I'm skimming over documentation relative to loudspeaker frequency response measurement.

Regarding open-field/far-field measurements, most authors mention issues with the ground reflection.

For example, in https://audioxpress.com/article/measuring-loudspeaker-low-frequency-response we can see this drawing:

enter image description here

Why couldn't we simply consider pointing the loudspeaker toward the sky, with the mic placed above it, to get rid of the ground reflection?

  • 1
    One would assume these tests are to be done indoors. – Tetsujin Mar 4 '20 at 10:08

speakers don't just emit sound towards the front, they emit sound in 360 degrees despite the efforts of designers. Higher frequency drivers are more directional than lower frequency drivers. The lower the frequency, the more 'omnidirectional' the device is.

By attempting to measure frequency response vertically above a speaker, you are going to experience interference generated by the reflection directly behind the device - i.e. the floor. In your scenario, the distance between the device and the floor is likely to be quite small (as I assume you are trying to avoid having to loft the mic and the device). If you imagine a distance of 1m to the floor, you will potentially experience interference in the 343Hz range.

Having a lofted device and measurement mic, you will likely have interference at a much lower frequency than if you were to attempt measurement vertically as you have suggested.

  • Thanks for the explanation, Mark. two remarks: (1) Indeed, I considered the speaker as emitting a "significant amount" of sound only to the front. I assume it has a similar polar pattern than directional microphones. (2) Actually I was planning to raise the speaker to 1m from the floor, and place the mic another 1m away from the speaker. Are my math correct if I say interferences will start at 115Hz? – Sylvain Leroux Mar 7 '20 at 14:19
  • 1
    The way I would work this out is that with nodes at 0, L/2, L and with the fact you are dealing with distances at 1m (direct) and 2.24m (via floor reflection). You will see nodes at the mic at the following frequencies: 343Hz, 686Hz (Direct), 153Hz, 306Hz (Reflected) plus additional frequencies up the comb. – Mark Mar 8 '20 at 1:38
  • Thanks, @Mark. Could you point me to some doc explaining how to obtain those results? – Sylvain Leroux Mar 9 '20 at 17:57
  • Ask it as a question on the stack and I'll try and answer it for you. – Mark Mar 10 '20 at 0:08
  • done: sound.stackexchange.com/q/47023/25426 The title is pretty awful, so don't hesitate to fix it if you may ;) – Sylvain Leroux Mar 10 '20 at 19:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.