Skip to content
BioSyn

OD600 Calculator

OD600 is a turbidity reading, not a concentration, so turning it into cells per millilitre needs one extra number: how many cells of your organism, in your medium, on your instrument are worth one unit of OD600. This calculator multiplies your blank-corrected reading by that factor — taken from published BioNumbers records for E. coli, S. cerevisiae and B. subtilis, or from a calibration you measure yourself — and then does the three things a single magic-number converter skips: it shows the spread between independent citations instead of pretending one value is the truth, it folds in the dilution you made before reading and reports the cuvette and the culture separately, and it tells you when the reading is above OD 1 and therefore no longer proportional to anything. Optional dry-weight factors give the same answer in g DCW/L, a volume gives total cells, and a calibration mode fits your own factor from paired counts.

A spectrophotometer set to 600 nm is not measuring a chemical at all. There is no chromophore absorbing at 600 nm in a bacterial suspension; the beam is dimmed because cells deflect light out of it. That is why the reading is called optical density and why the conversion is empirical: the answer depends on how big the cells are, how they cluster, what the detector sees around the cuvette, and how far the light travelled. The cited records behind this page say exactly that — BNID 100985 notes that turbidity "is due to light scattering, not absorption", that it is "sensitive to the geometry of the detector relative to the cuvette, as well as the size and shape of the cells", and that one "must calibrate each spectrophotometer for each strain".

The consequence is a number you cannot memorise your way around. Two papers on the same organism, Escherichia coli, published in the same reference database give 1×10⁹ cells/mL per OD600 unit (BNID 100985, from Ausubel’s Short Protocols) and 2×10⁸ (BNID 103625, Sezonov and colleagues on E. coli physiology in Luria-Bertani broth) — a five-fold difference for one species. Pages that print "OD600 1.0 = 8×10⁸ cells/mL" without saying which organism, which medium and which instrument are not wrong so much as incomplete, and the incompleteness is the whole error budget.

So this page keeps the factor visible. Pick an organism and condition and the result prints the record it used — the BNID and the paper behind it — and, because one citation is a starting point rather than a measurement, the range across the published records for that organism sits beside it. That range is not hedging: it is the size of the disagreement in the literature, and it is the number that tells you whether to bother calibrating. For budding yeast the published spread inside one species is nearly four-fold (3×10⁷ cells/mL per OD600 at BNID 100986 against 8×10⁶ for stationary-phase cells at 30 °C, BNID 106301).

The arithmetic is then a multiplication with three bookkeeping terms, and the calculator handles all of them: the blank that was subtracted from your reading, the dilution that stood between the culture and the cuvette, and the window the instrument can be trusted in. BNID 100986 states the working limit plainly — "best to measure at OD<1 to insure linearity" — and BNID 109702 shows what happens past it: Corynebacterium glutamicum is reported at 5.5 g CDW/L at OD600 15 and 9.6 g/L at OD 30, so each additional OD unit is worth 13% less dry mass at the top of that range than at the bottom. Anything read above the window is scaled back by the dilute-into-window mode, which tells you the fold to make and the reading to expect.

Two out-frames come out of the same result, because a bench conversation about density happens in both. Cells per millilitre is what an inoculation decision needs; grams of dry cell weight per litre is what a fermentation and a product-titration conversation uses, and those ratios (0.30 to 0.47 g DCW/L per OD600 for E. coli across four independent labs, BNID 109835, 109838, 109837 and 107924) are on the page with their own citations. Add the volume your culture sits in and the same reading becomes a total number of cells. Counting colonies, growth rates and molar concentrations are deliberately not here: viable counts are the CFU calculator, OD over time is the population doubling time calculator, and a defined compound with a real extinction coefficient is the Beer-Lambert calculator.

What the instrument reported for the cuvette you put in, at 600 nm.

The same medium in the same cuvette. Leave empty or 0 if the instrument was already zeroed on it.

Counted or planned density of the culture, before any dilution you make to read it.

Each option is a published BioNumbers record for that organism and condition; the BNID in the label is the one the answer uses.

Cells/mL that one blank-corrected OD600 unit is worth for your strain, medium and instrument.

The g DCW/L per OD600 ratio published for that organism, with the spread across citations.

How many times the cuvette sample was diluted — 4 for one part culture plus three parts medium.

Optional: the volume your culture occupies, for a total-cell number.

Above this reading turbidity stops tracking cell number. Defaults to 1, the limit the cited record advises.

Blank-corrected reading of the first sample.

Cell count of that same sample.

Cell density6×10⁸ cells/mL
OD600 used0.6
Factor1×10⁹ cells/mL per OD600
Across citations1.2×10⁸ – 6×10⁸ cells/mL
Dry weight— g DCW/L
Total cells—
Transmittance25.12 %T
Expected reading— OD600
Point scatter— %

Factor from BNID 100985 · Ausubel et al., Short Protocols in Molecular Biology, 5th ed., Vol. 1 pp. 1-6

What one OD600 unit is worth, by published record

Every number below is a BioNumbers record with the conditions it was measured under. Note how far apart they are for one species — that spread, not a rounding error, is why a calibration beats a citation.

Organism and conditioncells/mL per OD600As publishedBNID
E. coli — exponential culture1×10⁹1×10⁸ cells/mL at OD600 0.1100985
E. coli — Luria-Bertani broth2×10⁸2×10⁷ cells/mL at OD600 0.1103625
S. cerevisiae (budding yeast)3×10⁷3×10⁷ cells/mL at OD600 1100986
S. cerevisiae — stationary phase, 30 °C8×10⁶8×10⁶ cells/mL at OD600 1106301
S. cerevisiae — stationary phase, 20 °C1.1×10⁷1.1×10⁷ cells/mL at OD600 1106300
S. cerevisiae — stationary phase, 10 °C9×10⁶9×10⁶ cells/mL at OD600 1106299
B. subtilis — F medium, log-stationary5×10⁸5×10⁸ cells/mL at OD600 1105287
B. subtilis — soft agar (published range)1×10⁸–5×10⁸1×10⁷–5×10⁷ cells/mL at OD600 ~0.1105286

Dry-weight ratios published on the same basis, each from its own record: 0.33 g DCW/L for E. coli K-12 MG1655 (BNID 109838); 0.30 g DCW/L for E. coli K-12 W3110 (BNID 109835); 0.36 g DCW/L for E. coli DH5α / BLR / JM109 (BNID 109837); 0.47 g DCW/L for E. coli BL21(DE3) — MOPS minimal, aerobic (BNID 107924); 0.62 g DCW/L for S. pombe (fission yeast) (BNID 108275); 0.37 g DCW/L for C. glutamicum — at OD600 15, divided out of the 5.5 g/L it reached at OD 15 (BNID 109702); 0.32 g DCW/L for C. glutamicum — at OD600 30, divided out of the 9.6 g/L it reached at OD 30 (BNID 109702). The working window the calculator warns about is the OD 1 advised in the comment to BNID 100986.

How to use the od600 calculator

Choose what to find: cell density from a reading, the reading a known density should give, how far to dilute a too-dark sample, or your own calibration factor from paired counts.

Enter the OD600 reading the instrument reported, and the blank reading of the same medium in the same cuvette if it was not zeroed away. The subtraction happens before anything else.

Pick a conversion factor. Each option is a published BioNumbers record and its label carries the BNID it came from; the result names the organism, the condition and the paper.

Set the dilution factor if the cuvette sample was diluted — 4 for one part culture plus three parts medium. The primary answer is your undiluted culture; the cuvette value is reported separately.

Optionally choose a dry-weight factor to get g DCW/L as well as cells/mL, and enter the culture volume to get the total number of cells in the flask.

For the calibration mode, enter at least two pairs of a blank-corrected OD600 and the cell count of that same sample — from a hemocytometer, a Coulter counter or a plate count. Three pairs or more, spread across 0.2 to 1.0, are what make the fit worth using.

Read the range printed under the result. If it is wider than the decision you are making, calibrate; if it is narrower, the published factor is good enough and you saved an afternoon.

Take the density forward: a dilution into a fresh flask is the dilution calculator or the serial dilution calculator, and the whole workflow is walked in the OD600 to cell density guide.

OD600 formula

Correct the reading: OD = OD600(sample) − OD600(blank)
Cell density: cells/mL = OD × factor (factor = cells/mL per one OD600 unit)
Undo the dilution: cells/mL of culture = OD × factor × dilution factor
Total cells: N = cells/mL of culture × volume (mL)
Dry weight: g DCW/L = OD × dilution factor × (g DCW/L per OD600)
Range: [OD × DF × min factor, OD × DF × max factor] over that organism’s cited records
Calibration: k = Σ(ODᵢ·cellsᵢ) / Σ(ODᵢ²) (least squares through the origin)
Expected reading: OD = cells/mL ÷ factor ÷ dilution factor
Dilution into window: fold = ⌈OD ÷ ceiling⌉, predicted reading = OD ÷ fold (ceiling 1.0)
A factor is not a constant. It is the slope one laboratory measured for one strain in one medium on one instrument at one wavelength — which is why this page prints the citation next to every number and the disagreement between citations next to that.

Variables and units

SymbolMeaningUnit
OD600Turbidity reading at 600 nm; light scattered out of the beam, not absorbed—
ODBlank-corrected reading used in every equation below—
factorCells/mL worth one OD600 unit for that organism and conditioncells·mL⁻¹·OD⁻¹
DFFold the sample was diluted before the reading was taken×
VVolume the culture occupiesmL
NTotal cells in that volumecells
kCalibration factor fitted from your own paired countscells·mL⁻¹·OD⁻¹
ceilingAbove this reading turbidity stops being proportional to cell numberOD600

How od600 is calculated

  1. Subtract the blank: 0.68 − 0.08 = 0.60. Everything after this uses the corrected number.
  2. Multiply by the factor of the organism you picked: 0.60 × 1×10⁹ = 6.0×10⁸ cells/mL in the cuvette.
  3. Multiply by the dilution factor to leave the cuvette and return to the culture: 6.0×10⁸ × 10 = 6.0×10⁹ cells/mL.
  4. Multiply by the published range for the same organism to get the interval the honest answer sits in: 1.2×10⁸–6.0×10⁸ at the cuvette stage, ten times that for the culture.
  5. Convert in parallel to biomass with a g DCW/L ratio (0.60 × 0.33 = 0.198 g/L) and to total cells with the volume (6.0×10⁸ × 250 mL = 1.5×10¹¹ cells).
  6. Check the reading against the window before believing any of it: above OD 1 the calculator stops calling the result a measurement and starts telling you what to dilute.

Worked laboratory examples

Example 1 — The plain conversion: 0.60 OD600 of E. coli

Given: reading 0.68, blank 0.08 · factor BNID 100985 = 1×10⁹ cells/mL per OD600 · no dilution

  1. OD = 0.68 − 0.08 = 0.60.
  2. cells/mL = 0.60 × 1×10⁹ = 6.0×10⁸.
  3. The same reading against BNID 103625 (E. coli in LB, 2×10⁸ per OD600) would give 1.2×10⁸ — the five-fold spread the two citations actually disagree by, printed as a range: 1.2×10⁸–6.0×10⁸ cells/mL.
  4. Transmittance at that reading is 25.1%, which is why the beam still reaches the detector: this is a readable sample, no dilution needed.

Result: 6.0×10⁸ cells/mL (range 1.2×10⁸–6.0×10⁸ across the two cited E. coli records)

Example 2 — The dilution you made before reading

Given: OD 0.60 after blank · BNID 100985 factor · 1 part culture + 9 parts medium, DF = 10 · culture volume 250 mL

  1. The cuvette holds 0.60 × 1×10⁹ = 6.0×10⁸ cells/mL.
  2. The flask is ten times that: 6.0×10⁹ cells/mL — this is the number the calculator prints as the answer.
  3. Total cells: 6.0×10⁹ × 250 mL = 1.5×10¹² cells in the culture.
  4. Dry weight at the same time: the culture is worth 0.60 × 10 = 6.0 OD600 units per litre of cuvette-equivalent, so 6.0 × 0.33 g DCW/L per OD600 (BNID 109838, E. coli MG1655) = 1.98 g DCW/L.

Result: 6.0×10⁹ cells/mL in the culture, 1.5×10¹² cells total, 1.98 g DCW/L

Example 3 — A reading that is too dark to trust

Given: OD600 3.2 · linear ceiling 1.0 (BNID 100986: measure below OD 1 to ensure linearity)

  1. fold = ⌈3.2 ÷ 1.0⌉ = 4, so make a 1-in-4 dilution — 50 µL culture plus 150 µL fresh medium.
  2. The diluted cuvette should then read near OD600 0.80 (3.2 ÷ 4).
  3. Convert that reading normally and multiply by the four-fold dilution: 0.80 × 1×10⁹ × 4 = 3.2×10⁹ cells/mL.
  4. The rounding is deliberately upward. A prediction of 0.80 is comfortably inside the window; a prediction of 0.98 would be sitting on its edge, and the flattening documented at BNID 109702 — where each OD unit of C. glutamicum is worth 13% less dry mass at OD 30 than at OD 15 — is exactly the error being avoided.

Result: Dilute 4-fold, expect OD600 0.80, then 3.2×10⁹ cells/mL

Example 4 — Calibrating your own factor

Given: (0.20, 1.8×10⁸) · (0.50, 4.5×10⁸) · (1.00, 8.6×10⁸) — readings against counts of the same samples

  1. k = Σ(OD·cells) ÷ Σ(OD²) = (3.6×10⁷ + 2.25×10⁸ + 8.6×10⁸) ÷ (0.04 + 0.25 + 1.00) = 1.121×10⁹ ÷ 1.29 = 8.69×10⁸ cells/mL per OD600.
  2. The three per-point ratios are 9.0×10⁸, 9.0×10⁸ and 8.6×10⁸ — a 4.5% spread, which is tight enough that one straight line is a fair description of your culture.
  3. That fitted factor is 13% below the published 1×10⁹ of BNID 100985, so at OD 0.60 your strain is 5.2×10⁸ cells/mL where the literature factor would have said 6.0×10⁸.
  4. Feed 8.69×10⁸ back in as your own factor and every later reading on this instrument uses your strain, not somebody else’s.

Result: Your factor 8.69×10⁸ cells/mL per OD600, scatter 4.5%, 13% below the cited E. coli record

Example 5 — Yeast, where the same reading means a different organism-sized number

Given: OD600 1.0 · BNID 100986 S. cerevisiae 3×10⁷ cells/mL per OD600 · BNID 106301 stationary phase at 30 °C 8×10⁶

  1. One OD600 unit of budding yeast is 3×10⁷ cells/mL — more than thirty times fewer cells than an E. coli reading of the same size, because a yeast cell is several times wider and scatters far more light.
  2. The same species measured in stationary phase gives 8×10⁶ at 30 °C, 1.1×10⁷ at 20 °C, 9×10⁶ at 10 °C (BNID 106301, 106300, 106299) — a 3.75-fold disagreement inside one organism, and the three stationary temperatures sit within 1.375× of each other.
  3. Yeast cultures also grow past the window: 5.2 ± 0.2 OD600 is the maximum reported in YPD (BNID 106303), which is a five-fold dilution away from linearity by the BNID 100986 criterion.
  4. For dry weight, fission yeast publishes 0.62 g DCW/L per OD600 (BNID 108275) — nearly double the E. coli ratio, and a reminder that the biomass factor is organism-specific too.

Result: 3×10⁷ cells/mL at OD600 1.0 (BNID 100986), with 8×10⁶–1.1×10⁷ for stationary-phase cells

Want the reasoning behind the numbers? Read How to Convert OD600 to Cell Density.

How to interpret the result

The primary card is the density of the culture you started with, in cells/mL — the cuvette value is printed underneath it so the dilution is never silently folded in twice. The card after that is a range rather than a second opinion: it is the interval spanned by the published records for that organism, and reading it is how you decide whether the difference matters. If you are choosing between inoculating at 10⁶ or 10⁷ cells/mL, a five-fold citation spread is not an obstacle. If you are reporting a yield in a paper, it is the first thing a reviewer will ask about, and the answer is a calibration, not a citation.

Dry weight appears only when you ask for it, because it is a different measurement with its own spread. Four E. coli labs publishing OD600-to-dry-weight ratios give 0.30, 0.33, 0.36 and 0.47 g DCW/L per OD600 (BNID 109835, 109838, 109837, 107924) — a 1.6-fold range that reflects medium, strain and how the drying was done, since a residual water content of a few percent moves the number. The ratio you should use is the one measured in your medium, and the published ones are the bracket.

The transmittance figure is included for one practical reason: it is the physical meaning of the warning. OD600 0.6 lets 25% of the light through, OD 1 lets 10% through, and OD 3.2 leaves 0.06% — one part in sixteen hundred. At that point the reading is dominated by whatever light the detector was never supposed to see, and the arithmetic that happily multiplies it by a factor is producing fiction. The dilute-into-window mode exists to keep you out of that region rather than to help you compute inside it.

When the calibration mode reports its scatter, it is testing your own linearity claim. A fitted factor with a 4% spread says turbidity and cell number are proportional across the range you measured and one number is safe. A spread above 15% means the counts stopped tracking the reading — usually because the dense points were above the window, or because cells clustered — and the honest response is to measure fewer, darker-free points, not to average them.

Assumptions

  • The reading was taken at 600 nm against a blank of the same medium in the same cuvette type, in a instrument whose stated working range covers the reading.
  • Cells are suspended singly and evenly — clumps and biofilm fragments scatter like larger particles and inflate the reading relative to counts.
  • The chosen factor was measured for a comparable organism, medium, growth phase and instrument geometry, since every one of those enters the published spread.
  • Turbidity is proportional to cell number over the readings used, which holds inside the linear window and is what the calibration mode checks.
  • A dilution factor of 1 means the culture went into the cuvette undiluted.

Limitations

  • A published factor transfers between instruments only approximately: scattering depends on detector geometry, and the source records say the calibration must be repeated per spectrophotometer and per strain.
  • The table is six cell-count records and seven dry-weight records. Anything else — a Pseudomonas, a CHO line, a Streptomyces — must be converted with your own factor, and the page says so rather than extrapolating.
  • Viable counts are not the same quantity as total counts. A plate count reads colony-forming units, which is what the CFU calculator is for, and dead cells contribute turbidity but not colonies.
  • Cell size changes with growth phase and medium, so a factor fitted on exponential-phase cells will misreport stationary-phase cells even on the same instrument.
  • The 1.0 ceiling is the conservative published advice, not a property of your machine; some instruments stay linear somewhat higher and microvolume readers behave differently again.

Common mistakes

  • Using one number for every organism — a yeast culture at OD 1.0 is roughly 3×10⁷ cells/mL (BNID 100986), an E. coli culture roughly 1×10⁹ (BNID 100985).
  • Reading an undiluted culture at OD 3 and multiplying it by a factor; above the window the reading under-reports the density it is supposed to measure.
  • Reporting the cuvette density as the culture density after a 1-in-10 dilution, and losing the factor of ten.
  • Subtracting the blank twice — once on the instrument and again in the calculator — or not subtracting it at all when the medium itself is turbid (a YPD blank can read several hundredths on its own).
  • Quoting a citation factor to two significant figures in a manuscript when the literature disagrees by five-fold for the same species.
  • Treating OD600 and absorbance-based concentration as the same measurement: no ε exists for a cell, so the Beer-Lambert law cannot be applied to a culture.

FAQ

How do I convert OD600 to cells/mL?

Multiply the blank-corrected reading by a factor expressing how many cells per millilitre one OD600 unit is worth for your organism and conditions: cells/mL = OD × factor × dilution factor. Published factors are on this page with their sources — 1×10⁹ for E. coli (BNID 100985), 2×10⁸ for E. coli grown in Luria-Bertani broth (BNID 103625), 3×10⁷ for budding yeast (BNID 100986), 5×10⁸ for Bacillus subtilis (BNID 105287). If the decision is sensitive, measure your own factor with the calibration mode.

Is OD600 the same as absorbance?

It is reported on the same scale but produced differently. Absorbance at 600 nm in a cell culture is almost entirely light scattered out of the beam by particles, not electronic absorption by a molecule — which is why a 600 nm reading has no extinction coefficient, why detector geometry matters so much, and why the Beer-Lambert calculator is the wrong tool for a culture and the right tool for a pure compound.

Why do different sources give different numbers for E. coli?

Because the factor is measured, not derived. Strain, medium, growth phase, cuvette path length and instrument geometry all change how much light a cell removes from the beam, and the published records for one species differ by up to five-fold (BNID 100985 against BNID 103625). The range this calculator prints is that disagreement made explicit, and the only way to collapse it is to calibrate on your own instrument.

What is the highest OD600 I can trust?

The record behind the yeast factor in this table puts it at about 1: read below OD 1 to ensure linearity (BNID 100986). Above that, scattering saturates — the dry-weight data for C. glutamicum (BNID 109702) is worth 0.37 g/L per OD at OD 15 but 0.32 g/L per OD at OD 30, so the top of the range reports fewer cells than are really there. Use the dilute-into-window mode: it gives the fold and the reading to expect.

Can this convert OD600 to CFU/mL?

Not directly, and the page does not pretend otherwise. CFU counts viable cells that grow into colonies; OD600 counts light-blocking particles including dead cells and debris. The ratio between them is strain- and phase-specific, so the honest route is a plate count next to a reading — and the plate arithmetic itself is on the CFU calculator.

How do I make a calibration curve for my instrument?

Take three or more samples from the same culture at different dilutions, read each one blank-corrected at 600 nm, and count cells in the same samples with a hemocytometer, a particle counter or a plate count. Enter the pairs in calibration mode; the calculator fits k = Σ(OD·cells)/Σ(OD²) — a line through the origin, because zero cells must scatter nothing — and reports how far apart the individual ratios were. The guide walks the whole procedure with numbers.

Do I need to subtract a blank?

Yes, and media blanks are not zero. Broth contains its own suspended material, and a plastic cuvette adds a little. Enter the reading of the same medium in the same cuvette as the blank and the calculator subtracts it first; if your instrument was already zeroed on that blank, leave the field empty or at 0.

What does the culture volume do to the answer?

Nothing to the density, and everything to the total. Density is per millilitre and does not care how much you have; multiplying it by the volume the culture occupies gives the number of cells in the flask — useful when you are harvesting, normalising a protein expression, or working out how many cells you actually inoculated.

Scientific Sources

External links open in a new tab and carry rel="nofollow". We cite primary references (NIST, NIH/NCBI, peer-reviewed papers) and established manufacturer technical documentation (NEB, Thermo Fisher, Promega, IDT). Calculated results are estimates; your protocol and manufacturer guidance take precedence.

Last reviewed: 2026-10 · Method version: 1.0