The mental picture
A photo can be sharp yet show the wrong crop. A frequency response can similarly look clean while its conditions hide an important property. A short time window excludes later reflections but sacrifices low-frequency resolution.
The principle
Electrical impedance and acoustic response are different loudspeaker measurements. Acoustic work needs documented position, distance, excitation and suitable microphone calibration. Combining drivers requires a consistent timing reference.
Try it yourself
Measure rather than guess
Compare 5 ms and 20 ms windows. Characteristic frequency resolution improves, but the longer window may include more room reflections.
Model limits: Only the relation between window length, approximate frequency resolution and path difference. No simulation of a specific room or measurement system.
A worked example
A 5 ms window corresponds roughly to 200 Hz frequency resolution. Smoothing does not make a response below that range suddenly reliable. At 343 m/s, an extra reflection path of 1.715 m corresponds to about 5 ms.
A closer look · equations and assumptions
Δf ≈ 1/T is a guide, not a sharp validity boundary. Window shape and analysis change spectral behaviour. Near-field data can extend low-frequency information, but merging with far-field data requires correct scaling and consideration of different radiation conditions.
A common pitfall. Do not blindly add or compare curves taken at different levels, microphone positions or independent time offsets.
Workshop takeaway
Change one thing at a time and label it clearly. One simple reproducible measurement is worth more than ten ambiguous curves.
Sources & transparency
Original illustrations and educational calculations. No measurements of a specific product. Edited 11 October 2026.
Work it out · answers unfold
Write down your prediction first, then your calculation. Open the answer only when you are ready to compare.
01 · An unpowered resistor measures 998 Ω. Nominal value 1 kΩ. Find relative error.
Show worked answer
- Deviation: -2 Ω.
- Divide by nominal 1000 Ω.
- Result: -0.2%.
02 · Why may an in-circuit resistance measurement be too low?
Show worked answer
- Other components may form parallel paths.
- The meter sees their equivalent resistance.
- Measure unpowered; isolate one lead if necessary.
03 · A 10 kΩ/10 kΩ divider at 10 V is measured across its lower resistor using a 10 MΩ voltmeter. Estimate the reading.
Show worked answer
- Lower equivalent resistance: 9990.01 Ω.
- U = 10 · 9990.01/(10000 + 9990.01).
- Reading: 4.9975 V.
04 · The same 10 MΩ meter measures a 10 MΩ/10 MΩ divider at 10 V. Find the reading.
Show worked answer
- Lower resistor parallel meter: 5 MΩ.
- Divider: 10 MΩ upper, 5 MΩ lower.
- Reading: 3.333 V rather than unloaded 5 V.
