Hemocytometer Cell Count
Enter your Neubauer grid count, the corner squares you counted, and your dilution — get cells/mL, viability %, and the total viable cells in your slurry. Built on the lab formula cells/mL = (count ÷ squares) × dilution × 10⁴ by a former New Belgium yeast specialist.
cells/mL = (count ÷ squares) × dilution × 10⁴. Each large 1mm × 1mm corner square over the 0.1mm chamber depth = 10⁻⁴ mL. Viability and total cells are optional refinements on top of this base count.
Turn a grid count into a number you can pitch by.
Cells/mL = (cells counted ÷ squares counted) × dilution × 10,000. A hemocytometer count is the ground truth underneath every other yeast number on this site — pitch rate, viability, starter sizing all assume you know your actual cell density, and the only way to know it is to count it under a scope. This tool takes your raw Neubauer count, your dilution, and (optionally) a methylene-blue live/dead split, and returns cells/mL, viability, and total viable cells in the sample — no spreadsheet, no unit-conversion mistakes.
Everything runs client-side. Nothing is stored, nothing is sent — change any input and the result updates live.
The hemocytometer formula, explainedWhy the ×10,000 — and how the chamber gets you there.
The geometry: one corner square = 10⁻⁴ mL
A Neubauer-ruled hemocytometer has a fixed, known geometry. Each of the four large 1mm × 1mm triple-ruled corner squares (the ones used for yeast, as opposed to the smaller center grid used for blood cells) sits under a chamber exactly 0.1mm deep when the coverslip is properly seated. Volume is area × depth: 1mm × 1mm × 0.1mm = 0.1 mm³ = 0.1 µL = 0.0001 mL. That single fact — one counted square represents exactly 10⁻⁴ mL — is where the ×10,000 in the formula comes from. Everything else in the calculation is just averaging your count across the squares you used and undoing your dilution.
cells_per_square = cells_counted ÷ squares_counted
cells_per_mL = cells_per_square × dilution × 10,000
# optional refinements
viability_pct = live ÷ (live + dead) × 100
viable_cells_per_mL = cells_per_mL × (viability_pct ÷ 100)
total_cells = cells_per_mL × volume_mL
Why the dilution has to be right — not close
The dilution factor is a straight multiplier on the final answer, which means a bookkeeping slip there isn't a rounding error — it's an order of magnitude. Write down what you actually mixed (1 part slurry to N−1 parts diluent, expressed as 1:N) before you pipette, and re-derive N from your own notes rather than trusting memory when you sit down to calculate.
400 cells counted across 4 corner squares at a 1:100 dilution.
cells_per_square = 400 ÷ 4 = 100. cells_per_mL = 100 × 100 × 10,000 = 1.0 × 10⁸ cells/mL = 100 M/mL = 0.1 B/mL.
A thick repitch slurry is much denser than that, so you'd dilute further to get a countable field — the same 400-cells-in-4-squares count at a 1:1000 dilution instead reads 1.0 × 10⁹ cells/mL = 1.0 B/mL, ten times the concentration, from an identical microscope view. The dilution is doing all the work.
When the standard number is wrong.
A hemocytometer count is exact arithmetic on an inexact input — the microscope view. Here's where a technically-correct calculation still gives you a number you shouldn't trust.
Dilution bookkeeping is the #1 error — and it's 10×, not 10%
Confusing a 1:10 dilution with a 1:100 doesn't nudge your answer, it multiplies or divides your pitch by ten. Because the formula treats dilution as a pure linear multiplier, there's no internal check that catches the mistake — the math will happily return a confident, wrong answer. Write the dilution down at the bench, before you pipette, and check it against your notes when you sit down to calculate — not from memory.
Count enough cells for the statistics to mean something
A hemocytometer count is a sample, and small samples are noisy — Poisson counting statistics mean the relative error on a count shrinks as the count grows. Aim for roughly 100–400 total cells across the squares you use. Under about 100 and the noise swamps the result — two people counting the same chamber can land 20–30% apart. If instead you can't resolve individual cells because the field is packed (more than roughly 10 cells per small square), the sample is too dense to count accurately at all — dilute further and recount rather than trust a crowded field.
The edge rule: count two sides, skip the other two
Cells that straddle a boundary line between two adjacent squares get counted once by convention, not twice. The standard rule: count a cell if it touches the top or left boundary of its square, and skip it if it only touches the bottom or right. Apply this inconsistently — or not at all — and you systematically overcount, more so as cell density rises and more cells sit on lines.
Methylene blue tells you viability, not vitality — and it can lie both ways
Methylene blue works because living cells enzymatically reduce and decolorize the dye while dead cells can't, so dead cells stain blue and live ones stay clear. But on a very healthy population it can over-report viability — some marginal cells reduce the dye just long enough to read as "live" before failing shortly after. On a very dead sample the dye can also behave unpredictably. Methylene violet tracks flow-cytometry viability more closely if you need a second opinion. Either way, remember this measures viability — alive versus dead — not vitality, which is how metabolically ready the living cells actually are to ferment. A culture can be 95% viable and still ferment sluggishly if vitality is low.
Flocculent strains hide cells inside clumps
Highly flocculent strains clump before you ever load the chamber, and clumped cells hide from the count — you see the outer cells of a clump and miss the ones packed inside it, which undercounts the true population. Mix the sample thoroughly and, if needed, break clumps mechanically or with a mild disaggregation step before you pipette onto the hemocytometer, or your "counted" number is a floor, not the real density.
Notes from the labAt New Belgium, the hemocytometer was the number every other number deferred to. Pitch rate, viability, starter targets — they all assumed a cell count somewhere underneath was honest, and the fastest way to lose a batch to a preventable stall was a sloppy count treated as gospel. The mistakes were never exotic: someone wrote 1:100 on the sheet when they'd actually run 1:10, or counted a crowded field they should have diluted and recounted, or missed the clumps in a flocculent strain and reported a slurry thinner than it really was. None of that shows up as an error message — the arithmetic runs clean either way. The formula is only as good as the count you feed it — count enough cells, write the dilution down before you trust it, and recount anything that looks too dense to resolve.
Track your counts against how the batch finishes.
A cell count is a snapshot. The value compounds when every propagation and every repitch gets logged — count, viability, dilution — against how the fermentation actually performed. That's what a QC dashboard does automatically.