Cell Doubling Time Calculator

Background.

A cell doubling time calculator turns two measurements of the same culture into the two numbers that actually describe how fast it is growing: the doubling time (microbiologists call it the generation time) and the specific growth rate constant μ. You give it a reading at one timepoint, a reading at a later timepoint, and the interval between them. It fits the first-order growth model N(t) = N₀·e^(μt) and returns μ in reciprocal hours, the doubling time in hours and minutes, how many generations passed, the division rate, and the fold change. The readings can be blank-corrected OD₆₀₀ absorbance, cells per millilitre from a haemocytometer, or CFU per millilitre from a plate count — the unit cancels in the ratio N/N₀, so the only requirement is that both readings use the same unit and the same instrument settings.

Two conditions decide whether the answer means anything, and both belong here rather than buried in an FAQ. First, both readings must lie inside the exponential phase. Hall and colleagues put it plainly in the paper that accompanies the widely used GrowthRates program: only the exponential portions of a growth curve are useful for determining growth rates. A reading taken during lag, or after the culture has begun to decelerate towards stationary phase, will pull the fitted rate towards zero and inflate the doubling time. Second, if you are reading optical density, the reading has to be inside your instrument's linear range. Myers, Curtis and Curtis measured deviation from linearity for an E. coli suspension at around OD 0.8, and noted that dilution to below roughly OD 0.5 is often recommended. Above that, extra cells stop producing proportionally more signal and the culture looks slower than it is.

It also matters what a doubling time is a property of. It is not a property of the species. It is a property of that strain in that medium, at that temperature, with that aeration. Escherichia coli K-12 MG1655 doubles roughly every 20 minutes in Luria-Bertani broth at 37 °C — Sezonov, Joseleau-Petit and D'Ari measured exactly that during steady-state growth — while the same organism in a minimal salts medium with a poor carbon source may take several hours, and Mycobacterium tuberculosis takes most of a day. Reporting a doubling time without its condition is reporting half a result, which is why this page asks you to record temperature, medium and shaking speed alongside the number it gives you.

Optical density deserves one more caution, because it is the measurement most people feed into this calculator. OD is a light-scattering proxy for biomass, not a cell count, and it is instrument-specific. The 2020 interlaboratory study by Beal and colleagues across 244 laboratories states the problem directly: optical density cannot be compared between instruments without a standardized calibration protocol and is challenging to relate to actual cell count. For a doubling time this matters less than it does for a headcount, because a consistent scaling error cancels in the ratio — but a nonlinear one does not, and that is precisely what happens above the linear range.

The calculator implements the exact logarithmic form rather than the rounded shortcut many textbooks print. Where a course handout writes g = t ÷ (3.3·log₁₀(N/N₀)), the 3.3 is a rounded 1/log₁₀2 = 3.321928, and using it overstates every generation time by about 0.66 percent. On the worked example below that is the difference between 30.000 and 30.199 minutes. If your textbook answer sits slightly above this page's, that constant is almost certainly why.

What is cell doubling time calculator?

Doubling time, or generation time, is the interval a population needs to double in number while growing exponentially. It is the biological twin of half-life: where half-life describes first-order decay, doubling time describes first-order growth. The underlying model is dN/dt = μN, whose solution is N(t) = N₀·e^(μt). Rearranged for the rate constant, μ = ln(N/N₀)/Δt, with units of reciprocal time; setting N = 2N₀ gives the doubling time t_d = ln2/μ. The number of generations that passed is log₂(N/N₀), and the division rate v is simply 1/t_d. Because all four quantities — μ, t_d, n and v — are algebraic rearrangements of each other, quoting any one of them fixes the others; this page returns all four so you never have to convert by hand. In predictive microbiology the same constant appears as μ_max, defined by Zwietering and colleagues as the slope of the tangent to ln(N/N₀) against time at the inflection point of the growth curve, which is the same natural-logarithm basis used here. In mammalian cell culture the same quantity is called the population doubling time, and the count of doublings since thaw is the population doubling level. What doubling time is not: it is not a measure of how many cells you have, it is not a viability measurement, and it is not transferable between conditions.

How to use this calculator.

1. 01Take your first reading during the exponential phase — for a shake flask, once the culture has clearly left lag but is still well below the plateau.
2. 02If you are using OD₆₀₀, subtract a blank of the same sterile medium, and dilute any sample reading above about 0.5 so you stay inside the instrument's linear range.
3. 03Enter that blank-corrected value as the first reading (N₀).
4. 04Take a second reading later in the same exponential phase and enter it as N. It must be larger than the first.
5. 05Enter the interval between the two readings in hours — 1.5 for ninety minutes, 0.6667 for forty.
6. 06Read the doubling time in minutes as the headline result, and the specific growth rate μ beside it; they are the same measurement expressed two ways.
7. 07Check the generations figure for a sanity test: an eight-fold rise must be exactly three generations, a four-fold rise exactly two.
8. 08Record the temperature, medium and aeration next to the result — a doubling time without its condition is not reproducible.

The formula.

Exponential growth is defined by the statement that the rate of increase is proportional to the amount already present: dN/dt = μN, where μ is the first-order growth rate constant. Hall, Acar, Nandipati and Barlow state the model in exactly this form and note that α — their symbol for μ — is in reciprocal time units, so if time is measured in minutes the growth rate is per minute. Integrating gives N = N₀·e^(μt), and taking natural logs of both sides gives ln N = ln N₀ + μt. That is the whole method: μ is the slope of ln N against time, and in practice, as the same paper says, it is the slope of ln OD against time. With only two points that slope is ln(N/N₀) divided by the elapsed interval, which is what this calculator computes.

Doubling time follows by asking when N reaches 2N₀. Substituting gives 2 = e^(μ·t_d), so ln 2 = μ·t_d and t_d = ln2/μ = 0.693147/μ. Because ln 2 is a fixed number, μ and t_d carry identical information: a culture with μ = 1.386 h⁻¹ has a doubling time of 0.5 h and vice versa. The number of generations elapsed is the base-2 logarithm of the fold change, n = log₂(N/N₀) = ln(N/N₀)/ln2, and this equals Δt/t_d. The division rate v = 1/t_d expresses the same thing as divisions per hour.

Dimensional analysis confirms the arrangement. N/N₀ is a ratio of two quantities in the same unit, so it is dimensionless and its logarithm is a pure number. Dividing a pure number by a time gives reciprocal time, so μ is in h⁻¹. Dividing ln 2, a pure number, by h⁻¹ returns hours, so t_d is a time. Generations, being a pure number divided by a pure number, are dimensionless. The unit of the readings never enters the result, which is why OD₆₀₀ and CFU/mL can be used interchangeably in this calculator provided you do not mix them between the two readings.

On rounding: every intermediate here — the ratio, its natural logarithm, ln 2, the division — is carried at twenty significant digits and rounded exactly once, at the moment the result is returned. Nothing is rounded mid-calculation, and there are no bands or thresholds to fall across. The one genuine boundary is the domain edge at N = N₀. There, μ = 0 and t_d = ln2 ÷ 0 is undefined — a singularity, not a large number — so the calculator refuses the input rather than printing an infinity. It refuses N < N₀ for the same reason: a shrinking culture is not in exponential growth, and a negative doubling time is not a meaningful quantity.

One widely taught variant deserves an explicit note because it does not agree with this page to the last digit. Many introductory texts write the generation time as g = t ÷ (3.3 · log₁₀(N/N₀)). That is the same equation only if 3.3 equals 1/log₁₀2, which is 3.321928. Because 3.3 is smaller, the shortcut computes fewer generations and therefore a longer generation time — by 0.66 percent, systematically, in every result. On the worked example below the exact answer is 30.000 minutes and the 3.3 shortcut gives 30.199 minutes. This calculator uses the exact constant.

A worked example.

Frequently asked questions.

References& sources.

1. [1]Hall BG, Acar H, Nandipati A, Barlow M (2014). "Growth Rates Made Easy." Molecular Biology and Evolution 31(1):232–238. Model section: dN/dt = αN; N = N₀e^(αt); α is the slope of ln N versus t and is in reciprocal time units. Retrieved 2026-07-29.
2. [2]Zwietering MH, Jongenburger I, Rombouts FM, van 't Riet K (1990). "Modeling of the Bacterial Growth Curve." Applied and Environmental Microbiology 56(6):1875–1881. Independent second authority: defines the maximum specific growth rate μ_max as the tangent to ln(N/N₀) against time, and the lag time λ as that tangent's x-axis intercept. Retrieved 2026-07-29.
3. [3]Sezonov G, Joseleau-Petit D, D'Ari R (2007). "Escherichia coli physiology in Luria-Bertani broth." Journal of Bacteriology 189(23):8746–8749. Strain K-12 MG1655; "The doubling time during steady-state growth in Luria-Bertani broth was 20 min." Retrieved 2026-07-29.
4. [4]Myers JA, Curtis BS, Curtis WR (2013). "Improving accuracy of cell and chromophore concentration measurements using optical density." BMC Biophysics 6:4. Reports deviation from linearity for an E. coli sample at OD ≈ 0.8 and notes that dilution to below OD ≈ 0.5 is often recommended. Retrieved 2026-07-29.
5. [5]Beal J, Farny NG, Haddock-Angelli T, et al. (2020). "Robust estimation of bacterial cell count from optical density." Communications Biology 3:512. Abstract: optical density "cannot be compared between instruments without a standardized calibration protocol and is challenging to relate to actual cell count." Retrieved 2026-07-29.
6. [6]Buchanan RL, Whiting RC, Damert WC (1997). "When is simple good enough: a comparison of the Gompertz, Baranyi, and three-phase linear models for fitting bacterial growth curves." Food Microbiology 14(4):313–326. Three-phase linear model: specific growth rate is zero in lag and stationary phases and constant in the exponential phase. Publisher full text is paywalled; abstract and model statement verified 2026-07-29.