Gaussian Peak Fit Inputs
Paste your x,y data (two columns). Delimiters can be comma, space, tab, or semicolon.
Example Data Table
A simple single-peak dataset (copy into the input box if needed).
| x | y |
|---|---|
| 0 | 0.02 |
| 1 | 0.03 |
| 2 | 0.08 |
| 3 | 0.20 |
| 4 | 0.42 |
| 5 | 0.78 |
| 6 | 1.05 |
| 7 | 0.77 |
| 8 | 0.40 |
| 9 | 0.18 |
| 10 | 0.07 |
| 11 | 0.03 |
Formula Used
This calculator fits one to three Gaussian peaks, plus an optional baseline.
G(x) = A · exp( - (x − μ)² / (2σ²) )None: 0Constant: c0Linear: c0 + c1·xŷ(x) = baseline(x) + Σ Gᵢ(x)FWHM = 2·sqrt(2·ln 2)·σArea = A·σ·sqrt(2π)Fitting uses a damped least-squares method (Levenberg–Marquardt) to minimize the sum of squared residuals between y and ŷ.
How to Use This Calculator
- Paste your data as two columns: x and y.
- Select how many peaks you expect (1–3).
- Choose a baseline model if your signal has offset or drift.
- Adjust iterations, damping λ, and tolerance for tougher fits.
- Press Fit Gaussian Peaks to view parameters and stats.
- Use Download CSV or Download PDF to export results.
FAQs
1) What does Gaussian peak fitting estimate?
It estimates peak amplitude (A), center (μ), and width (σ) that best reproduce your measured signal, plus optional baseline terms.
2) When should I use a linear baseline?
Use a linear baseline when the background slowly rises or falls across x, such as drifting detector response or sloped chromatographic background.
3) How do I choose the number of peaks?
Start with the smallest number that visually explains the data. Increase peaks only if residuals stay structured and R² improves meaningfully.
4) Why can the fit look unstable?
Overlapping peaks, noisy data, and poor starting guesses can cause instability. Try higher damping λ, more iterations, or enforcing positive amplitudes.
5) What does σ represent physically?
σ is the standard deviation of the Gaussian shape along the x-axis. It relates to peak broadening mechanisms in instruments and separations.
6) How is peak area computed?
For each Gaussian, area equals A·σ·sqrt(2π). This is the integrated signal above baseline across all x.
7) What if my peaks are not Gaussian?
If peaks are skewed or have tails, Gaussian fits may bias parameters. Consider alternative shapes like Lorentzian, Voigt, or exponentially modified models.
8) Can I export the fitted curve for plotting?
Yes. The CSV export includes a fitted curve table plus a summary section, so you can plot y_fit versus x in any spreadsheet or graphing tool.