Understanding Fourier Magnitude
A Fourier transform changes a signal into frequency parts. Each part is called a bin. The real value shows cosine strength. The imaginary value shows sine strength. The magnitude combines both values into one size. This size helps you see which frequencies are strong.
Why This Calculator Helps
Manual spectral work can be slow. Data may contain offsets, leakage, and scaling mistakes. This calculator gives options for mean removal, window choice, and normalization. It also lists phase, power, and frequency for every selected bin. You can study a small range or review the full transform.
The tool accepts real or complex samples. Real samples are common in vibration, audio, finance, and sensor logs. Complex samples are useful in communications and advanced maths. You can paste values separated by commas, spaces, or new lines. The page then applies your selected preprocessing steps before running the transform.
Reading the Results
The magnitude column is the main value. Larger magnitudes show stronger frequency content. The frequency column depends on your sampling rate. A sampling rate of 800 Hz with eight samples creates bins 100 Hz apart. Bin zero is the average component. Higher bins show repeating patterns within the record.
Windowing changes edge behavior. Rectangular keeps the raw record. Hann, Hamming, and Blackman reduce sharp boundary jumps. They can lower leakage when the sample does not contain full cycles. Normalization changes the displayed scale. It does not change which bin is dominant.
Use the phase column when timing matters. Phase shows the angle of the complex transform result. Power is magnitude squared. It is useful when comparing energy-like strength between bins. The peak summary highlights the strongest selected bins, so you can inspect important components quickly.
Practical Workflow
Start with clean data. Remove obvious typing errors. Choose mean removal when the baseline is not important. Select a window when the signal begins and ends at different levels. Use no normalization for raw DFT values. Use division by sample count when you want a stable scale across records.
Export the table after checking the settings. CSV is useful for spreadsheets. PDF is useful for reports. Keep the original samples with the exported values. That habit makes reviews easier and reduces confusion.