Example data table
This sample set resembles a clean sine wave cycle pattern.
| Sample # | Value |
|---|---|
| 0 | 0.0000 |
| 1 | 0.7071 |
| 2 | 1.0000 |
| 3 | 0.7071 |
| 4 | 0.0000 |
| 5 | -0.7071 |
| 6 | -1.0000 |
| 7 | -0.7071 |
Formula used
The analyzer computes a discrete transform of your sample list. For an input sequence x[n] of length N, the transform is:
Frequency bins are mapped by:
Magnitude is:
When one-sided output is selected, interior bins are doubled. Windowing uses coherent gain for amplitude and mean-square gain for power.
How to use this calculator
- Paste your sampled values into the samples box.
- Enter the sampling rate in hertz for your data.
- Select a window to control spectral leakage.
- Choose padding and your preferred output scale.
- Set a peak search range and how many peaks to list.
- Press Submit to see results above the form.
- Use Download CSV or Download PDF for exports.
FAQs
1) Why do I see leakage around a single tone?
Leakage happens when a tone does not fit an integer number of cycles in the windowed record. A tapered window reduces leakage, but it broadens the peak. Try Hann or Blackman and increase record length when possible.
2) Does zero-padding increase true frequency resolution?
Zero-padding refines the frequency grid spacing, making peaks easier to locate and interpolate. It does not add new information, so it does not improve the underlying resolving power set by record length.
3) What sampling rate should I enter?
Use the rate used when the samples were collected. If your data comes from a sensor stream, it is often the device’s configured sample rate. A wrong sampling rate shifts all frequency labels and peak readouts.
4) What does “one-sided spectrum” mean?
For real signals, negative-frequency content mirrors positive frequencies. A one-sided view keeps bins from 0 to Nyquist and doubles interior bin values to preserve total energy, excluding DC and Nyquist.
5) Which scale should I choose: amplitude, RMS, or power?
Amplitude (peak) is useful for sine peak size. RMS matches common measurement readings. Power is a relative energy view across bins. dB variants compress dynamic range and help compare small components near large tones.
6) Why remove DC or detrend?
DC offset creates a strong 0 Hz bin that can dominate the output. Detrending removes slow drift that appears as low-frequency content. Use these when the offset or drift is not meaningful for your analysis.
7) Why is this analyzer slower for large inputs?
This tool uses a direct discrete transform for portability. It is reliable, but it scales roughly with N² operations. Keep FFT points modest or shorten the sample list for faster results and exports.
8) Can I paste time,value pairs?
Yes. If you paste rows like “0.00, 1.23”, the analyzer uses the last numeric value on each row. Ensure the sampling rate matches the time spacing, or compute it before analyzing.