Calibration Curve Calculator

Signal per unit concentration, from the best-fit line of your standards.
Signal where concentration = 0 (the blank).
The instrument reading (e.g. absorbance, peak area) for your unknown.
Chemistry

Use a linear calibration curve (y = mx + b) to convert an instrument signal into an unknown concentration. Enter the slope and intercept from your standards, plus the measured signal, and you get the concentration back instantly — the core step in spectrophotometry, chromatography, and most quantitative lab assays.

Formula

concentration = (signal − intercept) ÷ slope, from y = mx + b
  • A calibration curve plots known standard concentrations (x) against their measured signal (y) and fits a straight line y = mx + b.
  • Rearranging for the unknown: x = (y − b) ÷ m, where m is the slope and b is the y-intercept.
  • Get the slope and intercept from a linear regression of your standards (most spreadsheets and instruments report both).
  • Stay inside the calibrated range — extrapolating below the lowest or above the highest standard is unreliable.
  • A good calibration has R² ≥ 0.99; a poor fit means the (y − b) ÷ m result carries large uncertainty.
  • The concentration is returned in the same units you used for your standards (e.g. mg/L, ppm, µM).

Absorbance reading of 0.385

Inputs
  • Slope (m): 0.025
  • Y-Intercept (b): 0.01
  • Measured Signal (y): 0.385

With slope m = 0.025 and intercept b = 0.01, an absorbance of 0.385 gives concentration = (0.385 − 0.01) ÷ 0.025 = 0.375 ÷ 0.025 = 15 units.

Frequently asked questions

How do I get the slope and intercept?
Measure several standards of known concentration, plot signal (y) against concentration (x), and fit a straight line. The line's slope is m and where it crosses the y-axis is b. Spreadsheets do this with SLOPE() and INTERCEPT().
Why subtract the intercept?
The intercept is the signal you'd read at zero concentration (the blank). Subtracting it isolates the part of the signal that actually comes from your analyte before dividing by the slope.
Can I use this outside the standard range?
It's not recommended. Calibration lines are only validated between your lowest and highest standards. Outside that range the response often becomes non-linear, so dilute or concentrate the sample to fit the curve.
What if my curve isn't a straight line?
This calculator assumes a linear fit. If your data curves (common at high concentrations), restrict to the linear region or use a non-linear model instead.