear. Every kLa is the correlation's, flagged outside its fitted range, with no prediction interval.
• The yield check refuses only what the carbon and electron balances forbid; a feasible yield is not thereby a likely one.
• The model's feed law; pump calibration, line dead volume and feed density at temperature are yours to check.
• A complex medium is context; the model simulates one declared substrate; delivery is not verified.
• Matched by exact compound identity against a short ruled list; names are not read; glucose monohydrate is converted to anhydrous glucose by formula mass, assuming the exact monohydrate.
• An apparent ln(response) slope over your window; not the model's μmax; no OD-to-biomass conversion; no interval.
• Elapsed time from a declared event is not physiological equivalence; responses are not converted or overlaid on one axis; a simulation reference is context only.
• Estimates one parameter, conditional on every other value as entered; a wrong fixed value biases it without warning.
• Needs measured biomass as mass concentration and measured substrate; optical density alone is refused.
• No confidence interval; a profile is a numerical diagnostic.
• Values marked “Unverified” have not yet been checked against their sources; each is checked, or removed, before this app leaves preview.
• A link carries the preset and every parameter you changed, never results; a run you keep stays in this browser. No account needed.
• No offline use: the pages are served by the site, and nothing is installed.