Free · No login required
Comb Filter Frequency Calculator
Predict the first three cancellation frequencies and tooth spacing when a signal is mixed with one delayed copy.
Delay difference (ms): 0.01–1000
Inputs stay on this page and clear on refresh.
Comb Filter Frequency Calculator
Delay difference (ms): 1
- First cancellation (Hz)
- 500.000000
- Tooth spacing (Hz)
- 1000.000000
- Second cancellation (Hz)
- 1500.000000
- Third cancellation (Hz)
- 2500.000000
Enter the arrival-time difference, not the absolute latency of either signal. For a direct signal plus an equal-level, same-polarity delayed copy, cancellation occurs at f = (2k + 1) × 500 / d Hz, where d is the delay in milliseconds and k starts at zero. Adjacent cancellations are 1000 / d Hz apart. A 1 ms difference gives nulls at 500, 1500 and 2500 Hz, spaced by 1000 Hz. Doubling the delay halves every frequency. Use this theoretical pattern to investigate reflections or two microphones picking up one source. The input accepts 0.01–1000 ms; zero is excluded because it produces no finite cancellation frequencies in this model. Results may lie outside human hearing and are not limited to an audible band. Large delays may be heard as echoes. Six displayed decimals describe numerical precision, not measurement accuracy. This is a single feedforward delay model, with no feedback and no polarity inversion. Unequal levels reduce notch depth; source spectra, filters, room reflections and additional paths alter the actual response. It does not measure sound, prescribe EQ boosts or prove that a real system has complete cancellation. Confirm the arrival difference and response with suitable measurements. Copy, download or print the input, frequency pattern and assumptions together. Blank or invalid input removes the previous result.
Methodology
How it works
Enter the arrival-time difference, not the absolute latency of either signal. For a direct signal plus an equal-level, same-polarity delayed copy, cancellation occurs at f = (2k + 1) × 500 / d Hz, where d is the delay in milliseconds and k starts at zero. Adjacent cancellations are 1000 / d Hz apart. A 1 ms difference gives nulls at 500, 1500 and 2500 Hz, spaced by 1000 Hz. Doubling the delay halves every frequency. Use this theoretical pattern to investigate reflections or two microphones picking up one source. The input accepts 0.01–1000 ms; zero is excluded because it produces no finite cancellation frequencies in this model. Results may lie outside human hearing and are not limited to an audible band. Large delays may be heard as echoes. Six displayed decimals describe numerical precision, not measurement accuracy. This is a single feedforward delay model, with no feedback and no polarity inversion. Unequal levels reduce notch depth; source spectra, filters, room reflections and additional paths alter the actual response. It does not measure sound, prescribe EQ boosts or prove that a real system has complete cancellation. Confirm the arrival difference and response with suitable measurements. Copy, download or print the input, frequency pattern and assumptions together. Blank or invalid input removes the previous result.
Method and assumptions documented by Techrider.live
2026-10-22
Frequently asked questions
Why is zero delay excluded?
Two identical, same-polarity signals at zero delay reinforce at all frequencies; this model has no finite nulls then.
Does this predict notch depth?
No. Exact nulls assume equal levels and perfect coherence. Unequal levels and real acoustic paths change the response.