Calculator Inputs
Example Data Table
| Parameter | Example Value | Comment |
|---|---|---|
| Sample Rate | 4096 Hz | Used with N = 1024, so frequency resolution becomes 4 Hz. |
| Selected Bin | k = 32 | Bin center frequency equals 128 Hz. |
| Input FFT Bin | 12 - j3 | Complex input spectrum sample at the selected frequency bin. |
| Output FFT Bin | 18 - j6 | Complex output spectrum sample at the same frequency bin. |
| Window Coherent Gains | Input 0.50, Output 0.48 | Corrects amplitude scaling caused by different window choices. |
| Expected Magnitude Ratio | 1.597844 | Computed from corrected output divided by corrected input. |
| Expected Gain | 4.070686 dB | Uses 20 log10 of the magnitude ratio. |
| Expected Phase | -4.398705° | Argument of H[k] after correction and before extra compensation. |
| Estimated Group Delay | 95.458016 µs | Calculated from -phase divided by angular frequency. |
| Measured Signal Frequency | 129 Hz | Creates a 0.25-bin mismatch, which hints at leakage risk. |
Formula Used
1) Frequency resolution
Δf = Fs / N
2) Selected bin frequency
f[k] = k × Δf
3) Window-corrected spectra
Xc[k] = X[k] / Gx and Yc[k] = Y[k] / Gy
4) Transfer function estimate
H[k] = Yc[k] / Xc[k]
5) Magnitude, gain, and phase
|H[k]| = √(Re(H)^2 + Im(H)^2), Gain(dB) = 20 log10(|H[k]|), Phase = atan2(Im(H), Re(H))
6) Delay estimate at one frequency
τ = -φ / (2πf)
This calculator estimates a frequency-response value at one FFT bin. It does not build a full polynomial transfer function model. It is ideal for frequency-domain testing, swept-tone checks, vibration studies, filter verification, and sampled-system comparison work.
How to Use This Calculator
- Choose whether your FFT data is available as complex bins or polar values.
- Enter the sample rate, FFT size, and the bin index you want to inspect.
- Fill in the input and output spectrum values for that same bin.
- Add coherent gains if different windows were used on input and output signals.
- Set a phase compensation value when you need to remove cable or fixture phase shifts.
- Optionally add measured signal frequency and coherence to judge leakage and trustworthiness.
- Press Calculate Transfer Function to display the result table above the form.
- Use the CSV or PDF buttons to export the calculated result metrics.
FAQs
1) What does this calculator actually compute?
It computes a frequency-domain transfer-function estimate at one FFT bin. You get magnitude ratio, gain in decibels, complex response, phase, and an approximate delay for the selected frequency.
2) Why must the input bin be nonzero?
The transfer function is output divided by input. If the input FFT value is zero or extremely small, the division becomes unstable and the estimate becomes meaningless.
3) Why does the calculator ask for coherent gain?
Window functions change amplitude scaling. Coherent gain corrects that scaling so the transfer estimate better reflects the actual frequency response instead of the window’s attenuation effect.
4) What does bin mismatch mean?
Bin mismatch measures how far the measured signal frequency sits from the selected FFT bin center. Larger mismatch usually increases spectral leakage and can distort magnitude and phase estimates.
5) Is the delay result always valid?
It is a local estimate at one frequency. It works best for nearly linear phase behavior near that bin. At zero frequency, the simple delay formula is undefined.
6) What does coherence tell me?
Coherence is a quality indicator between zero and one. Values close to one usually suggest a cleaner, more repeatable relationship between the input and output at that frequency.
7) Should I use complex mode or polar mode?
Use complex mode when your FFT tool gives real and imaginary parts. Use polar mode when your instrument reports magnitude and phase directly. Both paths lead to the same transfer estimate.
8) Why is FFT size restricted to powers of two?
Most FFT workflows prefer powers of two because they are computationally efficient and standard across software and instruments. The calculator follows that common practice for consistency.